UA2014 - Underwater Acoustics Conference

Transcription

UA2014
2nd International Conference and Exhibition on Underwater Acoustics
2nd International Conference and Exhibition on
Underwater Acoustics
22nd to 27th June 2014
Rhodes. Greece
Program
&
Book of Abstracts
Program and Book of Abstracts
a
UA2014
Conference logo by Yorgis Androulakis, IACM-FORTH
b
UA2014
Table of Contents
Welcome ____________________________________________________________ 1
Conference Chairmen __________________________________________________ 2
Supported by _________________________________________________________ 2
Endorsed by __________________________________________________________ 2
Scientific Committee ___________________________________________________ 3
Session Organizers_____________________________________________________ 4
Conference Program _______________________________________________ 6
Monday 23rd June 2014 _________________________________________________ 6
Registration _________________________________________________________________ 6
Welcome ___________________________________________________________________ 6
Session 14: Innovative Approaches for Characterizing Ocean Bottom Properties __________ 7
Session 30: Underwater Unexploded Ordnance (UXO) Detection and Remediation ________ 8
Session 8: Advances in Acoustic Measurement Systems: Technologies and Applications ___ 9
Session 26a: Underwater Acoustic Measurement Facilities and Standards ______________ 10
Session 4a: Acoustic Tomography in Shallow Seas, Benthic and Terrestrial Waters _______ 11
Session 19a: Sonar Performance Modeling and Verification: Applications to Active and
Passive Sonar_______________________________________________________________ 12
Session 6a: Acoustics of Bubbles, Oil and Gas _____________________________________ 13
Session 26b: Underwater Acoustic Measurement Facilities and Standards ______________ 14
Session 4b: Acoustic Tomography in Shallow Seas, Benthic and Terrestrial Waters ______ 15
Session 19b: Sonar Performance Modeling and Verification: Applications to Active and
Passive Sonar_______________________________________________________________ 15
Session 6b: Acoustics of Bubbles, Oil and Gas _____________________________________ 16
Session 16: Outer Continental Shelf, Shelfbreak and Canyon Acoustics ________________ 17
Tuesday 24th June 2014 ________________________________________________ 18
Session 10: Comprehensive Nuclear-Test-Ban Treaty Monitoring _____________________ 18
Session 35a: Acoustic Modelling _______________________________________________ 19
Session 24a: Three-dimensional sound propagation models _________________________ 20
Session 31a: Unmanned Vehicles (AUV, USV and Gliders) for Underwater Acoustic
Surveillance and Monitoring___________________________________________________ 21
Session 35b: Acoustic Modelling _______________________________________________ 22
Session 1: Acoustic imaging ___________________________________________________ 23
Session 24b: Three-dimensional sound propagation models _________________________ 24
i
UA2014
Session 31b: Unmanned Vehicles (AUV, USV and Gliders) for Underwater Acoustic
Session 21a: Soundscapes and Measuring Noise __________________________________ 26
Session 29a: Underwater Communication and Networking __________________________ 27
Session 32a: Vector Sensors: Development and Applications ________________________ 28
Session 31c: Unmanned Vehicles (AUV, USV and Gliders) for Underwater Acoustic
Session 21b: Soundscapes and Measuring Noise __________________________________ 29
Session 29b: Underwater Communication and Networking __________________________ 30
Session 32b: Vector sensors: development and applications _________________________ 30
Wednesday 25th June 2014 _____________________________________________ 31
Plenary Lecture _____________________________________________________________ 31
Session 20a: Sonar Signal and Information Processing ______________________________ 31
Session 18: Sensitivity of underwater acoustic observables __________________________ 32
Session 9: Calibration of Sonar and Hydrophones __________________________________ 33
Session 7a: Acoustics of marine renewable energy developments ____________________ 34
Session 20b: Sonar Signal and Information Processing ______________________________ 35
Session 17a: Radiated Noise from Ships and Production Platforms ____________________ 36
Session 12a: Experimental and modelling validation of target strength measurements ___ 37
Session 7b: Acoustics of marine renewable energy developments ____________________ 38
Session 20c: Sonar Signal and Information Processing ______________________________ 39
Session 17b: Radiated Noise from Ships and Production Platforms ____________________ 40
Session 12b: Experimental and modelling validation of target strength measurements ___ 41
Session 7c: Acoustics of marine renewable energy developments ____________________ 42
Thursday 26th June 2014 _______________________________________________ 43
Session 11: Distributed Networked Systems for Surveillance _________________________ 43
Session 23: Tank Experiments _________________________________________________ 44
Session 5a: Acoustics in Polar Environments ______________________________________ 45
Session 25a: Towards Automatic Target Recognition: Detection, Classification and Modeling
(of Underwater Targets) ______________________________________________________ 46
Session 2a: Acoustic Monitoring of Marine Mammals ______________________________ 47
Session 33: Bioacoustics ______________________________________________________ 48
Session 5b: Acoustics in Polar Environments ______________________________________ 49
Session 25b: Towards Automatic Target Recognition: Detection, Classification and Modeling
(of Underwater Targets) ______________________________________________________ 50
Session 2b: Acoustic Monitoring of Marine Mammals ______________________________ 51
Session 13a: Habitat Mapping: Procedures and Results _____________________________ 52
Session 27a: Underwater Acoustic Studies in Asian Seas ____________________________ 53
Session 34a: Underwater Communication ________________________________________ 54
ii
UA2014
Session 2c: Acoustic Monitoring of Marine Mammals ______________________________ 55
Session 13b: Habitat Mapping: Procedures and Results _____________________________ 55
Session 27b: Underwater Acoustic Studies in Asian Seas ____________________________ 56
Session 34b: Underwater Communication _______________________________________ 56
Friday 27th June 2014__________________________________________________ 57
Session 15: Modeling Sonar Performance in Uncertain Environments _________________ 57
AB Wood Medal Lecture ______________________________________________________ 58
Session 22: Synthetic Aperture Sonar: State-of-the-art _____________________________ 58
Session 36: Acoustic Inversions ________________________________________________ 59
Closing Ceremony ___________________________________________________________ 59
Poster Sessions ___________________________________________________ 60
Tuesday 24th June 2014 ________________________________________________ 60
Wednesday 25th June 2014 ____________________________________________ 61
Thursday 26th June 2014 _______________________________________________ 61
Book of Abstracts _________________________________________________ 62
Session 1: Acoustic imaging ____________________________________________ 62
A Design Philosophy of Portable,High-frequence Image Sonar System _________________ 62
Synthetic Aperture Sonar Images Mosaic Based on SIFT and RANSAC Method___________ 62
Research on underwater target detection based on seafloor physiognomy-matching of side
scan sonar imagery __________________________________________________________ 63
Research on Target Classification for Side-Looking Sonar Based on Acoustic Model ______ 63
Design and Implementation of A Real-time 3-D Imaging Sonar Signal Processing System on
TMS320C6678 ______________________________________________________________ 63
Numerical Analysis for Ambient Noise Imaging with Acoustic Lens: Target Detection around
the Barge Moored in Uchiura Bay ______________________________________________ 64
Session 2: Acoustic Monitoring of Marine Mammals ________________________ 65
Passive acoustic monitoring of humpback whales in Exmouth Gulf using a sparse array of
DIFAR sensors ______________________________________________________________ 65
Acoustic monitoring of marine mammals in the Chukchi Sea – Three case studies _______ 65
Passive acoustic monitoring and bio-sonar characterization of Ganges river dolphin, India's
national aquatic animal ______________________________________________________ 66
Passive acoustic source localisation methods for the nonintrusive monitoring of echolocating
dolphins in the wild __________________________________________________________ 67
Bowhead whale calls localization using a single receiver and warping processing ________ 67
Passive Acoustic Monitoring of Marine Mammals using a Ternary Array _______________ 68
Acoustics as a tool to reveal population structure of the elusive blue whale ____________ 69
iii
UA2014
The use of passive acoustic data to predict beaked whale habitat in the California Current
Ecosystem _________________________________________________________________ 69
Remote Sensing of Large Herring Shoals by Baleen Whales __________________________ 70
Modelling the impact of ocean environment on automatic aural classification of marine
mammals. _________________________________________________________________ 70
Classifying Humpback Whale Individuals from their Nocturnal Feeding-Related Vocalizations
__________________________________________________________________________ 71
Classification of Beaked Whale Signal Recorded in Atlantic __________________________ 72
Acoustic broadband backscattering and classification of fish _________________________ 72
Session 4: Acoustic Tomography in Shallow Seas, Benthic and Terrestrial Waters _ 73
Passive acoustic tomography of fluid-mud processes on the Amazon continental shelf ___ 73
Horizontal Ocean Current Tomography with Iterative Model Weighting Constraint ______ 74
Estimation of temperature information based on ocean ambient noise measurement ____ 74
A Coastal Acoustic Tomography Experiment for Tidal Current Measurement in the South of
Jiaozhou Bay _______________________________________________________________ 75
Acoustic investigations of unsteady salinity intrusion in a diversion channel ____________ 76
Vertical profiling of temperature and velocity from the quite limited data set of coastal
acoustic tomography ________________________________________________________ 76
Tomographic mapping of coastal upwelling generated in Hiroshima Bay, Japan _________ 77
Session 5: Acoustics in Polar Environments ________________________________ 78
Acoustic Communications Experiments in the Fram Strait 2013 ______________________ 78
Measurements of the Ambient Noise Field in an Arctic, Glacial Fjord __________________ 78
Acoustic propagation in the Marginal Ice Zone and the implications for navigation of
underwater vehicles _________________________________________________________ 79
Zooplankton distribution studies combining acoustical and optical observations ________ 80
Three-dimensional Source Localization using an Ice-mounted Geophone ______________ 80
An acoustical study of gas bubbles escaping from melting growlers ___________________ 81
The soundscape of the Fram Strait Marginal Ice Zone ______________________________ 82
Under sea-ice acoustic noise and propagation measurements in Tethys Bay (Ross Sea,
Antarctica) _________________________________________________________________ 82
Sources of long-term ambient ocean sound near the Antarctic Peninsula ______________ 83
Session 6: Acoustics of Bubbles, Oil and Gas _______________________________ 84
A technique to measure the real surface tension on a bubble wall ____________________ 84
Passive acoustic quantication of gas releases _____________________________________ 84
Acoustical classification of the shallow sediment gaseous structures in the Southern Baltic
Sea _______________________________________________________________________ 85
Comparison of theories for acoustic wave propagation in gassy marine sediments _______ 85
Attenuation of low frequency underwater noise using arrays of air-filled resonators _____ 86
Numerical modelling of a bubble curtain _________________________________________ 87
iv
UA2014
The role of air bubbles in acoustic surface loss confirmed by historical data (1949-2005)
about attenuation excess of sound in oceanic surface channels ______________________ 87
Session 7: Acoustics of marine renewable energy developments ______________ 89
Designing practical on-site calibration protocols for acoustic systems; Key elements and
pitfalls ____________________________________________________________________ 89
An environmental survey around the Narec Offshore Anemometry Hub (NOAH) – a
comparison between acoustic measurement instruments. __________________________ 89
Methods for measurement of long term radiated noise from a wave energy system _____ 90
Environmental inversion with an autonomous hydrophone in a wave energy device
deployment site ____________________________________________________________ 91
Field deployments of a self-contained subsea platform for acoustic monitoring of the
environment around marine renewable energy structures __________________________ 91
Underwater Sounds from Drillships and Support Vessels During Exploratory Drilling Offshore
Alaska ____________________________________________________________________ 92
Underwater sound due to a subsea high speed turbo-compressor ____________________ 93
New methods in impact pile driving noise attenuation _____________________________ 93
Soil vibration due to offshore pile driving and induced underwater noise ______________ 94
Model results for offshore piling acoustics featuring an elastic sediment with a depthdependent shear wave speed _________________________________________________ 94
The evanescent pressure waves associated with ground roll waves from seabed impacts _ 95
Effective Reduction of Offshore Piling Noise ______________________________________ 95
Underwater noise assessment of wave energy devices _____________________________ 97
Underwater sound levels at a wave energy device testing facility in Falmouth Bay, UK ____ 97
Cabled observatory enabled acoustic monitoring of hydrothermal discharge ___________ 98
Session 8: Advances in Acoustic Measurement Systems: Technologies and
Applications _________________________________________________________ 99
The SMO antenna: status and first results ________________________________________ 99
Processing strategies for evaluating the ship radiated noise using an underwater vertical
array ____________________________________________________________________ 100
Development of ultra sideband transducer ______________________________________ 101
Basic Study of Rhomboidal Acoustic Lens Constructed with Phononic Crystal __________ 101
Session 9: Calibration of Sonar and Hydrophones __________________________ 103
Low Frequency Tank Calibration by comparison __________________________________ 103
Adulteration of underwater acoustic measurements ______________________________ 103
The calibration of hydroacoustic channel of mobile measurement module ____________ 104
Calibrating hydrophones at very low frequencies _________________________________ 104
Array Shape Estimation Using Measurements of Heading and Depth Sensors __________ 104
Session 10: Comprehensive Nuclear-Test-Ban Treaty Monitoring _____________ 106
v
UA2014
Results from ocean currents and acoustic propagation modelling studies in support of the
installation of CTBTO Hydroacoustic station HA04, Crozet Islands, France _____________ 106
Using correlation matrices to identify temporal characteristics of ambient noise _______ 106
Antarctic’s Siren Call: The Sound of Icebergs _____________________________________ 107
Evanescent wave coupling in a geophysical system: Airborne acoustic signals from the Mw
8.1 Macquarie Ridge earthquake ______________________________________________ 107
Basin Scale Time-Domain Modelling for CTBTO Tracking ___________________________ 108
Session 11: Distributed Networked Systems for Surveillance _________________ 109
Adaptive Bayesian behaviors for AUV surveillance networks ________________________ 109
Problems in globally optimizing underwater surveillance networks in communications
limited environments _______________________________________________________ 109
Parameter estimation for non-cooperative multistatic sonar _______________________ 110
Optimal Area Coverage in Autonomous Sensor Networks __________________________ 110
Acoustic communication and localization in AUV cooperative surveys ________________ 111
Session 12: Experimental and modelling validation of target strength measurements
__________________________________________________________________ 112
Concepts for reliable Target Echo Strength measurements and improved target
representation ____________________________________________________________ 112
Model Tank Measurements and Using a Random Noise Feld to Determine the Scattering
Properties of an Object ______________________________________________________ 113
On the design and construction of drifting-mine test targets for sonar, radar and electrooptical detection experiments ________________________________________________ 113
Open water target strength measurements and validation _________________________ 113
Finite element modeling of acoustic radiation force for elastic objects ________________ 115
Improved modeling accuracy of the Elastic Object Response by Inclusion of Higher Order ReScattering ________________________________________________________________ 115
Acoustic scattering from partially buried cylinders: Measurement validation and
interpretation using physical acoustics and finite element models ___________________ 116
Performance Modelling and Experimental Validation for a Sediment-Penetrating SAS ___ 116
An efficient numerical target strength prediction model: validation against analytic solutions
_________________________________________________________________________ 117
Session 13: Habitat Mapping: Procedures and Results ______________________ 118
Towards Joint Use Of Side Scan Sonar And Sub-Bottom Profiler Data For The Automatic
Quantification Of Marine Habitats. Case Study: Lourdas Gulf, Kefalonia Isl., Greece. _____ 118
Automatic classification of bedforms using phase differencing bathymetric sonar ______ 119
Diurnal variation in sediment backscattering properties caused by photosynthesis of
microphytobenthos (Southern Baltic Sea) _______________________________________ 119
Do fish and blue algae blooms coexist in space? __________________________________ 120
vi
UA2014
Acoustic mapping of submerged macrophytes in selected lakes of the Drawieński National
Park _____________________________________________________________________ 120
High-resolution multibeam mapping of habitats in the extremely shallow waters of the
Venice Lagoon _____________________________________________________________ 121
Development of Acoustic Colour technique using multi-frequency swath acoustic
backscatter _______________________________________________________________ 123
MARTA: an AUV for Underwater Cultural Heritage ________________________________ 123
High-resolution multibeam bathymetry applied to underwater research: a case study from
the Lagoon of Venice _______________________________________________________ 124
Session 14: Innovative Approaches for Characterizing Ocean Bottom Properties _ 125
Experimental studies on passive bottom loss estimation from a compact array mounted on
an autonomous underwater vehicle ___________________________________________ 125
Estimating seabed scattering mechanisms via Bayesian model selection ______________ 125
Attenuation of sound in mud sediments________________________________________ 126
Recursive Bayesian synthetic aperture geoacoustic inversion in the presence of motion
dynamics _________________________________________________________________ 126
Bayesian geoacoustic inversion of airgun modal dispersion using a single hydrophone in the
Chukchi Sea _______________________________________________________________ 127
Nonlinear inversion of ship radiated noise in shallow water ________________________ 127
Session 15: Modeling Sonar Performance in Uncertain Environments__________ 129
Mitigation Methods and Techniques for Enhancing Sonar Operational Confidence ______ 129
Modelling multistatic sonobuoy fields in uncertain environments ___________________ 129
Propagation of Acoustic Waves through a Spatially Fluctuating Medium: Theoretical Study of
the Physical Phenomena. ____________________________________________________ 129
Passive Sonar Performance Characterization and Transmission Loss Measurement Using a
Calibrated Mobile Acoustic Source ____________________________________________ 130
Measurement and Model Forecast Comparison of Acoustic Signal-Excess Fluctuations __ 131
Adding error bounds to energy flux estimates of trasmission loss ____________________ 131
Session 16: Outer Continental Shelf, Shelfbreak and Canyon Acoustics_________ 132
Acoustical diagnostics of processes on the shelf of the Black Sea ____________________ 132
A Normal Mode Approach to Modelling Airgun Signals in Australian Coastal Waters ____ 132
Observations of Horizontal Coupling in the Monterey Bay Canyon ___________________ 133
Frequency shift of broadband acoustic signals and its relation to the nonlinear internal
waves in shallow water ______________________________________________________ 133
The “Integrated Ocean Dynamics and Acoustics” (IODA) hybrid modeling effort ________ 133
Session 17: Radiated Noise from Ships and Production Platforms _____________ 135
General characteristics of shipping underwater noise _____________________________ 135
Analysis of underwater acoustic noise measured at the ship bow during sea trials ______ 135
vii
UA2014
Sound generation by the ship propeller in the subcavitation range __________________ 136
Measurement System to Assess Underwater Noise from Vessels and Marine Activities __ 136
Integrated tool for the acoustic assessment and monitoring of marine activities and
operations ________________________________________________________________ 137
SIRAMIS : Preliminary Analysis of Acoustic and Seismic Ship Signatures _______________ 137
Prediction of propeller radiated noise by onboard measurement ____________________ 139
Source level estimates of small cargo ships from measurements in a fjord_____________ 139
Pressure signals of low frequency generated by the ship's hull. _____________________ 140
Session 18: Sensitivity of underwater acoustic observables __________________ 141
Multiple-Source Localization and Environmental Sensitivity ________________________ 141
Sound-speed tomography using angle sensitivity-kernels in an ultrasonic waveguide ____ 141
Travel-time sensitivity kernels in a shallow water environment______________________ 142
A hybrid approach for ocean acoustic tomography based on statistical characterization of
the acoustic signal and the identification of modal arrivals. _________________________ 142
Non-Perturbative Evaluations of Time Sensibility Kernels using Alternative Definitions of
Propagation Delay __________________________________________________________ 143
Session 19: Sonar Performance Modeling and Verification: Applications to Active
and Passive Sonar ___________________________________________________ 144
Overview of the reverberation component of TREX13 _____________________________ 144
Correlation of reverberation with bottom sand waves along the TREX reverberation track 144
A fast algorithm for the computation of incoherent propagation loss for variable water
depth: a validation study ____________________________________________________ 145
Sonar Performance Modeling and Verification: Applications to Active and passive Sonar _ 146
Analysis of sonar detection performance in South China Sea for ASW using ASORPS_____ 146
Sediment acoustics: The need for improvement __________________________________ 148
Dynamic Self-Organizing algorithm for unsupervised segmentation of sidescan sonar images
_________________________________________________________________________ 148
Performance analysis of single receiver Matched-Mode processing for source localization149
Session 20: Sonar Signal and Information Processing _______________________ 150
Performance of Tomographic Arrays in the Arctic using Cramer-Rao Bound
on
Performance as Implemented with OASES ______________________________________ 150
Source motion parameter estimation using direct and multipath arrivals at a pair of
hydrophones ______________________________________________________________ 150
Compressive sensing in acoustics and seismology ________________________________ 151
Sparsity and super-resolution in sound source localization with sensor arrays __________ 151
DoA Estimation Algorithm as Applied to Wideband Processing ______________________ 151
Estimating source spectra from recordings made in a reverberant underwater channel __ 152
Multivariate Distributions of Clutter Levels for Automated Classifiers ________________ 153
Seafloor classification using statistical modeling of wavelet subbands ________________ 153
viii
UA2014
Classifying continuous active sonar echoes for target recognition ____________________ 154
Signal Change Detection Method Used for Mine-Like Objects Segmentation in Sonar Images
_________________________________________________________________________ 154
Underwater moving targets detection using the sequence of acoustic image __________ 155
Passive sonar denoising for diver detection in presence of snapping shrimp ___________ 155
Target AOU Growth Containment using High Duty Cycle Sonar ______________________ 157
Target Doppler Estimation and Range Bias Compensation using LFM High Duty Cycle Sonar
_________________________________________________________________________ 157
A Doppler estimation technique is based on the signals with good correlation properties:
Experimental results ________________________________________________________ 158
Ontology Design for Cooperative Underwater Target Tracking ______________________ 158
Session 21: Soundscapes and Measuring Noise ____________________________ 159
Applying the dynamic soundscape to estimates of signal detection __________________ 159
Passive calibration of soundscapes ____________________________________________ 159
Ship Noise mapping in the North Sea ___________________________________________ 160
Practical spreading laws: the snakes and ladders of shallow water acoustics ___________ 160
Signal Grouping by Correlation of Cepstra _______________________________________ 161
Feasibility of reef health monitoring using passive acoustics ________________________ 161
Sound Maps Of The Dutch North Sea For Natural And Anthropogenic Sound Sources ____ 163
Analysis of soundscapes in the East coast waters of the UK _________________________ 163
Session 22: Synthetic Aperture Sonar: State-of-the-art _____________________ 165
Change detection in topographic structures using interferometric synthetic aperture sonar
_________________________________________________________________________ 165
Comparison of Fusion Approaches for the Displace Phase Centre Antenna Method _____ 165
Alternative SAS processing for gas seep detection ________________________________ 166
Session 23: Tank Experiments __________________________________________ 167
Source level measurement in deep water conditions: Are free field condition met whatever
the source frequency? ______________________________________________________ 167
Calibration of ultrasound transducer heads using short preprocessed ultrasonic pulses __ 167
De-Coherence Effects in Underwater Acoustics: Scaled Experiments. _________________ 168
Acoustic Echo Reduction and Insertion Loss of Tiles _______________________________ 168
A high intensity pulsed laser as a wide band acoustic source for underwater acoustic
applications _______________________________________________________________ 169
Session 24: Three-dimensional sound propagation models __________________ 171
Three-dimensional ray modelling of high-frequency under-ice shallow-water sound
propagation _______________________________________________________________ 171
Numerical applications of a higher order square-root Helmholtz operator splitting method
on modeling three-dimensional sound propagation _______________________________ 171
ix
UA2014
Numerically exact 3D propagation _____________________________________________ 172
An explicit analytical solution for the problem of adiabatic sound propagation along an
underwater canyon with penetrable bottom ____________________________________ 172
Data driven three-dimensional modeling of shallow water waveguide during broadband
acoustic propagation in the presence of internal waves ____________________________ 173
Three-dimensional Split-Step Pade Modelling (Peregrine) __________________________ 173
Mode parabolic equations with mode interaction for the 3D modeling of sound propagation
_________________________________________________________________________ 174
A coastal wedge propagation model including shear in an absorptive bottom __________ 174
Incorporating cross-multiplied terms in a three-dimensional parabolic equation model __ 175
Benchmarking a Three-Dimensional Gaussian Beam Tracing Model __________________ 175
Session 25: Towards Automatic Target Recognition: Detection, Classification and
Modeling (of Underwater Targets) ______________________________________ 176
Iterative Target Recognition for Port Protection System ___________________________ 176
The Main Peculiarities of Automatic Target Recognition ___________________________ 176
Independent views in MIMO sonar systems _____________________________________ 177
Automatic classification for mid-frequency anti-submarine warfare sonars – recognizing
pipelines _________________________________________________________________ 178
Acoustic obstacle detection for safe AUV surfacing _______________________________ 178
Tracking underwater objects using large MIMO sonar systems ______________________ 179
An information-based Johnson’s criteria model for UUV system performance prediction _ 180
Quantifying the complexity in sonar images for MCM performance estimation _________ 180
Towards Automatic Target Recognition in Low-Frequency Sub-Sediment Sonar Imagery _ 181
A GPU Sonar Simulator for Automatic Target Recognition __________________________ 181
Efficient Superellipse Fitting based Contour Extraction for mine-like shape Recognition __ 182
Identifying Contents of Low Profile Targets in a Cluttered Environment _______________ 182
Session 26: Underwater Acoustic Measurement Facilities and Standards _______ 183
The Evolution of Acoustic Transducer Calibration Underwater Sound Reference Division
(USRD) ___________________________________________________________________ 183
Provision Of Standards At Simulated Ocean Conditions ____________________________ 183
Recent advances in methods for the calibration of linear hydrophone arrays at low
frequency ________________________________________________________________ 184
The method of complex free-field calibration of a pressure gradient receiver __________ 184
Underwater techniques to characterize the near scattered acoustic vector field ________ 186
Calibration of hydrophones in the frequency range 1 kHz to 200 kHz using optical method
_________________________________________________________________________ 186
The design of acoustic absorbers for test tank linings______________________________ 186
Calibration methods of the interferometric fiber-optic hydrophone __________________ 187
The calibration and characterisation of autonomous underwater recorders ___________ 187
x
UA2014
Long term underwater third octave sound levels at a busy UK port __________________ 187
Session 27: Underwater Acoustic Studies in Asian Seas _____________________ 189
Observation of ambient noise induced by the internal solitary wave in the center of Kuroshio
northeast of Taiwan ________________________________________________________ 189
Some characteristics of bottom scattering provided by single-mode reverberation______ 189
Measured Channel Impulse Responses for a Mobile Source in the Northeastern Sea off
Taiwan ___________________________________________________________________ 190
Observe Seismic Activities and Ambient Noise of Underwater Acoustic Data from MACHO
Hydrophone ______________________________________________________________ 191
An Overview of Ocean Ambient Noise around Taiwan: Measurement and Analysis______ 192
Large-amplitude internal solitary wave of the second mode in Luzon Strait: observations and
numeric modeling of its propagation ___________________________________________ 193
Reverberation Modeling in Range-dependent Waveguide __________________________ 193
Session 29: Underwater Communication and Networking ___________________ 194
High rate uplink acoustic communication from AUVs to surface platform _____________ 194
Clock Synchronization in Underwater Acoustic Networks During Payload Data Exchange _ 194
Multi-user communication by adaptive time reversal in deep ocean _________________ 195
Stealth underwater acoustic communications based upon steganography techniques___ 196
The design of wide band transducers for underwater acoustic communication _________ 196
Measurement and Modelling of Fading in Ultrasonic Underwater Channels ___________ 197
Joint Pilot based Channel Estimation with Selected-Mapping to Reduce PAPR in Underwater
Acoustic MIMO-OFDM System without Side Information __________________________ 198
Ultrasonic Diversity OFDM Transceiver architecture with Impulsive Noise Cancelling for
shallow sea communication __________________________________________________ 198
PIC-DDFE-IDMA Detection For Uplink Shallow Water Acoustic Channels ______________ 199
Session 30: Underwater Unexploded Ordnance (UXO) Detection and Remediation
__________________________________________________________________ 200
First results for buried object detection from the “sounding ammunition (SOAM)” 2013
experiment _______________________________________________________________ 200
Target detection with low-cost imaging sonars ___________________________________ 200
Low frequency feature extraction for target discrimination on a Biosonar dataset ______ 201
State of the Art in Commercial Offshore Unexploded Ordnance Detection ____________ 201
Offshore Unexploded Ordnance Recovery and Disposal____________________________ 202
Using A 3D Subbottom Profiler For AUV-Based Pipeline Detection And Localization _____ 203
Session 31: Unmanned Vehicles (AUV, USV and Gliders) for Underwater Acoustic
Surveillance and Monitoring ___________________________________________ 204
Monitoring the underwater acoustic pressure field using two spatially-separated
hydrophones with application to forward-aft sensors onboard an undersea glider ______ 204
xi
UA2014
Passive acoustics embedded on gliders – Weather observation through ambient noise. __ 205
Small vessel detection through the use of an underwater glider _____________________ 205
Automated Detection of Fishing Vessels using Smart Hydrophones on an Underwater Coastal
Glider ____________________________________________________________________ 206
NeXOS objectives in multi-platform underwater passive acoustics ___________________ 208
The fusion of digital terrain models measured from multiple acoustic sensors – Application
to the DAURADE autonomous underwater vehicle ________________________________ 208
Real time improvement of the seabed mapping with AUV-borne sensors using statistical
analysis __________________________________________________________________ 209
Real-time acoustic monitoring of baleen whales from autonomous platforms __________ 210
SONOBOT - an autonomous unmanned surface vehicle for hydrographic surveys,
hydroacoustic communication and positioning in tasks of underwater acoustic surveillance
and monitoring ____________________________________________________________ 210
The Persistent Maritime Monitoring System (PMMS)______________________________ 211
Virtual Ocean Testbed For Autonomous Undersea Sensing Networks _________________ 213
Networking underwater, surface and air vehicles: tools and experimentation __________ 213
CMRE's use of AUVs and USVs for the demonstration of network concepts for multistatic
active ASW _______________________________________________________________ 214
Session 32: Vector Sensors: Development and Applications__________________ 215
An experimental study on DEMON spectrum direction estimation of multi-target with a low
frequency vector hydrophone ________________________________________________ 215
Signal Processing for circular Vector-Sensor Array mounted around a cylindrical baffle __ 215
Broadband DOA estimation in phase modal space for circular acoustic Vector-Sensor Array
_________________________________________________________________________ 216
Patch near-field acoustical holography based on vector hydrophone array ____________ 216
A Robust Noise Sources Localization and Identification method based on Vector Sensor
Array ____________________________________________________________________ 217
Research Vector Hydrophone Measurement System ______________________________ 217
Measurement of the vector field and its application on Geo-acoustic inversion ________ 219
Particle filtering of vertical impedance due to moving ship noise for sediment
characterization ___________________________________________________________ 219
Session 33: Bioacoustics ______________________________________________ 221
Bioacoustic Absorption Spectroscopy of physoclists _______________________________ 221
Bluefin tuna behavioural response to anthropogenic noise in floating sea cages ________ 221
Acoustical biomass estimation results in Mediterranean aquaculture sea cages ________ 222
Sound propagation over an elastic bottom –particle motions caused by seismic interface
waves ____________________________________________________________________ 222
Acoustics marine survey of distribution and intensity of species in Arabian Sea during
monsoon variation (by mid-water trawl survey) __________________________________ 222
xii
UA2014
Session 34: Underwater Communication _________________________________ 224
Data Smoothing algorithms for Phased Array Doppler Log __________________________ 224
Comparison of Modulation Techniques for Parametric Underwater Communications ___ 224
Inquiring Flooding Algorithm For Underwater Acoustic Sensor Self-Organization Network 225
DSP implementation of Turbo Equalization based Underwater Acoustic Modem________ 225
Orthogonal Multicarrier Underwater Acoustic Communication Experiments in River, Lake
and Shallow Sea ___________________________________________________________ 226
Study on the impact of multi-path channel to the horizontal uniform linear array
beamforming ______________________________________________________________ 226
Development/Introduction of the bio-logging system to realize high data recovery rate using
acoustic communication _____________________________________________________ 228
Full-duplex, relative clock based and collision free protocol for underwater acoustic
networks _________________________________________________________________ 228
Session 35: Acoustic Modelling_________________________________________ 230
Comparison of RAM-based Calculations of Acoustic Propagation with Measured Data in the
East China Sea _____________________________________________________________ 230
Single global empirical equation for prediction of the peak pressure level of airgun arrays
signals in different marine environments _______________________________________ 230
Peculiarities of the broadband interference pattern in a shallow sea with a substantially
varying bottom relief _______________________________________________________ 231
Acoustic remote sensing of internal Kelvin waves dynamics in a stratified lake _________ 231
On the importance of uncertain sea bottom parameters for the prediction of pile driving
noise ____________________________________________________________________ 232
Modeling pulse propagation in a wedge environment with range-dependent geoacoustic
parameters _______________________________________________________________ 232
Finite Difference Time Domain Method for Acoustic Waves in Attenuate and Absorptive
Medium for Layered Underwater Acoustic Environments __________________________ 233
Sound fluctuations in the presence of nonlinear internal waves moving along acoustic track
in shallow water ___________________________________________________________ 233
Statistical Characterization of Wideband Channel Impulse Response Observations in Shallow
Water ____________________________________________________________________ 234
Modeling Reverberation Time Series Based on Full Wave Reverberation model ________ 234
Acoustic Backscattering from Layered Interfaces using Finite Elements _______________ 235
Session 36: Acoustic Inversions ________________________________________ 236
Geoacoustic inversion using pile driving pulse and surface ship noise of opportunity based
on single vector sensor ______________________________________________________ 236
Inferring ocean temperature variations from shipping noise ________________________ 236
Marine mammal’s directivity in geoacoustic inversion scheme ______________________ 237
xiii
UA2014
Bayesian reconstruction of seafloor shape from side-scan sonar measurements using a
Markov Random Field _______________________________________________________ 237
AB Wood Medal Lecture ______________________________________________ 239
From the pole to the equator: Utilizing a screw dislocation in an acoustic wavefront. ____ 239
Posters ____________________________________________________________ 240
Multi-ary code shift keying direct sequence spread spectrum using in underwater acoustic
communication ____________________________________________________________ 240
Direct-sequence spread spectrum underwater acoustic communications with Turbo
equalization in time-varying channels __________________________________________ 240
Underwater Acoustic Communication System Simulation Based On Gaussian Beam Method
_________________________________________________________________________ 241
The Study of Time Delay Estimation Technology Based On The Cross-spectrum Method _ 242
The Study of Passive Ranging Technology Based on Three elements Vector Array _______ 242
Adaptive Despeckling Method For Sas Images In NSCT Domain ______________________ 243
Influence of ship radiated noise level directivity on the assessment of underwater noise
maps ____________________________________________________________________ 244
The fluid noise analysis on underwater high speed small vehicles ____________________ 244
The comparison of two way sound propagation in Hashirimizu Port __________________ 245
A combined GIS-2DFTT multi-parameter analysis of very high resolution bathymetric data: a
case study from the Venice lagoon ____________________________________________ 246
Passive acoustic detections of odontocetes in the Ionian and Aegean Seas, Greece. _____ 246
Sound Pressure Field Focused by Off-Axis Aplanatic Straubel Acoustic Mirror __________ 247
Preliminary investigation on the potential of using low power ultrasound to induce low
frequency vibrations on an immersed object. ____________________________________ 248
Parallel Activities-Meeting Point ____________________________________ 249
Monday 23rd June 2014, 17:40-18:40 ____________________________________ 249
Institute of Acoustics, Underwater Acoustics Group, AGM meeting __________________ 249
Wednesday 25th June, 14:15-18:00 ______________________________________ 249
CTBTO’s Hydroacoustic Network ______________________________________________ 249
Conference Venue Map ___________________________________________ 251
Exhibition ______________________________________________________ 252
Teledyne RESON A/S _________________________________________________ 252
RTSYS _____________________________________________________________ 254
Ocean Sonics _______________________________________________________ 256
NORBIT____________________________________________________________ 258
xiv Program and Book of Abstracts
UA2014
EvoLogics GmbH ____________________________________________________ 260
MarSensing Lda. ____________________________________________________ 262
Wildlife Acoustics ___________________________________________________ 264
Sessions Timetable _______________________________________________ 266
Sessions List ____________________________________________________ 267
Social Program __________________________________________________ 269
Index of Session Organizers & Presenters _____________________________ 270
xv
UA2014
Welcome
Dear Participant,
A warm welcome to Greece, to Rhodes and to UA2014. the 2 nd International Conference and
Exhibition on Underwater Acoustics, which was born out of the merger of the ECUA and UAM
Conferences.
It is the ambition of the Scientific Committee and the Organizers of UA2014 to be able to continue
the long history of successes of the ECUA and UAM Conferences.
This Book of Abstracts includes 282 abstracts; among them a very large number of papers
offering significant contributions to the most important fields of Underwater Acoustics. Invited
as well as contributed papers are all allotted 20 minutes for presentations, including time for
questions.
At UA2014 we have established a ‘Meeting Point’, to facilitate holding of specialists’ meetings,
shorter seminars, handing out of medals and awards, establishing of scientific co-operations,
working out of applications for funding from EU and other bodies etc. as parallel activities. The
activities hosted by the ‘Meeting Point’ this year are a CTBTO Expert Group Forum on Innovation
and the Annual Meeting of the Underwater Acoustics Group of the UK Institute of Acoustics .We
are always very happy for our cooperation with the UK Institute of Acoustics and for hosting their
medal awards, and this year the AB Wood Medal Award will be presented on Friday morning,
June 27th.
We also continue the competition among young scientists. An award will be given during the
closing ceremony on Friday morning to the best paper presented by a graduate student.
It is obvious that the success of UA2014 is due to the efforts and hard work of the Structured
Session Organizers and their invitees, and we are most grateful to them.
We would like to thank the Office of Naval Research Global for their continuous support and
financial contributions throughout the history of these Conferences. Teledyne-RESON has also
for this year, as well as for the previous 5 UAM and UA conferences, provided the conference bags,
which is greatly appreciated. We would also like to thank Evologics GmbH for their financial
contribution to this year’s event. Our exhibitors are very welcome and their participation in
UA2014 is much appreciated.
We also appreciate the sponsoring of UA2014 by two major societies, EAA and ASA. And of
course, we would like to acknowledge the continuous support of IACM-FORTH. Finally, a warm
thank you to all members of the Scientific Committee for their helpful suggestions and advices..
We wish you an interesting, fruitful and enjoyable UA2014
John S. Papadakis and Leif Bjørnø
1
UA2014
Conference Chairmen
John S. Papadakis and Leif Bjørnø
Technical Issues
Yorgis Androulakis, IACM - FORTH
Secretariat
Yiota Rigopoulou, IACΜ - FORTH
Supported by
Office of Naval Research Global, London, UK
IACM-FORTH
RESON A/S, Denmark
EvoLogics GmbH, Germany
Endorsed by
Institute of Acoustics, UK
European Acoustics Association, EAA
Acoustical Society of America, ASA
2
UA2014
Scientific Committee
John Papadakis - FORTH, Greece (Co-Chairman)
Leif Bjørnø - UltraTech Holding, Denmark (Co-Chairman)
Michael Ainslie - TNO, The Netherlands
Tuncay Akal - Tubitak, Turkey
Philippe Blondel - University of Bath, UK
David Bradley - ARL, Penn State University, USA
Mike Buckingham - MPL, Scripps, USA
Andrea Caiti - ISME/DII, Italy
Chris Capus - Heriot-Watt Univ., Edinburgh, Scotland
Ross Chapman - University of Victoria, Canada
Chi-Fang Chen - National Taiwan University, Taiwan
Nick Chotiros - ARL, Univ. of Texas at Austin, USA
Peter Gerstoft - MPL, Scripps, USA
Grazyna Grelowska - Naval Academy, Gdynia, Poland
Hirouyki Hachiya - Tokyo Inst. Technology, Japan
Roy-Edgar Hansen - FFI, Norway
Jean-Pierre Hermand - Univ. Libre de Bruxelles, Belgium
Charles Holland - ARL, Penn State University, USA
Kirk Jenne - NSWC HQ, Washington DC, USA
Sergio Jesus - Univ. Algarve, Faro, Portugal
Arata Kaneko - Hiroshima University, Japan
Eugeniusz Kozaczka - Techn. Univ. Gdansk, Poland
Tim Leighton - University of Southampton, UK
Tony Lyons - ARL, Penn State University, USA
Qihu Li - Institute of Acoustics, Acad. Sinica, China
Alain Maguer - CMRE, NATO Sci. & Techn. Org., Italy
Purnima Ratilal - Northeastern Univ., Boston, USA
Jean-Pierre Sessarego - CNRS-LMA, Marseille, France
Martin Siderius - Portland State University, USA
Dick Simons - Delft Univ. Technology, The Netherlands
Emmanuel Skarsoulis - F.O.R.T.H., Greece
Michael Taroudakis - University of Crete & F.O.R.T.H., Greece
Alessandra Tesei - NATO - STO CMRE, Italy
Peter Thorne - National Oceanographic Centre, Liverpool, UK
3
UA2014
Session Organizers
4
Michael Ainslie - TNO, The Netherlands
Tuncay Akal - Tubitak, Turkey
Philippe Blondel - University of Bath, UK
Daniel Brown - APL, Pennsylvania State Univ., USA
Andrea Caiti - ISME/DII, Italy
Ross Chapman - University of Victoria, Canada
Chi-Fang Chen - National Taiwan University, Taiwan
Lee Culver - ARL, Penn State University, USA
Julian Deeks - Dstl Portsdown West, UK
Eric Delory - PLOCAN, Spain
Frank Ehlers - WTD 71 - FWG 630, Germany
Dale Ellis - Canada
Dominique Fattaccioli - DGA Techniques Navales, France
John Fawcett - DRDC Atlantic, Canada
Brian Ferguson - Defence Science and Technology Organisation, Australia
Alexander Gavrilov - Curtin University, Australia
Peter Gerstoft - MPL, Scripps, USA
Ryan Goldhahn - NATO -CMRE,
Grazyna Grelowska - Naval Academy, Gdynia, Poland
Johannes Groen - Atlas Elektronik, GmbH
Hiroyuki Hachiya - Tokyo Inst. Technology, Japan
Roy-Edgar Hansen - FFI, Norway
Georgios Haralabus - CTBTO
Kevin Heaney - OASIS Inc, USA
Jean-Pierre Hermand - Univ. Libre de Bruxelles, Belgium
Oliver Hinton - Newcastle University, UK
Charles Holland - ARL, Penn State University, USA
Alan Hunter - TNO, The Netherlands
Wolfgang Jans - WTD 71, Germany
Sergio Jesus - Univ. Algarve, Faro, Portugal
Arata Kaneko - Hiroshima University, Japan
Tim Leighton - University of Southampton, UK
Paul Lepper - Loughborough University, UK
Jiyuan Liu - Institute of Acoustics, Chinese Academy of Sciences
Bo Lövgren - Saab Dynamica AB, Sweden
Jim Lynch - Woods Hole Oceanographic Institution
Alain Maguer - CMRE, NATO Sci. & Techn. Org., Italy
Jennifer Miksis-Olds - ARL, Penn State University, USA
Peter Nielsen - CMRE, NATO
David Nunn - Dstl Porton Down, UK
Yan Pailhas - Heriot-Watt University, UK
Anthony Paolero - Naval Undersea Warfare Center, USA
Michael Porter - HLS Research Inc., USA
Mark Prior - CTBTO
Purnima Ratilal - Northeastern Univ., Boston, USA
Mike Richardson - NRL, Stennis Space Center, USA
Stephen Robinson - Acoustics Group, National Physical Laboratory Teddington, UK
Stefan Schael - WTD 71, Germany
Arne Schulz - WTD 71 - FWG 630, Germany
UA2014
Jean-Pierre Sessarego - CNRS-LMA, Marseille, France
Martin Siderius - Portland State University, USA
Emmanuel Skarsoulis - F.O.R.T.H., Greece
Chris Strode - NATO Undersea Research Center (NURC)
Frederic Sturm - LMFA, France
Jaroslaw Tegowski - Institute of Oceanography, University of Gdansk, Poland
Alessandra Tesei - NATO - STO CMRE, Italy
Charalampos Tsimenidis - Newcastle University, UK
Duncan Williams - Dstl Porton Down, UK
Mario Zampolli - CTBTO
Adam Zielinski - University of Victoria, Canada
5
UA2014
Conference Program
Monday 23rd June 2014
Registration
Location: Lobby, Registration Area - 08:30-10:00
Welcome
Location: Lecture Room A – Monday 23rd June 2014, 10:00-10:30
Coffee break: 10:30-10:45
6
UA2014
Session 14: Innovative Approaches for Characterizing Ocean Bottom Properties
Organizer: Martin Siderius, Sergio Jesus, Peter Nielsen, Jean-Pierre Hermand and
Ross Chapman
Location: Lecture Room A - Monday 23rd June 2014
Chair: Sergio Jesus and Ross Chapman
10:45 - 11:05
Lanfranco Muzi, Martin Siderius, Peter Nielsen
Experimental studies on passive bottom loss estimation from a compact array mounted
on an autonomous underwater vehicle (invited)
Presenter: Lanfranco Muzi
11:05 - 11:25
Gavin Steininger, Stan Dosso, Charles Holland, Jan Detmer
Estimating seabed scattering mechanisms via Bayesian model selection
Presenter: Gavin Steininger
11:25 - 11:45
Allan D. Pierce, William L. Siegmann, Joseph O. Fayton
Attenuation of sound in mud sediments (invited)
Presenter: Allan D. Pierce
11:45 - 12:05
Bien Aik Tan, Peter Gerstoft, Caglar Yardim, William S. Hodgkiss
Recursive Bayesian synthetic aperture geoacoustic inversion in the presence of motion
dynamics (invited)
Presenter: Peter Gerstoft
12:05 - 12:25
Graham Warner, Stan Dosso, David Hannay, Jan Dettmer
Bayesian geoacoustic inversion of airgun modal dispersion using a single hydrophone in
the Chukchi Sea
Presenter: Graham Warner
12:25 - 12:45
Steven E. Crocker, Peter L. Nielsen, James H. Miller, Martin Siderius
Nonlinear inversion of ship radiated noise in shallow water (invited)
Presenter: Steven E. Crocker
7
UA2014
Session 30: Underwater Unexploded Ordnance (UXO) Detection and
Remediation
Organizer: Mike Richardson and Wolfgang Jans
Location: Lecture Room B - Monday 23rd June 2014
Chair: Wolfgang Jans
10:45 - 11:05
David Rose, Wolfgang Jans
First results for buried object detection from the “sounding ammunition (SOAM)” 2013
experiment (invited)
Presenter: David Rose
11:05 - 11:25
Christian De Moustier, Joe Calantoni
Target detection with low-cost imaging sonars (invited)
Presenter: Christian De Moustier
11:25 - 11:45
Tara Leblanc, John Fawcett, Yan Pailhas, Duncan Williams
Low frequency feature extraction for target discrimination on a Biosonar dataset (invited)
Presenter: Tara Leblanc
11:45 - 12:05
Kay Winkelmann
State of the Art in Commercial Offshore Unexploded Ordnance Detection (invited)
Presenter: Kay Winkelmann
12:05 - 12:25
Jan Koelbel
Offshore Unexploded Ordnance Recovery and Disposal (invited)
Presenter: Jan Koelbel
12:25 - 12:45
Aneta Nikolovska
Using A 3D Subbottom Profiler For AUV-Based Pipeline Detection And Localization
(invited)
Presenter: Max Abildgaard
8
UA2014
Applications
Organizer: Alessandra Tesei
Location: Lecture Room C - Monday 23rd June 2014
Chair: Alessandra Tesei
10:45 - 11:05
S. Viola, G. Riccobene, F. Simeone
The SMO antenna: status and first results (invited)
Presenter: S. Viola
11:05 - 11:25
Tomaso Gaggero, Mirko Bassetti, Erica Firenze, Alessandra Tesei, Andrea Trucco
Processing strategies for evaluating the ship radiated noise using an underwater vertical
array (invited)
Presenter: Tomaso Gaggero
11:25 - 11:45
Toyoki Sasakura
Development of ultra sideband transducer
Presenter: Toyoki Sasakura
11:45 - 12:05
Takenobu Tsuchiya, Tetsuo Anada, Nobuyuki Endoh, Sayuri Matsumoto, Kazuyoshi Mori
Basic Study of Rhomboidal Acoustic Lens Constructed with Phononic Crystal
Presenter: Takenobu Tsuchiya
9
UA2014
Session 26a: Underwater Acoustic Measurement Facilities and Standards
Organizer: Anthony Paolero and Stephen Robinson
Location: Lecture Room D - Monday 23rd June 2014
Chair: Stephen Robinson
10:45 - 11:05
Anthony Paolero, Victor Evora
The Evolution of Acoustic Transducer Calibration Underwater Sound Reference Division
(USRD) (invited)
Presenter: Anthony Paolero
11:05 - 11:25
G A Beamiss, S P Robinson, Gary Hayman
Provision Of Standards At Simulated Ocean Conditions (invited)
Presenter: G A Beamiss
11:25 - 11:45
Steven E. Crocker
Recent advances in methods for the calibration of linear hydrophone arrays at low
frequency (invited)
11:45 - 12:05
Alexander Isaev, Anton Matveev
The method of complex free-field calibration of a pressure gradient receiver (invited)
Presenter: Anton Matveev
Lunch break: 12:45 – 14:15
10
UA2014
Session 4a: Acoustic Tomography in Shallow Seas, Benthic and Terrestrial
Waters
Organizer: Jean-Pierre Hermand, Arata Kaneko and Hiroyuki Hachiya
Chair: Xiao-Hua Zhu
14:15 - 14:35
Jean-Pierre Hermand, Qunyan Ren
Passive acoustic tomography of fluid-mud processes on the Amazon continental shelf
(invited)
Presenter: Jean-Pierre Hermand
14:35 - 14:55
Chen-Fen Huang, Naokazu Taniguchi, Jin-Yuan Liu
Horizontal Ocean Current Tomography with Iterative Model Weighting Constraint
(invited)
Presenter: Chen-Fen Huang
14:55 - 15:15
Ju Lin, Wang Huan, Wang Fengbao
Estimation of temperature information based on ocean ambient noise measurement
(invited)
Presenter: Ju Lin
15:15 - 15:35
Xiao-Hu Zhu, Xiaopeng Fan, Ju Lin, Chuanzheng Zhang
A Coastal Acoustic Tomography Experiment for Tidal Current Measurement in the South
of Jiaozhou Bay (invited)
Presenter: Xiao-Hu Zhu
11
UA2014
Session 19a: Sonar Performance Modeling and Verification: Applications to
Active and Passive Sonar
Organizer: Michael Ainslie, Charles Holland, Dale Ellis and Kevin Heaney
Chair: Michael Ainslie
14:15 - 14:35
Brian T. Hefner, Dajun Tang
Overview of the reverberation component of TREX13 (invited)
Presenter: Brian T. Hefner
14:35 - 14:55
Dale D Ellis, Sean P Pecknold, John R Preston, Jie Yang
Correlation of reverberation with bottom sand waves along the TREX reverberation track
(invited)
Presenter: Dale Ellis
14:55 - 15:15
H. Ozkan Sertlek, Michael A. Ainslie
A fast algorithm for the computation of incoherent propagation loss for variable water
depth: a validation study (invited)
Presenter: H.Özkan Sertlek
15:15 - 15:35
Xavier Cristol, Bruno Chalindar
Sonar Performance Modeling and Verification: Applications to Active and passive Sonar
(invited)
Presenter: Xavier Cristol, Bruno Chalindar
15:35 - 15:55
Chiao-Ming Peng, Yuan-Ying Chang, Yin-Ying Fang, Chung-Wu Wang, Chi-Fang Chen
Analysis of sonar detection performance in South China Sea for ASW using ASORPS
(invited)
Presenter: Chiao-Ming Peng
12
UA2014
Session 6a: Acoustics of Bubbles, Oil and Gas
Organizer: Tim Leighton and Lee Culver
Chair: Tim Leighton and Lee Culver
14:15 - 14:35
Tim Leighton, Mengyang Zhu, Peter Birkin
A technique to measure the real surface tension on a bubble wall (invited)
Presenter: Tim Leighton
14:35 - 14:55
Benoit J. P. Berges, Timothy G. Leighton, Paul R. White
Passive acoustic quantication of gas releases
Presenter: Benoît J. P. Bergès
14:55 - 15:15
Piotr Majewski, Zygmunt Klusek
Acoustical classification of the shallow sediment gaseous structures in the Southern Baltic
Sea (invited)
Presenter: Piotr Majewski
15:15 - 15:35
H. Dogan, T. G. Leighton, P. R. White
Comparison of theories for acoustic wave propagation in gassy marine sediments
Presenter: H. Dogan
15:35 - 15:55
Mark S. Wochner, Kevin M. Lee, Preston S. Wilson
Attenuation of low frequency underwater noise using arrays of air-filled resonators
(invited)
Presenter: Mark S. Wochner
13
UA2014
Session 26b: Underwater Acoustic Measurement Facilities and Standards
Chair: Anthony Paolero
14:15 - 14:35
Robert J. Barton Iii, Georges A. Dossot, Kevin B. Smith
Underwater techniques to characterize the near scattered acoustic vector field (invited)
Presenter: Robert Barton
14:35 - 14:55
Shiquan Wang, Yi Chen, Yongjun Huang, Yuebing Wang
Calibration of hydrophones in the frequency range 1 kHz to 200 kHz using optical method
(invited)
Presenter: Shiquan Wang
14:55 - 15:15
Yuebing Wang, Huifeng Zheng
The design of acoustic absorbers for test tank linings (invited)
Presenter: Yuebing Wang
15:15 - 15:35
Yi Chen, Jun Zhang, Min Zhang
Calibration methods of the interferometric fiber-optic hydrophone (invited)
Presenter: Yi Chen
15:35 - 15:55
Gary Hayman, Stephen P. Robinson, Paul A. Lepper
The calibration and characterisation of autonomous underwater recorders (invited)
Presenter: Stephen P. Robinson
15:55 - 16:15
Joanne K Garrett, Matthew J Witt, Lars Johanning
Long term underwater third octave sound levels at a busy UK port
Presenter: Joanne K Garrett
Coffee break: 16:15 – 16:30
14
UA2014
Session 4b: Acoustic Tomography in Shallow Seas, Benthic and Terrestrial
Waters
Chair: Jean-Pierre Hermand
16:30 - 16:50
Kiyosi Kawanisi, Mahdi Razaz
Acoustic investigations of unsteady salinity intrusion in a diversion channel (invited)
Presenter: Kiyosi Kawanisi
16:50 - 17:10
Arata Kaneko, Chuanzheng Zhang, Xiaohua Zhu, Noriaki Gohda
Vertical profiling of temperature and velocity from the quite limited data set of coastal
acoustic tomography (invited)
Presenter: Arata Kaneko
17:10 - 17:30
Chuanzheng Zhang, Arata Kaneko, Xiaohua Zhu, Noriaki Gohda
Tomographic mapping of coastal upwelling generated in Hiroshima Bay, Japan
Presenter: Chuanzheng Zhang
Session 19b: Sonar Performance Modeling and Verification: Applications to
Active and Passive Sonar
Chair: Dale Ellis
16:30 - 16:50
Nicholas P. Chotiros, Marcia J. Isakson
Sediment acoustics: The need for improvement (invited)
Presenter: Nicholas P. Chotiros
16:50 - 17:10
Ahmed Nait-Chabane, Benoit Zerr, Gilles Le Chenadec
Dynamic Self-Organizing algorithm for unsupervised segmentation of sidescan sonar
images
Presenter: Ahmed Nait-Chabane
17:10 - 17:30
Yann Le Gall, Francois-Xavier Socheleau, Julien Bonnel
Performance analysis of single receiver Matched-Mode processing for source localization
(invited)
Presenter: Yann Le Gall
15
UA2014
Session 6b: Acoustics of Bubbles, Oil and Gas
Chair: Tim Leighton and Lee Culver
16:30 - 16:50
Tobias Bohne, Cristina Diaz-Cereceda, Tanja Griessmann, Raimund Rolfes
Numerical modelling of a bubble curtain
Presenter: Tobias Bohne
16:50 - 17:10
Xavier Cristol
The role of air bubbles in acoustic surface loss confirmed by historical data (1949-2005)
about attenuation excess of sound in oceanic surface channels
Presenter: Xavier Cristol
16
UA2014
Session 16: Outer Continental Shelf, Shelfbreak and Canyon Acoustics
Organizer: Jim Lynch
Chair: YT Lin
16:30 - 16:50
Valeriy Goncharov, Boris Kuryanov, Andrey Serebryany
Acoustical diagnostics of processes on the shelf of the Black Sea
Presenter: Andrey Serebryany
16:50 - 17:10
Matthew W Koessler, Alexander N Gavrilov, Alec J Duncan
A Normal Mode Approach to Modelling Airgun Signals in Australian Coastal Waters
Presenter: Matthew W Koessler
17:10 - 17:30
Kevin B. Smith, James M. Upshaw, Timothy Kubisak
Observations of Horizontal Coupling in the Monterey Bay Canyon
Presenter: Kevin B. Smith
17:30 - 17:50
Mohsen Badiey, Lin Wan
Frequency shift of broadband acoustic signals and its relation to the nonlinear internal
waves in shallow water
Presenter: Mohsen Badiey
17:50 - 18:10
Tf Duda, Yt Lin, Ae Newhall, Kr Helfrich, Wg Zhang, M Badiey, Pfj Lermusiaux, Jf Lynch
The “Integrated Ocean Dynamics and Acoustics” (IODA) hybrid modeling effort
Presenter: YT Lin
17
UA2014
Tuesday 24th June 2014
Session 10: Comprehensive Nuclear-Test-Ban Treaty Monitoring
Organizer: Georgios Haralabus and Mario Zampolli
Location: Lecture Room A - Tuesday 24th June 2014
Chair: Georgios Haralabus and Mario Zampolli
8:30 - 8:50
Mario Zampolli, Georgios Haralabus, Jeremy P. Stanley, Lucie Pautet, Mark K. Prior, Patrick
Marsaleix, Florent Lyard, Kevin D. Heaney, Richard Campbell
Results from ocean currents and acoustic propagation modelling studies in support of the
installation of CTBTO Hydroacoustic station HA04, Crozet Islands, France (invited)
Presenter: Mario Zampolli, Georgios Haralabus
8:50 - 9:10
Stephen M. Nichols, David L. Bradley
Using correlation matrices to identify temporal characteristics of ambient noise (invited)
Presenter: Stephen M. Nichols
9:10 - 9:30
Haru Matsumoto, Robert.P. Dziak, Delwayne Bohnenstiehl, Jean Tournadre, Joe Haxel, T-K
Lau, Matt Fowler
Antarctic’s Siren Call: The Sound of Icebergs (invited)
Presenter: Haru Matsumoto
9:30 - 9:50
Evers, L.G., D. Brown, K. Heaney, J.D. Assink, P.S.M. Smets, M. Snellen
Evanescent wave coupling in a geophysical system: Airborne acoustic signals from the Mw
8.1 Macquarie Ridge earthquake (invited)
Presenter: Läslo Evers
9:50 - 10:10
Kevin D. Heaney, Richard L. Campbell
Basin Scale Time-Domain Modelling for CTBTO Tracking (invited)
Presenter: Kevin D. Heaney
18
UA2014
Session 35a: Acoustic Modelling
Location: Lecture Room B - Tuesday 24th June 2014
Chair: Marcia Isakson
8:30 - 8:50
Clare Nadig, David Bradley
Comparison of RAM-based Calculations of Acoustic Propagation with Measured Data in
the East China Sea
Presenter: Clare Nadig
8:50 - 9:10
Marta Galindo-Romero, Alexander Gavrilov, Alec J Duncan
Single global empirical equation for prediction of the peak pressure level of airgun arrays
signals in different marine environments
Presenter: Marta Galindo-Romero
9:10 - 9:30
E.L. Borodina, Yu.V. Petukhov
Peculiarities of the broadband interference pattern in a shallow sea with a substantially
varying bottom relief
Presenter: E.L. Borodina
9:30 - 9:50
Boris Katsnelson, Andrey Lunkov, Ilia Ostrovsky
Acoustic remote sensing of internal Kelvin waves dynamics in a stratified lake
Presenter: Boris Katsnelson
9:50-10:10
Tristan Lippert, Kristof Heitmann, Marcel Ruhnau, Stephan Lippert, Otto Von Estorff
On the importance of uncertain sea bottom parameters for the prediction of pile driving
noise
Presenter: Tristan Lippert
10:10 - 10:30
Melanie E. Austin, N. Ross Chapman
Modeling pulse propagation in a wedge environment with range-dependent geoacoustic
parameters
Presenter: Melanie Austin
19
UA2014
Session 24a: Three-dimensional sound propagation models
Organizer: Michael Porter and Frederic Sturm
Location: Lecture Room D - Tuesday 24th June 2014
Chair: Michael Porter and Frederic Sturm
8:30 - 8:50
Sven Ivansson, Ilkka Karasalo, Erland Sangfelt
Three-dimensional ray modelling of high-frequency under-ice shallow-water sound
propagation (invited)
Presenter: Sven Ivansson
8:50 - 9:10
Ying-Tsong Lin
Numerical applications of a higher order square-root Helmholtz operator splitting method
on modeling three-dimensional sound propagation (invited)
Presenter: Ying-Tsong Lin
9:10 - 9:30
Ahmad T. Abawi
Numerically exact 3D propagation (invited)
Presenter: Ahmad T. Abawi
9:30 - 9:50
Pavel Petrov
An explicit analytical solution for the problem of adiabatic sound propagation along an
underwater canyon with penetrable bottom (invited)
Presenter: Pavel Petrov
9:50 - 10:10
Data driven three-dimensional modeling of shallow water waveguide during broadband
acoustic propagation in the presence of internal waves
10:10 - 10:30
Three-dimensional Split-Step Pade Modelling (Peregrine)
20
UA2014
Session 31a: Unmanned Vehicles (AUV, USV and Gliders) for Underwater
Acoustic Surveillance and Monitoring
Organizer: Alain Maguer, Brian Ferguson and Eric Delory
Chair: Alain Maguer, Eric Delory
10:45 - 11:05
Brian G. Ferguson, Kam W. Lo
Monitoring the underwater acoustic pressure field using two spatially-separated
hydrophones with application to forward-aft sensors onboard an undersea glider (invited)
Presenter: Brian Ferguson
11:05 - 11:25
Pierre Cauchy, Pierre Testor, Laurent Mortier, Marie-Noelle Bouin
Passive acoustics embedded on gliders – Weather observation through ambient noise.
(invited)
Presenter: Pierre Cauchy
11:25 - 11:45
Tesei, A., Been, R., Troiano, L., Dymond, R., Maguer, A.
Small vessel detection through the use of an underwater glider (invited)
Presenter: Tesei A.
11:45 - 12:05
Ray Mahr, Mark Wood
Automated Detection of Fishing Vessels using Smart Hydrophones on an Underwater
Coastal Glider (invited)
Presenter: Ray Mahr, Mark Wood
21
UA2014
Session 35b: Acoustic Modelling
Chair: Ross Chapman
10:45 - 11:05
J. Lakziz, S.Othmane, S.Ouaskit
Finite Difference Time Domain Method for Acoustic Waves in Attenuate and Absorptive
Medium for Layered Underwater Acoustic Environments
Presenter: S.Ouaskit
11:05 - 11:25
Boris Katsnelson, Valery Grigorev, Jixing Win
Sound fluctuations in the presence of nonlinear internal waves moving along acoustic
track in shallow water
11:25 - 11:45
W.S. Hodgkiss, H.C. Song, D.E. Ensberg
Statistical Characterization of Wideband Channel Impulse Response Observations in
Shallow Water
Presenter: W.S. Hodgkiss
11:45 - 12:05
J.R.Wu, L.Ma
Modeling Reverberation Time Series Based on Full Wave Reverberation model
Presenter: J.R.Wu
12:05 - 12:25
Marcia Isakson, Anthony Bonomo, Nicholas Chotiros
Acoustic Backscattering from Layered Interfaces using Finite Elements
Presenter: Marcia Isakson
22
UA2014
Session 1: Acoustic imaging
Organizer: Jiyuan Liu
Location: Lecture Room C - Tuesday 24th June 2014
Chair: Jiyuan Liu
10:45 - 11:05
Yuan Li,Bing Li, Zhibo Zhang ,Haibo Zheng, Qihu Li
A Design Philosophy of Portable,High-frequence Image Sonar System (invited)
Presenter: Yuan Li
11:05 - 11:25
Zelin Jiang, Wei Liu, Pengfei Zhang, Yuetao Zhou, Zhijie Wang, Yong Huang, Jiyuan Liu,
Chunhua Zhang
Synthetic Aperture Sonar Images Mosaic Based on SIFT and RANSAC Method (invited)
Presenter: Zelin Jiang
11:25 - 11:45
Li Peng, Xu Feng , Zhang Chun
Research on underwater target detection based on seafloor physiognomy-matching of
side scan sonar imagery (invited)
Presenter: Li Peng
11:45 - 12:05
Xudong An, Feng Xu, Juan Yang, Jia Liu, Peng Li
Research on Target Classification for Side-Looking Sonar Based on Acoustic Model
(invited)
Presenter: Xudong An
12:05 - 12:25
Peng Wang,Yong Huang, Jiyuan Liu
Design and Implementation of A Real-time 3-D Imaging Sonar Signal Processing System on
TMS320C6678 (invited)
Presenter: Peng Wang
12:25 - 12:45
Kazuyoshi Mori, Hanako Ogasawara, Toshiaki Nakamura, Takenobu Tsuchiya, Nobuyuki
Endoh
Numerical Analysis for Ambient Noise Imaging with Acoustic Lens: Target Detection
around the Barge Moored in Uchiura Bay (invited)
Presenter: Kazuyoshi Mori
23
UA2014
Session 24b: Three-dimensional sound propagation models
Chair: Michael Porter and Frederic Sturm
10:45 - 11:05
M.Yu. Trofimov, S.B. Kositskiy, A.D. Zakharenko
Mode parabolic equations with mode interaction for the 3D modeling of sound
propagation
Presenter: M.Yu. Trofimov
11:05 - 11:25
Piotr Borejko
A coastal wedge propagation model including shear in an absorptive bottom
Presenter: Piotr Borejko
11:25 - 11:45
Frederic Sturm
Incorporating cross-multiplied terms in a three-dimensional parabolic equation model
Presenter: Frédéric Sturm
11:45 - 12:05
Michael B. Porter
Benchmarking a Three-Dimensional Gaussian Beam Tracing Model
Presenter: Michael B. Porter
Lunch break: 12:45 – 14:15
24
UA2014
Session 31b: Unmanned Vehicles (AUV, USV and Gliders) for Underwater
Chair: Alain Maguer, Brian Ferguson
14:15 - 14:35
Eric Delory, Daniel Toma, Joaquin Del Rio, Pablo Ruiz, Luigi Corradino, Patrice Brault,
Frederic Fiquet
NeXOS objectives in multi-platform underwater passive acoustics (invited)
Presenter: Eric Delory
14:35 - 14:55
Ridha Fezzani, Benoit Zerr, Michel Legris, Ali Mansour, Yann Dupas
The fusion of digital terrain models measured from multiple acoustic sensors –
Application to the DAURADE autonomous underwater vehicle
Presenter: Ridha Fezzani
14:55 - 15:15
Naveed Islam, Ahmed Nait-Chabane, Benoit Zerr, Yann Dupas
Real time improvement of the seabed mapping with AUV-borne sensors using statistical
analysis
Presenter: Naveed Islam
15:15 - 15:35
Mark Baumgartner, David Fratantoni
Real-time acoustic monitoring of baleen whales from autonomous platforms (invited)
Presenter: Mark Baumgartner
15:35 - 15:55
K.G. Kebkal, I. Glushko, T.Tietz, R.Bannasch, O.G.Kebkal, M.Komar, S.G.Yakovlev
SONOBOT - an autonomous unmanned surface vehicle for hydrographic surveys,
hydroacoustic communication and positioning in tasks of underwater acoustic
surveillance and monitoring (invited)
Presenter: K.G. Kebkal
15:55 - 16:15
Willcox S., Leroy F., Wyatt P.
The Persistent Maritime Monitoring System (PMMS) (invited)
Presenter: Leroy F., Willcox S.
25
UA2014
Session 21a: Soundscapes and Measuring Noise
Organizer: Jennifer Miksis-Olds, Mark Prior and Kevin Heaney
Chair: Mark Prior
14:15 - 14:35
Jennifer L. Miksis-Olds, Julia A. Vernon, Kevin Heaney
Applying the dynamic soundscape to estimates of signal detection (invited)
Presenter: Jennifer L. Miksis-Olds
14:35 - 14:55
Thomas Folegot, Dominique Clorennec, Cedric Gervaise, Yann Stephan
Passive calibration of soundscapes (invited)
Presenter: Thomas Folegot
14:55 - 15:15
M.E.G.D. Colin, M.A. Ainslie, A.A.F.M. Beeks, A. M. Von Benda-Beckmann, C. Booth , N.
Bouton, C. Harris, C.A.F. De Jong, H.O. Sertlek, H.W. Slabbekoorn
Ship Noise mapping in the North Sea (invited)
Presenter: M.E.G.D. Colin
15:15 - 15:35
Michael A. Ainslie, Peter H. Dahl, Christ A. F. De Jong, Robert M. Laws
Practical spreading laws: the snakes and ladders of shallow water acoustics (invited)
Presenter: Michael A. Ainslie
15:35 - 15:55
Mark K Prior, David Brown
Signal Grouping by Correlation of Cepstra (invited)
Presenter: Mark Prior
15:55 - 16:15
A.J. Hunter, L. Fillinger, M.C. Clarijs
Feasibility of reef health monitoring using passive acoustics (invited)
Presenter: A.J. Hunter
26
UA2014
Session 29a: Underwater Communication and Networking
Organizer: Charalampos Tsimenidis and Oliver Hinton
Chair: Charalampos Tsimenidis
14:15 - 14:35
Xiao Zhang, Adam Zielinski, Jingwei Yin
High rate uplink acoustic communication from AUVs to surface platform (invited)
Presenter: Adam Zielinski
14:35 - 14:55
K.G. Kebkal, O.G. Kebkal
Clock Synchronization in Underwater Acoustic Networks During Payload Data Exchange
(invited)
14:55 - 15:15
Takuya Shimura, Yukihiro Kida, Mitsuyasu Deguchi, Takami Mori, Yoshitaka Watanabe,
Hiroshi Ochi
Multi-user communication by adaptive time reversal in deep ocean (invited)
Presenter: Takuya Shimura
15:15 - 15:35
Jm Passerieux
Stealth underwater acoustic communications based upon steganography techniques
(invited)
Presenter: Jm Passerieux
15:35 - 15:55
The design of wide band transducers for underwater acoustic communication (invited)
15:55 - 16:15
A. Sanchez, E. Robles, F. J. Rodrigo, F. Ruiz-Vega, U. Fernandez-Plazaola, J. F. Paris
Measurement and Modelling of Fading in Ultrasonic Underwater Channels
Presenter: J. F. Paris
27
UA2014
Session 32a: Vector Sensors: Development and Applications
Organizer: Tuncay Akal, Sergio Jesus and Jean-Pierre Hermand
Chair: Desen Yang
14:15 - 14:35
Hualin Lan, Dajun Sun, Jidan Mei, Tingting Teng
An experimental study on DEMON spectrum direction estimation of multi-target with a
low frequency vector hydrophone (invited)
Presenter: Hualin Lan
14:35 - 14:55
D.S. Yang, Z.R. Zhu
Signal Processing for circular Vector-Sensor Array mounted around a cylindrical baffle
(invited)
Presenter: D.S. Yang
14:55 - 15:15
Z.R. Zhu, D.S. Yang
Broadband DOA estimation in phase modal space for circular acoustic Vector-Sensor
Array (invited)
Presenter: Z.R. Zhu
15:15 - 15:35
Hu Bo, Yang Desen, Sun Yu
Patch near-field acoustical holography based on vector hydrophone array (invited)
Presenter: Hu Bo
15:35 - 15:55
Jie Shi, Desen Yang, Shengguo Shi
A Robust Noise Sources Localization and Identification method based on Vector Sensor
Array (invited)
Presenter: Jie Shi
15:55 - 16:15
Hong Lianjin, Fang Erzheng
Research Vector Hydrophone Measurement System (invited)
Presenter: Hong Lianjin
28
UA2014
Session 31c: Unmanned Vehicles (AUV, USV and Gliders) for Underwater
Chair: Alain Maguer, Brian Ferguson and Eric Delory
16:30 - 16:50
Henrik Schmidt, Erin Fishell
Virtual Ocean Testbed For Autonomous Undersea Sensing Networks (invited)
Presenter: Henrik Schmidt
16:50 - 17:10
Ricardo Martins, Joao Borges De Sousa
Networking underwater, surface and air vehicles: tools and experimentation (invited)
Presenter: Ricardo Martins
17:10 - 17:30
Kevin D. Lepage, Ryan Goldhahn
CMRE's use of AUVs and USVs for the demonstration of network concepts for multistatic
active ASW (invited)
Presenter: Kevin Lepage
Session 21b: Soundscapes and Measuring Noise
Chair: Kevin Heaney
16:30 - 16:50
H.Ozkan Sertlek, Geert Aarts, Sophie Brasseur, Hans Slabbekoorn, Carel Ten Cate, Michael A.
Ainslie
Sound Maps Of The Dutch North Sea For Natural And Anthropogenic Sound Sources
(invited)
16:50 - 17:10
L. S. Wang, S. P. Robinson, P.M. Harris, D. Partridge, J. F. Borsani, K. Brookes.
Analysis of soundscapes in the East coast waters of the UK (invited)
Presenter: L. S. Wang
17:10 – 17:40
Open Discussion
Chair: Jennifer Miksis-Olds
29
UA2014
Session 29b: Underwater Communication and Networking
Chair: Charalampos Tsimenidis
16:30 - 16:50
Siyu Xing, Gang Qiao, Wei Wang, Songzuo Liu, Zhimeng Zhu
Joint Pilot based Channel Estimation with Selected-Mapping to Reduce PAPR in
Underwater Acoustic MIMO-OFDM System without Side Information
Presenter: Siyu Xing
16:50 - 17:10
Tran Minh Hai, Yasuto Matsuda, Taisaku Suzuki, Tomohisa Wada
Ultrasonic Diversity OFDM Transceiver architecture with Impulsive Noise Cancelling for
shallow sea communication
Presenter: Tran Minh Hai
17:10 - 17:30
S.N. Qader, C.C. Tsimenidis, M. Johnston, B.S. Sharif
PIC-DDFE-IDMA Detection For Uplink Shallow Water Acoustic Channels
Presenter: C.C. Tsimenidis
Session 32b: Vector sensors: development and applications
Chair: Desen Yang
16:30 - 16:50
Fenghua Li, Liangming Zhu, Renhe Zhang
Measurement of the vector field and its application on Geo-acoustic inversion (invited)
Presenter: Fenghua Li
16:50 - 17:10
Qunyan Ren, Jean-Pierre Hermand
characterization (invited)
Presenter: Qunyan Ren
30
UA2014
Wednesday 25th June 2014
Plenary Lecture
Chair: Peter Gerstoft
Prof. Ross Chapman, University of Victoria, Canada
Reflections on ocean bottom reflections
Location: Lecture Room A - Wednesday 25th June 2014 - 09:30-10:30
Session 20a: Sonar Signal and Information Processing
Organizer: Ryan Goldhahn, Duncan Williams, Julian Deeks and Peter Gerstoft
Location: Lecture Room A - Wednesday 25th June 2014
Chair: Ryan Goldhahn, Julian Deeks and Peter Gerstoft
10:45 - 11:05
Arthur B. Baggeroer, Henrik Schmidt
on
Performance as Implemented with OASES (invited)
Presenter: Arthur Baggeroer
11:05 - 11:25
Kam W. Lo, Brian G. Ferguson
Source motion parameter estimation using direct and multipath arrivals at a pair of
hydrophones (invited)
11:25 - 11:45
Peter Gerstoft, Caglar Yardim, Angeliki Xenaki, Christoph Mecklenbrauker
Compressive sensing in acoustics and seismology (invited)
11:45 - 12:05
Angeliki Xenaki, Peter Gerstoft, Klaus Mosegaard
Sparsity and super-resolution in sound source localization with sensor arrays (invited)
Presenter: Angeliki Xenaki
12:05 - 12:25
Gleb Sidelnikov
DoA Estimation Algorithm as Applied to Wideband Processing (invited)
Presenter: Gleb Sidelnikov
12:25 - 12:45
Kay L. Gemba, Eva-Marie Nosal
Estimating source spectra from recordings made in a reverberant underwater channel
(invited)
Presenter: Kay L. Gemba
Program and Book of Abstracts 31
UA2014
Session 18: Sensitivity of underwater acoustic observables
Organizer: Emmanuel Skarsoulis
Location: Lecture Room B - Wednesday 25th June 2014
Chair: Emmanuel Skarsoulis
10:45 - 11:05
Stan E. Dosso, Michael J. Wilmut
Multiple-Source Localization and Environmental Sensitivity (invited)
Presenter: Stan E. Dosso
11:05 - 11:25
Florian Aulanier, Hacheme Ayasso, Philippe Roux, Barbara Nicolas, Jerome Mars.
Sound-speed tomography using angle sensitivity-kernels in an ultrasonic waveguide
(invited)
Presenter: Barbara Nicolas
11:25 - 11:45
E.K. Skarsoulis, B.D. Cornuelle, M.A. Dzieciuch
Travel-time sensitivity kernels in a shallow water environment (invited)
Presenter: E.K. Skarsoulis
11:45 - 12:05
Michael Taroudakis, Costas Smaragdakis
A hybrid approach for ocean acoustic tomography based on statistical characterization of
the acoustic signal and the identification of modal arrivals. (invited)
Presenter: Michael Taroudakis
12:05 - 12:25
Xavier Cristol
Non-Perturbative Evaluations of Time Sensibility Kernels using Alternative Definitions of
Propagation Delay
32
UA2014
Session 9: Calibration of Sonar and Hydrophones
Organizer: Bo Lövgren and Stefan Schael
Location: Lecture Room C - Wednesday 25th June 2014
Chair: Bo Lövgren and Stefan Schael
10:45 - 11:05
Bo Lovgren
Low Frequency Tank Calibration by comparison (invited)
Presenter: Bo Lövgren
11:05 - 11:25
Stefan Schael
Adulteration of underwater acoustic measurements (invited)
Presenter: Stefan Schael
11:25 - 11:45
Krystian Buszman, Ignacy Gloza, Rafal Jozwiak, Karol Listewnik
The calibration of hydroacoustic channel of mobile measurement module (invited)
Presenter: Krystian Buszman
11:45 - 12:05
Tom Dakin
Calibrating hydrophones at very low frequencies
Presenter: Tom Dakin
12:05 - 12:25
Xiuting Yang, Min Li, Yanhong Hu
Array Shape Estimation Using Measurements of Heading and Depth Sensors
Presenter: Xiuting Yang, Min Li, Yanhong Hu
33
UA2014
Session 7a: Acoustics of marine renewable energy developments
Organizer: Stephen Robinson, Paul Lepper and Philippe Blondel
Location: Lecture Room D - Wednesday 25th June 2014
Chair: Stephen Robinson, Paul Lepper and Philippe Blondel
10:45 - 11:05
Michael Butler, Jenny Norris, Paul Lepper
Designing practical on-site calibration protocols for acoustic systems; Key elements and
pitfalls (invited)
Presenter: Michael Butler
11:05 - 11:25
Peter Dobbins, Federica Pace, Irene Vollmy, Silvana Neves, Sophie Nedelec
An environmental survey around the Narec Offshore Anemometry Hub (NOAH) – a
comparison between acoustic measurement instruments. (invited)
Presenter: Peter Dobbins
11:25 - 11:45
Paul A. Lepper, Stephen P. Robinson
Methods for measurement of long term radiated noise from a wave energy system
(invited)
Presenter: Paul Lepper
11:45 - 12:05
Cristiano Soares, Erica Cruz, Friedrich Zabel, Andre Moura
Environmental inversion with an autonomous hydrophone in a wave energy device
deployment site (invited)
Presenter: Cristiano Soares
12:05 - 12:25
Benjamin Williamson, Philippe Blondel, James Waggitt, Paul Bell, Beth Scott
Field deployments of a self-contained subsea platform for acoustic monitoring of the
environment around marine renewable energy structures (invited)
Presenter: Philippe Blondel
12:25 - 12:45
Melanie E. Austin
Underwater Sounds from Drillships and Support Vessels During Exploratory Drilling
Offshore Alaska
Lunch break: 12:45 – 14:15
34
UA2014
Session 20b: Sonar Signal and Information Processing
14:15 - 14:35
Bruce Newhall, Anna Slowikowski
Multivariate Distributions of Clutter Levels for Automated Classifiers (invited)
Presenter: Bruce Newhall
14:35 - 14:55
N.-E. Lasmar, A. Baussard, G. Le Chenadec
Seafloor classification using statistical modeling of wavelet subbands (invited)
Presenter: N.-E. Lasmar
14:55 - 15:15
Stefan M. Murphy, Paul C. Hines, Kevin Dunphy
Classifying continuous active sonar echoes for target recognition (invited)
Presenter: Stefan M. Murphy
15:15 - 15:35
Ivan Aleksi, Dieter Kraus, Zeljko Hocenski
Signal Change Detection Method Used for Mine-Like Objects Segmentation in Sonar
Images (invited)
Presenter: Ivan Aleksi
15:35 - 15:55
Teng Tingting, Sun Dajun, Lan Hualin, Mei Jidan
Underwater moving targets detection using the sequence of acoustic image (invited)
Presenter: Teng Tingting
15:55 - 16:15
L. Fillinger, A.J. Hunter, M.C. Clarijs
Passive sonar denoising for diver detection in presence of snapping shrimp (invited)
Presenter: L. Fillinger
35
UA2014
Session 17a: Radiated Noise from Ships and Production Platforms
Organizer: Grazyna Grelowska
Chair: Grazyna Grelowska
14:15 - 14:35
G. Grelowska, E. Kozaczka, S. Kozaczka, W. Szymczak
General characteristics of shipping underwater noise (invited)
Presenter: G. Grelowska
14:35 - 14:55
Federico Traverso, Andrea Trucco
Analysis of underwater acoustic noise measured at the ship bow during sea trials (invited)
Presenter: Federico Traverso
14:55 - 15:15
E. Kozaczka, G. Grelowska, W. Szymczak, S. Kozaczka
Sound generation by the ship propeller in the subcavitation range (invited)
Presenter: W. Szymczak
15:15 - 15:35
Jesse Spence, Andrew Patterson, Ray Fischer
Measurement System to Assess Underwater Noise from Vessels and Marine Activities
Presenter: Jesse Spence
15:35 - 15:55
Andre Moura, Erica Cruz, Diogo Lopes, Raul Hospital-Bravo, Josep Sarrate, Pedro Diez
Integrated tool for the acoustic assessment and monitoring of marine activities and
operations
Presenter: André Moura, Raúl Hospital-Bravo
15:55 - 16:15
L. Fillinger, A. Mantouka, C. De Jong, I. Gloza, A. Sanchez, E. Moya, S. Schael, T. Lennartsson,
G. Petit, R. Fardal, H. Hasenpflug, A.L.D. Beckers
SIRAMIS : Preliminary Analysis of Acoustic and Seismic Ship Signatures (invited)
36
UA2014
Session 12a: Experimental and modelling validation of target strength
measurements
Organizer: Duncan Williams, David Nunn and Alan Hunter
Chair: David Nunn and Alan Hunter
14:15 - 14:35
David Prowse
representation (invited)
Presenter: David Prowse
14:35 - 14:55
Sandrine Rakotonarivo, Selda Yildiz, Philippe Roux, Earl Williams, W. A. Kuperman
Model Tank Measurements and Using a Random Noise Feld to Determine the Scattering
Properties of an Object (invited)
Presenter: W. A. Kuperman
14:55 - 15:15
Henry Dol
On the design and construction of drifting-mine test targets for sonar, radar and electrooptical detection experiments (invited)
Presenter: Henry Dol
15:15 - 15:35
Keith Page, Charles Taylor, Ben Hodder
Open water target strength measurements and validation (invited)
Presenter: Keith Page, Charles Taylor, Ben Hodder
15:35 – 16:15
Open Discussion
37
UA2014
Session 7b: Acoustics of marine renewable energy developments
14:15 - 14:35
Bas Binnerts,Pieter Van Beek,Sander Von Benda-Beckmann, Erik Nennie
Underwater sound due to a subsea high speed turbo-compressor
Presenter: Bas Binnerts
14:35 - 14:55
John T. Dardis, Per G. Reinhall
New methods in impact pile driving noise attenuation
Presenter: John T. Dardis Ii
14:55 - 15:15
Katja Reimann, Jurgen Grabe
Soil vibration due to offshore pile driving and induced underwater noise (invited)
Presenter: Katja Reimann
15:15 - 15:35
Michael Wood, Victor Humphrey
Model results for offshore piling acoustics featuring an elastic sediment with a depthdependent shear wave speed (invited)
Presenter: Michael Wood
15:35 - 15:55
Hazelwood R, Macey P
The evanescent pressure waves associated with ground roll waves from seabed impacts
(invited)
Presenter: Dick Hazelwood
15:55 - 16:15
Karl-Heinz Elmer
Effective Reduction of Offshore Piling Noise (invited)
Presenter: Karl-Heinz Elmer
38
UA2014
Session 20c: Sonar Signal and Information Processing
16:30 - 16:50
Doug Grimmett
Target AOU Growth Containment using High Duty Cycle Sonar (invited)
Presenter: Doug Grimmett
16:50 - 17:10
Doug Grimmett
Target Doppler Estimation and Range Bias Compensation using LFM High Duty Cycle
Sonar (invited)
17:10 - 17:30
Alexander V. Burenin, Vladimir V. Bezotvetnykh, Michail V. Lebedev, Evgeny A. Voytenko,
Dmitry S. Strobykin
A Doppler estimation technique is based on the signals with good correlation properties:
Experimental results (invited)
Presenter: Burenin Alexandr Victorovich
17:30 - 17:50
Frank Ehlers
Ontology Design for Cooperative Underwater Target Tracking (invited)
Presenter: Frank Ehlers
39
UA2014
Session 17b: Radiated Noise from Ships and Production Platforms
Chair: Grazyna Grelowska
16:30 - 16:50
Jong-Hoon Jeon, Won-Ho Joo
Prediction of propeller radiated noise by onboard measurement
Presenter: Jong-Hoon Jeon
16:50 - 17:10
S. Coward, H Dong, D Tollefsen
Source level estimates of small cargo ships from measurements in a fjord
Presenter: Dag Tollefsen
17:10 - 17:30
J. Bielanski
Pressure signals of low frequency generated by the ship's hull. (invited)
Presenter: J. Bielanski
40
UA2014
Session 12b: Experimental and modelling validation of target strength
measurements
Chair: David Nunn and Alan Hunter
16:30 - 16:50
Ahmad T. Abawi
Finite element modeling of acoustic radiation force for elastic objects (invited)
16:50 - 17:10
Marten .J.J. Nijhof, Aubrey L. Espana, Kevin L. Williams
Improved modeling accuracy of the Elastic Object Response by Inclusion of Higher Order
Re-Scattering (invited)
Presenter: Marten Nijhof
17:10 - 17:30
Aubrey L. Espana, Marten J. J. Nijhof, Kevin L. Williams, Daniel S. Plotnick, Philip L. Marston
Acoustic scattering from partially buried cylinders: Measurement validation and
interpretation using physical acoustics and finite element models (invited)
Presenter: Aubrey L. Espana
17:30 - 17:50
A.J. Hunter, I. Mulders, M.J.J. Nijhof, B.A.J. Quesson, A.L.D. Beckers
Performance Modelling and Experimental Validation for a Sediment-Penetrating SAS
(invited)
17:50 - 18:10
Laurent Fillinger, Marten J.J. Nijhof, Christ A.F. De Jong
An efficient numerical target strength prediction model: validation against analytic
solutions (invited)
Presenter: Laurent Fillinger, Marten J.J. Nijhof
18:10 – 18:30
Open Discussion
41
UA2014
Session 7c: Acoustics of marine renewable energy developments
16:30 - 16:50
Erica Cruz, Ines Machado, Teresa Simas
Underwater noise assessment of wave energy devices
Presenter: Erica Cruz
16:50 - 17:10
Joanne K. Garrett, Matthew J. Witt
Underwater sound levels at a wave energy device testing facility in Falmouth Bay, UK
(invited)
Presenter: Joanne Garrett
17:10 - 17:30
Karen Bemis, Guangyu Xu, Darrell Jackson, Peter Rona, Russ Light
Cabled observatory enabled acoustic monitoring of hydrothermal discharge
Presenter: Karen Bemis
42
UA2014
Thursday 26th June 2014
Session 11: Distributed Networked Systems for Surveillance
Organizer: Frank Ehlers and Arne Schulz
Location: Lecture Room A - Thursday 26th June 2014
Chair: Frank Ehlers
8:30 - 8:50
Kevin Lepage, Juri Sildam
Problems in globally optimizing underwater surveillance networks in communications
limited environments (invited)
8:50 - 9:10
Martin Michaelis, Martina Brotje, Frank Ehlers
Parameter estimation for non-cooperative multistatic sonar (invited)
Presenter: Martin Michaelis
9:10 - 9:30
Mark R. Balthasar, Sara Al-Sayed, Stefan Leier, Abdelhak M. Zoubir
Optimal Area Coverage in Autonomous Sensor Networks (invited)
Presenter: Mark R. Balthasar
9:30 - 9:50
Ryan Goldhahn, Paolo Braca, Kevin Lepage
Adaptive Bayesian behaviors for AUV surveillance networks (invited)
Presenter: Ryan Goldhahn
9:50 - 10:10
Andrea Caiti, Francesco Di Corato, Davide Fenucci, Benedetto Allotta, Fabio Bartolini,
Riccardo Costanzi, Jonathan Gelli, Niccolo Monni, Marco Natalini, Luca Pugi, Alessandro
Ridolfi
Acoustic communication and localization in AUV cooperative surveys (invited)
Presenter: Andrea Caiti
43
UA2014
Session 23: Tank Experiments
Organizer: Jean-Pierre Sessarego and Dominique Fattaccioli
Location: Lecture Room B - Thursday 26th June 2014
Chair: Dominique Fattaccioli
8:30 - 8:50
A. Guillou, J-P Sessarego, R. Guillermin, D. Fattaccioli
Source level measurement in deep water conditions: Are free field condition met
whatever the source frequency?
Presenter: J-P. Sessarego
8:50 - 9:10
Panagiotis Papadakis, George Piperakis, Spyros Kouzoupis
Calibration of ultrasound transducer heads using short preprocessed ultrasonic pulses
(invited)
Presenter: Panagiotis Papadakis
9:10 - 9:30
G. Real, J.-P. Sessarego, X. Cristol, D. Fattaccioli
De-Coherence Effects in Underwater Acoustics: Scaled Experiments. (invited)
Presenter: Gaultier Real
9:30 - 9:50
Shan Victor Pereira, D D Ebenezer, S K Bhattacharyya
Acoustic Echo Reduction and Insertion Loss of Tiles
Presenter: Shan Victor Pereira
9:50 - 10:10
J.-P. Sessarego, R. Guillermin, A. Jarnac, A. Houard, Y. Brelet, J. Carbonnel, Y.B. Andre, A.
Mysyrowicz, D. Fattaccioli
A high intensity pulsed laser as a wide band acoustic source for underwater acoustic
applications (invited)
44
UA2014
Session 5a: Acoustics in Polar Environments
Organizer: Jaroslaw Tegowski and Alexander Gavrilov
Location: Lecture Room C - Thursday 26th June 2014
Chair: Jaroslaw Tegowski and Alexander Gavrilov
8:30 - 8:50
Lee Freitag, Peter Koski, Hanne Sagen, Gaute Hope, Kevin Heaney
Acoustic Communications Experiments in the Fram Strait 2013 (invited)
Presenter: Lee Freitag
8:50 - 9:10
Grant B. Deane, Jaroslaw Tegowski, Oskar Glowacki Mateusz Moskalik, Philippe Blondel.
Measurements of the Ambient Noise Field in an Arctic, Glacial Fjord (invited)
Presenter: Jaroslaw Tegowski
9:10 - 9:30
Gaute Hope, Hanne Sagen, Dag Tollefsen, Hans-Christian Tengesdal
Acoustic propagation in the Marginal Ice Zone and the implications for navigation of
underwater vehicles
Presenter: Gaute Hope
9:30 - 9:50
Lukasz Hoppe, Joanna Szczucka, Emilia Trudnowska
Zooplankton distribution studies combining acoustical and optical observations
Presenter: Lukasz Hoppe
9:50 - 10:10
Stan E. Dosso
Three-dimensional Source Localization using an Ice-mounted Geophone
10:10 - 10:30
Jaroslaw Tegowski, Grant Deane, Philippe Blondel, Oskar Glowacki, Mateusz Moskalik
An acoustical study of gas bubbles escaping from melting growlers
45
UA2014
Session 25a: Towards Automatic Target Recognition: Detection, Classification
and Modeling (of Underwater Targets)
Organizer: John Fawcett, Johannes Groen, Wolfgang Jans and Yan Pailhas
Location: Lecture Room D - Thursday 26th June 2014
Chair: Wolfgang Jans and Johannes Groen
8:30 - 8:50
Mohamad Nakcha, Einas Alhaji, Mohamd Alcahaita
Iterative Target Recognition for Port Protection System
Presenter: Mohamad Nakcha, Einas Alhaji
8:50 - 9:10
Andrei Mashoshin
The Main Peculiarities of Automatic Target Recognition (invited)
Presenter: Andrei Mashoshin
9:10 - 9:30
Yan Pailhas, Yvan Petillot
Independent views in MIMO sonar systems (invited)
Presenter: Yan Pailhas
9:30 - 9:50
Karl Thomas Hjelmervik
Automatic classification for mid-frequency anti-submarine warfare sonars – recognizing
pipelines
Presenter: Karl Thomas Hjelmervik
9:50 - 10:10
Imen Karoui, Isabelle Quidu, Michel Legris
Acoustic obstacle detection for safe AUV surfacing (invited)
Presenter: Isabelle Quidu
10:10 - 10:30
Yan Pailhas, Emmanuel Delande, Jeremie Houssineau, Yvan Petillot, Daniel Clark
Tracking underwater objects using large MIMO sonar systems (invited)
46
UA2014
Session 2a: Acoustic Monitoring of Marine Mammals
Organizer: Purnima Ratilal and Adam Zielinski
Chair: Adam Zielinski
10:45 - 11:05
A. Gavrilov, A. Ghiotto, C. Jenner, M. Jenner , C. Gruenthal
Passive acoustic monitoring of humpback whales in Exmouth Gulf using a sparse array of
DIFAR sensors (invited)
Presenter: A. Gavrilov
11:05 - 11:25
Roberto Racca, Julien Delarue, Heloise Frouin-Mouy, David Hannay, Bruce Martin, Xavier
Mouy, Jennifer Wladichuk
Acoustic monitoring of marine mammals in the Chukchi Sea – Three case studies
Presenter: Roberto Racca
11:25 - 11:45
R. Bahl, H. Sugimatsu, T. Ura, J. Kojima, S. Behera
Passive acoustic monitoring and bio-sonar characterization of Ganges river dolphin,
India's national aquatic animal (invited)
Presenter: R. Bahl
11:45 - 12:05
Eric L. Ferguson
Passive acoustic source localisation methods for the nonintrusive monitoring of
echolocating dolphins in the wild (invited)
Presenter: Eric L. Ferguson
12:05 - 12:25
J. Bonnel, A. Thode
Bowhead whale calls localization using a single receiver and warping processing (invited)
Presenter: J. Bonnel
12:25 - 12:45
Liang Zhang, Adam Zielinski, Jidan Mei, Ping Cai
Passive Acoustic Monitoring of Marine Mammals using a Ternary Array (invited)
47
UA2014
Session 33: Bioacoustics
Chair: Orest Diachok
10:45 - 11:05
Orest Diachok
Bioacoustic Absorption Spectroscopy of physoclists (invited)
Presenter: Orest Diachok
11:05 - 11:25
Vicent Puig, Victor Espinosa, Ester Soliveres, Pedro Poveda, Fernando De La Gandara, Jaime
Ramis, Patricia Ordonez, Isabel Perez-Arjona, Jose L. Cort
Bluefin tuna behavioural response to anthropogenic noise in floating sea cages
Presenter: Victor Espinosa, Vicent Puig
11:25 - 11:45
Ester Soliveres, Vicent Puig, Patricia Ordónez, Isabel Pérez-Arjona, Miguel Ardid, Jaime
Ramis , Pedro Poveda, Vicent D. Estruch, Dolores Lópezc, del Mar Agrasod, Rocio Roblesd,
Víctor Espinosa
Acoustical biomass estimation results in Mediterranean aquaculture sea cages
Presenter: Victor Espinosa
11:45 - 12:05
Jens M. Hovem
Sound propagation over an elastic bottom –particle motions caused by seismic interface
waves
Presenter: Jens M. Hovem
12:05 - 12:25
Karthikeyan, Abhay
Acoustics marine survey of distribution and intensity of species in Arabian Sea during
monsoon variation (by mid-water trawl survey)
Presenter: Karthikeyan Muniraj
48
UA2014
Session 5b: Acoustics in Polar Environments
Chair: Jaroslaw Tegowski and Alexander Gavrilov
10:45 - 11:05
Hanne Sagen, Dag Tollefsen, Hans C. Tengesdal
The soundscape of the Fram Strait Marginal Ice Zone (invited)
11:05 - 11:25
Andrea Bordone, Roberto Bozzano, Sara Pensieri, Paola Picco, Elisabetta Schiano, Federico
Traverso, Andrea Trucco
Under sea-ice acoustic noise and propagation measurements in Tethys Bay (Ross Sea,
Antarctica)
11:25 - 11:45
R. P. Dziak, H. Matsumoto, D. R. Bohnenstiehl, M. Park, W S Lee, H. Klinck, M.J. Fowler, T-K
Lau, J.H. Haxel, D. K. Mellinger, K. M. Stafford
Sources of long-term ambient ocean sound near the Antarctic Peninsula (invited)
Presenter: R.P. Dziak
49
UA2014
Session 25b: Towards Automatic Target Recognition: Detection, Classification
and Modeling (of Underwater Targets)
Chair: Johannes Groen and Yan Pailhas
10:45 - 11:05
Richard Brothers
An information-based Johnson’s criteria model for UUV system performance prediction
(invited)
Presenter: Richard Brothers
11:05 - 11:25
Marc Geilhufe, Oivind Midtgaard
Quantifying the complexity in sonar images for MCM performance estimation (invited)
Presenter: Marc Geilhufe
11:25 - 11:45
A.J. Hunter, R. Van Vossen, A.L.D. Beckers
Towards Automatic Target Recognition in Low-Frequency Sub-Sediment Sonar Imagery
(invited)
11:45 - 12:05
Jo Inge Buskenes, Jon Petter Asen, Herman Midelfart, Oyvind Midtgard
A GPU Sonar Simulator for Automatic Target Recognition
Presenter: Jo Inge Buskenes
12:05 - 12:25
Daniel Kohntopp, Benjamin Lehmann, Dieter Kraus
Efficient Superellipse Fitting based Contour Extraction for mine-like shape Recognition
(invited)
Presenter: Daniel Köhntopp
12:25 - 12:45
Yan Pailhas, Keith Brown, Chris Capus, Nicolas Valeyrie
Identifying Contents of Low Profile Targets in a Cluttered Environment (invited)
Lunch break: 12:45 – 14:15
50
UA2014
Session 2b: Acoustic Monitoring of Marine Mammals
Chair: Purnima Ratilal
14:15 - 14:35
Naysa E. Balcazar, Holger Klinck , Sharon L. Nieukirk, David K. Mellinger , Joy S. Tripovich,
Robert P. Dziak , Tracey L. Rogers
Acoustics as a tool to reveal population structure of the elusive blue whale (invited)
Presenter: Naysa Balcazar
14:35 - 14:55
Tina M. Yack, Alyson Fleming, Jay Barlow, Jessica Redfern1, Elizabeth Becker, Peter Klimley,
Marcel Holyoak
The use of passive acoustic data to predict beaked whale habitat in the California Current
Ecosystem
Presenter: Tina M. Yack
14:55 - 15:15
Dong Hoon Yi, Peter Tyack, Nicholas Makris
Remote Sensing of Large Herring Shoals by Baleen Whales (invited)
Presenter: Dong Hoon Yi
15:15 - 15:35
Carolyn M. Binder, Paul C. Hines
Modelling the impact of ocean environment on automatic aural classification of marine
mammals.
Presenter: Carolyn M. Binder
15:35 - 15:55
Wei Huang, Fan Wu, Nicholas Makris, Purnima Ratilal
Classifying Humpback Whale Individuals from their Nocturnal Feeding-Related
Vocalizations
Presenter: Purnima Ratilal
51
UA2014
Session 13a: Habitat Mapping: Procedures and Results
Organizer: Philippe Blondel and Andrea Caiti
Chair: Philippe Blondel
14:15 - 14:35
Fakiris Elias, Zoura Despina, Ferentinos George, Papatheodorou George
Towards Joint Use Of Side Scan Sonar And Sub-Bottom Profiler Data For The Automatic
Quantification Of Marine Habitats. Case Study: Lourdas Gulf, Kefalonia Isl., Greece.
(invited)
Presenter: Fakiris Elias
14:35 - 14:55
Diego Porpilho, Antonio Henrique Da Fontoura Klein, Rafael S. V. De Camargo, Michel
Franco Volpato Prado, Jarbas Bonetti, Andrew Short, Elias Fakiris.
Automatic classification of bedforms using phase differencing bathymetric sonar
Presenter: Diego Porpilho
14:55 - 15:15
Natalia Gorska, Ewa Kowalska-Duda, Filip Pniewski, Adam Latala, Jacek Marszal, Jan
Schmidt
Diurnal variation in sediment backscattering properties caused by photosynthesis of
microphytobenthos (Southern Baltic Sea) (invited)
Presenter: Ewa Kowalska-Duda
15:15 - 15:35
Malgorzata Godlewska, Katarzyna Izydorczyk, Shaowen Ye, Bronislaw Dlugoszewski, Marta
Cendrowska
Do fish and blue algae blooms coexist in space? (invited)
Presenter: Malgorzata Godlewska
15:35 - 15:55
Marta Cendrowska , Wojciech Puchalski, Mikolaj Adamczyk, Malgorzata Godlewska
Acoustic mapping of submerged macrophytes in selected lakes of the Drawieński National
Park (invited)
15:55 - 16:15
F. Foglini, F. Madricardo, P. Blondel, A. Kruss, M. Sigovini
High-resolution multibeam mapping of habitats in the extremely shallow waters of the
Venice Lagoon (invited)
Presenter: Federica Foglini
52
UA2014
Session 27a: Underwater Acoustic Studies in Asian Seas
Organizer: Chifang Chen
Chair: Chifang Chen
14:15 - 14:35
Yiing-Jang Yang, Wen-Der Liang, Jeff Chih-Hao Wu, Hsian-Chih Chan, Ruey-Chang Wei, ChiFang Chen
Observation of ambient noise induced by the internal solitary wave in the center of
Kuroshio northeast of Taiwan (invited)
Presenter: Yiing-Jang Yang
14:35 - 14:55
J. Zeng, W.Y. Zhao, D.Y. Peng, H.F. Li , Y.Ge, T.F. Gao , E.C. Shang
Some characteristics of bottom scattering provided by single-mode reverberation (invited)
Presenter: J. Zeng
14:55 - 15:15
Linus Y.S Chiu, Andrea Chang, Chifang Chen
Measured Channel Impulse Responses for a Mobile Source in the Northeastern Sea off
Taiwan (invited)
Presenter: Linus Y.S Chiu
15:15 - 15:35
Yin-Ying Fang, Shih-En Chou, Chifang Chen, Chien-Kang Huang
Observe Seismic Activities and Ambient Noise of Underwater Acoustic Data from MACHO
Hydrophone
Presenter: Yin-Ying Fang
15:35 - 15:55
Jeff Chih-Hao Wu, Chi-Fang Chen, Ruey-Chang We
An Overview of Ocean Ambient Noise around Taiwan: Measurement and Analysis
Presenter: Jeff Chih-Hao Wu
53
UA2014
Session 34a: Underwater Communication
Chair: Victor Humphrey
14:15 - 14:35
Songzuo Liu, Gang Qiao,Zongxin Sun, Wei Sun, Lu Ma
Data Smoothing algorithms for Phased Array Doppler Log
Presenter: Songzuo Liu
14:35 - 14:55
Karsten Wiedmann, Tobias Weber
Comparison of Modulation Techniques for Parametric Underwater Communications
Presenter: Karsten Wiedmann
14:55 - 15:15
Xiao Dong, Wei Liping, Chen Yan, Chen Geng, Ma Li
Inquiring Flooding Algorithm For Underwater Acoustic Sensor Self-Organization Network
Presenter: Xiao Dong
15:15 - 15:35
Bo Peng, Hefeng Dong
DSP implementation of Turbo Equalization based Underwater Acoustic Modem
Presenter: Bo Peng
15:35 - 15:55
Feng Zhou, Gang Qiao, Yanling Yin, Lu Ma, Songzuo Liu, Zongxin Sun
Orthogonal Multicarrier Underwater Acoustic Communication Experiments in River, Lake
and Shallow Sea
Presenter: Feng Zhou
15:55 - 16:15
Jidan Mei, Dajun Sun, Yunfei Lv, Junjie Shi, Tingting Teng, Hualin Lan
Study on the impact of multi-path channel to the horizontal uniform linear array
beamforming
Presenter: Jidan Mei
54
UA2014
Session 2c: Acoustic Monitoring of Marine Mammals
Chair: Roberto Racca
16:30 - 16:50
Odile Gerard
Classification of Beaked Whale Signal Recorded in Atlantic
Presenter: Odile Gerard
16:50 - 17:10
Zhang Chun, Xu Feng, Zhang Qiao
Acoustic broadband backscattering and classification of fish
Presenter: Zhang Chun
Session 13b: Habitat Mapping: Procedures and Results
Chair: Andrea Caiti
16:30 - 16:50
Garret Duffy, Yuri Rzhanov, Eli Rinde, Berit Horvei, Colin Brown
Development of Acoustic Colour technique using multi-frequency swath acoustic
backscatter (invited)
Presenter: To Be Decided
16:50 - 17:10
B. Allotta, F. Bartolini, R. Conti, R. Costanzi, J. Gelli, N. Monni, M. Natalini, L. Pugi, A. Ridolfi
MARTA: an AUV for Underwater Cultural Heritage (invited)
Presenter: Benedetto Allotta
17:10 - 17:30
F. Madricardo, F. Foglini, A. Kruss
High-resolution multibeam bathymetry applied to underwater research: a case study
from the Lagoon of Venice (invited)
Presenter: F. Madricardo
55
UA2014
Session 27b: Underwater Acoustic Studies in Asian Seas
Chair: Chifang Chen
16:30 - 16:50
Andrey Serebryany, Andrey Belogortsev, Cho-Teng Liu
Large-amplitude internal solitary wave of the second mode in Luzon Strait: observations
and numeric modeling of its propagation (invited)
16:50 - 17:10
J.R.Wu T.F.Gao E.C.Shang
Reverberation Modeling in Range-dependent Waveguide (invited)
Presenter: J.R.Wu
Session 34b: Underwater Communication
Chair: Victor Humphrey
16:30 - 16:50
Yoshinori Miyamoto,Kazushi Miyashita, Nobuaki Arai, Takashi Kitagawa, Keiichi Uchida,
Hiromichi Mitamura, Toyoki Sasakura
Development/Introduction of the bio-logging system to realize high data recovery rate
using acoustic communication
Presenter: Yoshinori Miyamoto
16:50 - 17:10
Gang Qiao, Jiarong Zhang
Full-duplex, relative clock based and collision free protocol for underwater acoustic
networks
Presenter: Gang Qiao, Jiarong Zhang
56
UA2014
Friday 27th June 2014
Session 15: Modeling Sonar Performance in Uncertain Environments
Organizer: Georgios Haralabus and Chris Strode
Location: Lecture Room A - Friday 27th June 2014
Chair: Georgios Haralabus and Chris Strode
8:30 - 8:50
Mark Gammon, Marcel Lefrancois, Chris Strode
Mitigation Methods and Techniques for Enhancing Sonar Operational Confidence (invited)
Presenter: Chris Strode
8:50 - 9:10
Andrew Holden
Modelling multistatic sonobuoy fields in uncertain environments (invited)
Presenter: Andrew Holden
9:10 - 9:30
G. Real, X. Cristol , J.-P. Sessarego, D. Fattaccioli
Propagation of Acoustic Waves through a Spatially Fluctuating Medium: Theoretical Study
of the Physical Phenomena. (invited)
9:30 - 9:50
V. Premus, P. Abbot, M. Helfrick, T. Paluszkiewicz
Passive Sonar Performance Characterization and Transmission Loss Measurement Using a
Calibrated Mobile Acoustic Source (invited)
Presenter: V. Premus
9:50 - 10:10
Yong-Min Jiang, Christopher Strode
Measurement and Model Forecast Comparison of Acoustic Signal-Excess Fluctuations
(invited)
Presenter: Yong-Min Jiang, Christopher Strode
10:10 - 10:30
Kevin D. Lepage
Adding error bounds to energy flux estimates of trasmission loss (invited)
Presenter: Kevin D. Lepage
57
UA2014
AB Wood Medal Lecture
Location: Lecture Room B - Friday 27th June 2014
Chair: Peter Dobbins
8:30 – 9:00
Brian T. Hefner
From the pole to the equator: Utilizing a screw dislocation in an acoustic wavefront.
(invited)
Session 22: Synthetic Aperture Sonar: State-of-the-art
Organizer: Roy-Edgar Hansen and Daniel Brown
Location: Lecture Room C - Friday 27th June 2014
Chair: Roy-Edgar Hansen
8:30 - 8:50
Roy E Hansen, Torstein O Saeboe, Ole J Lorentzen, Oivind Midtgaard
Change detection in topographic structures using interferometric synthetic aperture
sonar (invited)
Presenter: Roy E Hansen
8:50 - 9:10
Johannes Groen, Holger Schmaljohann, Wolfgang Jans, Ursula Holscher-Hobing
Comparison of Fusion Approaches for the Displace Phase Centre Antenna Method
(invited)
Presenter: Johannes Groen
9:10 - 9:30
Ann E. A. Blomberg, Andreas Austeng, Roy Edgar Hansen
Alternative SAS processing for gas seep detection (invited)
Presenter: Ann E. A. Blomberg
58
UA2014
Session 36: Acoustic Inversions
Location: Lecture Room D - Friday 27th June 2014
Chair: Michael Taroudakis
8:30 - 8:50
Dajun Sun, Junjie Shi, Yunfei Lv, Hualin Lan, Jidan Mei
Geoacoustic inversion using pile driving pulse and surface ship noise of opportunity based
on single vector sensor
Presenter: Dajun Sun
8:50 - 9:10
Ana Bela Santos, Paulo Felisberto, Sergio M. Jesus
Inferring ocean temperature variations from shipping noise
Presenter: Ana Bela Santos
9:10 - 9:30
Amelie Barazzutti, Cedric Gervaise
Marine mammal’s directivity in geoacoustic inversion scheme
Presenter: Amelie Barazzutti
9:30 - 9:50
Philipp Woock, Alexey Pak
Bayesian reconstruction of seafloor shape from side-scan sonar measurements using a
Markov Random Field
Presenter: Philipp Woock
Closing Ceremony
Location: Lecture Room A - Friday 27th June 2014 – 10:45-11:15
10:45 - 11:00
Young Scientist Award for the best paper presented by a graduate student
Evaluating Committee: Prof. David Bradley, Prof. Mike Buckingham, Prof. Michael
Taroudakis
11:00-11:15
Closing
59
UA2014
Poster Sessions
Location: Poster session area
Tuesday 24th June 2014
10:00-13:00
Sun Zongxin, Yu Yang, Qiao Gang, Zhou Feng,
Multi-ary code shift keying direct sequence spread spectrum using in underwater acoustic
communication
Presenter: Sun Zongxin
10:00-13:00
Jianchun Huang, Shengming Guo, Li Ma, Zhongyuan Guo, Geng Chen
Direct-sequence spread spectrum underwater acoustic communications with Turbo
equalization in time-varying channels
Presenter: Jianchun Huang
10:00-13:00
Bing Li , Yuan Li ,Hai Bo Zheng ,Zhi Bo Zhang, Qihu Li
Underwater Acoustic Communication System Simulation Based On Gaussian Beam
Method
Presenter: Bing Li
14:15 - 17:15
Zhibo Zhang, Changyu Sun, Yuan Li, Haibo Zheng, Bing Li, Xizhong Bao, Qihu Li
The Study of Time Delay Estimation Technology Based On The Cross-spectrum Method
Presenter: Zhibo Zhang, Yuan Li, Haibo Zheng, Bing Li, Qihu Li
14:15 - 17:15
Haibo Zheng, Yuan Li,Bing Li, Zhibo Zhang,Xizhong Bao,Qihu Li
The Study of Passive Ranging Technology Based on Three elements Vector Array
Presenter: Haibo Zheng
14:15 - 17:15
Sheng Yan, Minggang Liu, Chaohuan Hou
Adaptive Despeckling Method For Sas Images In NSCT Domain
Presenter: Sheng Yan
60
UA2014
Wednesday 25th June 2014
10:00 - 13:00
Thomas Leissing, Christian Audoly, Celine Rousset
Influence of ship radiated noise level directivity on the assessment of underwater noise
maps
Presenter: Thomas Leissing
10:00 - 13:00
Lanyue Zhang, Yunsheng Li, Sichun Li, Lijing Sun, Bo Wang
The fluid noise analysis on underwater high speed small vehicles
Presenter: Lanyue Zhang
10:00 - 13:00
Hanako Ogasawara, Kazuyoshi Mori, Toshiaki Nakamura
The comparison of two way sound propagation in Hashirimizu Port
Presenter: Hanako Ogasawara
14:15 - 17:15
A. Kruss, F. Madricardo, J. Tegowski, F. Foglin, L. Janowski
A combined GIS-2DFTT multi-parameter analysis of very high resolution bathymetric data:
a case study from the Venice lagoon
Presenter: A. Kruss
14:15 - 17:15
Nikoletta Diogou, Holger Klinck, Julie Oswald, Evangelos Papathanassiou, Stratis
Georgakarakos, Jeffrey Nystuen
Passive acoustic detections of odontocetes in the Ionian and Aegean Seas, Greece.
Presenter: Nikoletta Diogou
14:15 - 17:15
Yuji Sato, Hanako Ogasawara, Koichi Mizutani, Toshiaki Nakamura
Sound Pressure Field Focused by Off-Axis Aplanatic Straubel Acoustic Mirror
Presenter: Toshiaki Nakamura
Thursday 26th June 2014
10:00 - 13:00
Spyros Kouzoupis, Panagiotis Papadakis, George Piperakis
Preliminary investigation on the potential of using low power ultrasound to induce low
frequency vibrations on an immersed object.
Presenter: Spyros Kouzoupis
61
UA2014
Book of Abstracts
Session 1: Acoustic imaging
Organizer: Jiyuan Liu
Location: Lecture Room C, Tuesday 24th June 2014, 10:45 - 12:45
A Design Philosophy of Portable,High-frequence Image Sonar System
Yuan Li ,Bing Li , Zhibo Zhang ,Haibo Zheng, Qihu Li
Presenter: Yuan Li
A portable ,high-frequence sonar system for 2D imaging is developing in Institute of Acoustics,
Chinese Academy of Sciences.It consists of bistatic transducer arrays (separate transducer and 80element receive hydrophone array), signal acquisition and processing unit,weak signal amplifier
unit,power amplifier unit and so on. All the units are packaged in a watertight vessel. Through
Ethernet port,the portable sonar system can be integrated into low power, compact marine
vehicles. Prototype has been built and tested in water.It’s main acoustic feature : Frequency : 650
kHz ,number of Beams:80;beam width: 45° ? 15°; beam spacing: 0.75°; range resolution:40mm.
Pulse-echo with a high signal-to-noise ratio (SNR) of more than 10dB has been achieved. Due to
high resolution and possessing good imaging capabilities,this sonar system has dual-use
application in both military and commercial markets.The paper presents the experimental facilities,
some of which are still under development, as well as the results of trials and scope of future work.
Synthetic Aperture Sonar Images Mosaic Based on SIFT and RANSAC Method
Zelin Jiang, Wei Liu, Pengfei Zhang, Yuetao Zhou, Zhijie Wang, Yong Huang, Jiyuan Liu,
Chunhua Zhang
Presenter: Zelin Jiang
Combing a number of acoustic pings, the Synthetic Aperture Sonar (SAS) system forms a large
virtual synthetic array, improving the azimuth resolution. With a constant and high azimuth
resolution, it turns to be potential for the image mosaic based on image features. Thus a SAS image
mosaic method based on image features is proposed, which uses the SIFT operator to extract the
interest points and uses the RANSAC algorithm to do finely matching. The method was applied to
the real SAS image, indicating this effective solution.
62
UA2014
Research on underwater target detection based on seafloor physiognomymatching of side scan sonar imagery
Li Peng, Xu Feng , Zhang Chun
Presenter: Li Peng
This papers proposed an underwater target detection methods based on seafloor physiognomymatching of side scan sonar imagery, which uses the seafloor image physiognomy-matching result
between the historical sonar imagery and the measured one to detect and confirm the underwater
target. When the side scan sonar scans the same region of seafloor at different time. The sonar
imageries are different in geometry distorting, compressing and stretching changes due to the
differences of the height of the tow body, the orientation of the navigation and the gesture. These
changes could result in the deviating or disappearing of the target position on the sonar imagery,
but the seafloor physiognomy feature would not disappear. This article preprocessed the measured
sonar imagery with noise reduction, background equilibration and geometrical correction, and then
used the Shift matching algorithm, a invariant characteristic method, to extract the key terrain
feature of the sonar imagery. With the matching result of the terrain feature, the target could be
detected. The terrain feature matching algorithm of the side scan sonar imagery has a significant
meaning to the harbor safety and underwater target confirming.
Research on Target Classification for Side-Looking Sonar Based on Acoustic
Model
Xudong An, Feng Xu, Juan Yang, Jia Liu, Peng Li
Presenter: Xudong An
The acoustic model of the side looking sonar had been established based on the combination of
sonar motion, sound propagation and the target models. Based on this model, the combination
features from the highlight and shadow of the targets had been improved. The targets had been
identified by matching the features of targets from the measured data and the acoustic model.
Because the state of side looking sonar motion and the sound rays had been corrected in this model,
the impact of changes in the marine environment had been reduced by using the method of the
target classification. Also, the combination features improved the steadiness of the target
classification.
Design and Implementation of A Real-time 3-D Imaging Sonar Signal
Processing System on TMS320C6678
Peng Wang,Yong Huang, Jiyuan Liu
Presenter: Peng Wang
In this paper, the design and implementation of a real-time 3-D imaging sonar signal processing
system on TMS320C6678 are presented. owing to large number of data channels and beams, the
computational load is prohibitive for real-time 3-D imaging sonar system. In this system, a parallel
frequency domain beamforming algorithm on TMS320C6678, which is efficient and
63
UA2014
computationally advantageous, is proposed to process. And the procedures of sonar signal
processing and data transmissions are illustrated.
Numerical Analysis for Ambient Noise Imaging with Acoustic Lens: Target
Detection around the Barge Moored in Uchiura Bay
Kazuyoshi Mori, Hanako Ogasawara, Toshiaki Nakamura, Takenobu Tsuchiya,
Nobuyuki Endoh
Presenter: Kazuyoshi Mori
We already designed and made a prototype system with an aspherical lens for Ambient Noise
Imaging (ANI). That was deployed through the barge ‘OKI SEATEC II’ moored in Uchiura Bay on
November 8-13, 2010. We successfully detected some silent targets under snapping shrimp
dominant noises. The noise spatial distribution was also observed using two tetrahedron arrays.
Now, we are planning to the second sea trial at the same site in November of 2014. The main
objective is to verify whether the target can be successfully imaged in some conditions for the
direction of imaging system and the noise distributions. In this study, we tried to estimate the sound
pressures of target scatterings detected in some situations. We conducted some numerical
simulations of sound propagation using the finite difference time domain method. Arranging many
noise sources on the similar spatial distribution observed in 2010, the sound pressures received at
the focal array are calculated when the field of view looks the area where the noise sources do not
exist. In this case, the prototype system is arranged horizontally. The sound pressures
corresponding with the on-target directions are greater than those for the off-target direction,
because the receivers for only the on-target directions receive the target scatterings. On the other
hand, the sound pressures are calculated when the field of view looks the area where the noise
sources exist. Then, the prototype system is arranged vertically and looks down toward the sea
bottom. The on-target pressures are weaker than the off-target pressures, because the noises
directly received by the off-target receivers are much greater than the pressures of the target
scatterings received by the on-target receivers. These will contribute the determination of the
arrangement of the prototype system and targets in the planned sea trial.
64
UA2014
Session 2: Acoustic Monitoring of Marine Mammals
Session 2a
Location: Lecture Room A, Thursday 26th June 2014, 10:45 - 12:45
Passive acoustic monitoring of humpback whales in Exmouth Gulf using a
sparse array of DIFAR sensors
A. Gavrilov, A. Ghiotto, C. Jenner, M. Jenner , C. Gruenthal
Presenter: A. Gavrilov
Exmouth Gulf and the coastal shelf north of it in Western Australia are known as breeding grounds
for the Western Australian population of humpback whales that visit this area yearly from early
June to late October. The marine environment over this part of Australia’s continental shelf can
potentially be affected by intense offshore operations that currently take place as a result of oil and
gas exploration, the development of already discovered fields and associated port infrastructure.
Effective means are needed to prevent physiological and behavioural impacts of such operations
on humpback whales, including collisions with vessels and the effect of man-made underwater
noise. To examine the feasibility and assess the expected capability of a passive acoustic monitoring
system to detect and localise humpback whales approaching offshore operation sites, ports and
frequently used vessel corridors, a sparse array of DIFAR sensors was deployed over the shelf off
Onslow and then in the gulf. The array consisted of three to six drifting sonobuoys deployed from
about 1 km to nearly 15 km apart. Acoustic monitoring was accompanied by visual sightings. The
detection range and localisation accuracy were determined using bearing data from the DIFAR
sensors. Some new types of sound from humpback whales were recorded. It was also observed that
pods of cow and calf stayed silent most of the time; however, they were often escorted by vocalising
male whales. The research was commissioned by Chevron as part of the Wheatstone LNG Project’s
environmental monitoring program.
Acoustic monitoring of marine mammals in the Chukchi Sea – Three case
studies
Roberto Racca, Julien Delarue, Heloise Frouin-Mouy, David Hannay, Bruce Martin,
Xavier Mouy, Jennifer Wladichuk
Presenter: Roberto Racca
We present an overview of some recent acoustic monitoring studies conducted in the eastern
Chukchi Sea off the north coast of Alaska, each of them exemplifying a different analytical objective
and application of underwater instrumentation technology in a remote and challenging natural
environment. Three projects are examined, spanning a considerable range of approaches and
overall complexity. In the first two case studies ambient noise and marine mammal vocalization
data were collected with an array of over 30 autonomous recorders covering an ocean surface of
65
UA2014
75,000 square kilometres over the summer months and, with a reduced number of stations, under
pack ice during winter. These data were used for an analysis of the movement patterns of several
marine mammal species and the detectability of their calls in noise, and for research on the spatiotemporal distribution of male bearded seal vocalizations and the seasonal and diel patterns of their
acoustic repertoire. In the third case study two real-time, directional, telemetered acoustic
recording systems deployed around walrus haul-outs were used to monitor underwater calls, crossfix their location of origin and determine in post-processing their absolute source level and other
acoustic properties. Taking these examples as a basis, we examine the research possibilities offered
by long-term archival or telemetered acoustic recording systems capable of operating in extreme
environmental conditions and we speculate on future study objectives and new technological
developments to pursue them.
Passive acoustic monitoring and bio-sonar characterization of Ganges river
dolphin, India's national aquatic animal
R. Bahl, H. Sugimatsu, T. Ura, J. Kojima, S. Behera
Presenter: R. Bahl
Pioneering studies have been conducted for the Ganges River dolphin by a team of collaborators
consisting of technologists and biologists from India and Japan. This paper presents the
development of techniques for passive acoustic monitoring of these dolphins, declared as the
"National Aquatic Animal" by India. Initial studies were conducted on a lone free-swimming adult
animal confined in a river gorge in the eastern state of Odisha (Orissa) over a wide bandwidth of
30-180 kHz. The data were recorded by a 3-hydrophone SBL array with 1.6 m baseline and a small
triangular 3-hydrophone SSBL array. These measurements revealed various bio-sonar parameters:
click duration of about 40 microseconds, inter-click intervals 20-60 milliseconds, click centre
frequency 65 kHz, average on-axis source level of around 172.5 dB re 1 microPascal and estimated
beam width of about 10-14 degrees. Subsequently, in association with WWF-India an unbroken
series of in-situ long-term studies have been conducted since 2008 at Karnavas village (Narora) on
the banks of the River Ganges southeast of New Delhi. This is a habitat for a number of dolphins
(15-18) including calves and young adults. A 6-hydrophone array has been designed to monitor the
dolphin movements in 3-D in a radius of about 100 metres. The click data is continuously recorded
for six months typically during November to May of the following year. The data recording is
remotely controlled and monitored from a server via GPRS/internet to enable near-real time 24x7
monitoring of the dolphins. These multi-year studies have provided more refined information on
the bio-sonar characteristics of the dolphins and their behavioural aspects such as diurnal
movements, calf-tending, migration patterns etc. It has also for the first time given an insight into
ways to discriminate young calves from adults and sub-adults based on the click patterns. These
inter-disciplinary scientific studies are contributing to development of strategies for the
conservation of India's national aquatic animal.
66
UA2014
Passive acoustic source localisation methods for the nonintrusive monitoring
of echolocating dolphins in the wild
Eric L. Ferguson
Presenter: Eric L. Ferguson
Dolphins probe the underwater environment using sequences of short-duration underwater sound
pulses (or clicks) for echolocation that enable them to navigate and avoid collisions with natural
objects, and to detect and discriminate between prey, predators, and companions. Passive sonar
source localisation algorithms, which are based on measuring the range differences of a click at
pairs of sensors in an acoustic receiving array, are shown to provide estimates of the instantaneous
source positions of all clicks in a sequence, even when three dolphins are echolocating at the same
time. The conventional localisation method of passive ranging by wavefront curvature provides
estimates of the instantaneous range (with respect to the middle sensor) and relative bearing (with
respect to the array axis) of the source. However, the range estimates, when calculated using the
conventional method, are found to have significant bias due to a small lateral displacement of the
middle sensor away from the array axis. A modified method, which overcomes the range bias
problem when the known sensor positions are not collinear, produced identical values for the source
positions when calculated using two other methods for passive sonar source localisation by range
differences, namely the line of position and circle intersection methods. The source localisation
method is then extended from two to three dimensions by applying a single sensor multipath time
delay method to estimate the source altitude. This method requires measuring the time delay
between the signal arriving by an indirect (sea surface reflected) propagation path and the signal
arriving by the direct travel path from source to sensor. With this method, the altitude (or
corresponding depth) of an echolocating dolphin is determined to within ±5 cm, even for closelyspaced echolocating dolphins in a pod at ranges in excess of 275 m. Finally, the nonintrusive passive
acoustic source localisation methods presented here constitute a powerful scientific toolset for
studying the behaviour of echolocating dolphins in the wild.
Bowhead whale calls localization using a single receiver and warping
processing
J. Bonnel, A. Thode
Presenter: J. Bonnel
In the shallow Arctic water, the low frequency vocalizations of bowhead whales propagate
according to normal mode theory, with the environment acting as a dispersive waveguide. When
recorded a few kilometers away from the whale, the call consists in a discrete set of normal modes.
Each mode propagates dispersively with its own group speed, so that the whale range can be
inferred from the relative arrival time of the modal arrivals. However, dense vertical line arrays are
usually required to properly isolate the mode arrivals. In this presentation, we show that it is
possible to isolate the mode using a single hydrophone and a non-linear sampling scheme called
warping. Once the modes are filtered using warping, the source range is estimated using classical
67
UA2014
dispersion-based inverse methods, and the original call structure is recovered as a by-product. The
modal filtering using warping is restricted to up/down-sweep like sources, but this situation is
relatively common for bowhead calls. However, warping allows filtering modes at short range, even
when no modal arrival can be distinguished on the received spectrogram. This method is applied
on 12 bowhead whale vocalizations recorded near Kaktovik (Alaska) in 2010, with ranges between
3 and 35 km. The single-receiver range estimates are consistent with estimated ranges previously
obtained via other methods [Work supported by North Pacific Research Board and Shell Exploration
Passive Acoustic Monitoring of Marine Mammals using a Ternary Array
Liang Zhang, Adam Zielinski, Jidan Mei, Ping Cai
There is a growing interest in the monitoring of vocalising marine mammals. Their behaviour is
studied in the context of climatic changes and human civil and military activities. Passive acoustic
surveying methods are preferred since they do not affect animals’ natural behaviour. For this
reason, a variety of passive methods have been proposed but most are expensive and involve
complicated logistics. In this paper, a passive, simple and flexible ternary array is proposed for
tracking marine mammals from a small vessel or an anchored platform or floating platform. The
system consists of three surface buoys, equipped with array’s hydrophones, pressure sensors and
global positioning system (GPS) receivers. One hydrophone is a vector hydrophone. The floating
buoys are deployed from the boat, connected in chain with a rope, and float with the currents and
winds. The array can stay in a position for a continuous and prolonged length of time without the
necessity of human operators. The received data are post-processed to track single or multiple
marine mammals. A simpler version of the system utilizes an electrical cable connecting all buoys
with the platform. The cable can be used to supply power to each buoy and transmit data from
them. It is envisioned that the array of approximately 40m length will allow sufficient accuracy of
target range estimation, while target angular positions are obtained using the vector hydrophone.
The simulation results support that expectation.
68
UA2014
Session 2b
Location: Lecture Room A, Thursday 26th June 2014, 14:15 – 15:55
Acoustics as a tool to reveal population structure of the elusive blue whale
Naysa E. Balcazar, Holger Klinck , Sharon L. Nieukirk, David K. Mellinger , Joy S.
Tripovich, Robert P. Dziak , Tracey L. Rogers
Presenter: Naysa Balcazar
Our inability to directly observe animals in complex environments has limited our understanding of
elusive species. The blue whale, although the largest animal that has ever lived, has elusive
behaviour. Their pelagic habitat, wide dispersal and low population densities make field
observations difficult. The sub-species the pygmy blue whale, listed as data deficient, occurs in the
southeast Indian Ocean, yet little is known about their occurrence in the southwest Pacific Ocean.
Pygmy blue whales (Balaenoptera musculus brevicauda) produce regionally-specific calls-dialectsincluding the Madagascan, Sri Lankan, Australian, New Zealand and Solomon type calls. We
recorded year-round passive acoustic data at five sites, three in the southeast Indian Ocean and
two in the southwest Pacific Ocean (2009 to 2012) and used automated methods to detect
occurrence of different call types. Over a three year period two types of pygmy blue whale calls
(Australia and New Zealand) were detected, where the ‘Australian’ dialect dominates the southeast
Indian Ocean the ‘New Zealand’ dialect dominate the southwest Pacific Ocean. Distribution
patterns divide at the Bass Strait (southeast Australia) which appears to be a separation boundary.
Differences in temporal occurrence patterns between the ocean basins suggest the whales use
theses areas differently. Here acoustics plays a vital role in providing not only evidence of a
previously unknown population, but also insight into differences in population structure and
migration patterns across the ocean basins. We propose that these “acoustic populations” should
be considered when assessing conservation needs of blue whales in the Indian and Pacific Oceans.
The use of passive acoustic data to predict beaked whale habitat in the
California Current Ecosystem
Tina M. Yack, Alyson Fleming, Jay Barlow, Jessica Redfern1, Elizabeth Becker, Peter
Klimley, Marcel Holyoak
Presenter: Tina M. Yack
Beaked whales spend the majority of their time at depth, typically occur in small groups, and behave
inconspicuously at the surface. These factors make them extremely difficult to detect using
standard visual survey methods. To date, beaked whale habitat models have been limited in utility
due primarily to small samples of visual observations available to inform models. Recent
advancements in acoustic detection capabilities have made passive acoustic monitoring a viable
alternative method to monitor beaked whales. We used beaked whale acoustic encounters to
inform Generalized Additive Models (GAMs) of encounter rates for beaked whales in the California
Current Ecosystem (CCE) and compare these to visual based models. Acoustic and visual based
69
UA2014
models were independently developed for a ‘small beaked whale species’ group and Baird’s beaked
whales (Berardius bairdii). Distributions were modeled using a combination of fixed spatial features
and dynamic oceanographic variables. Two models for each species group were evaluated for both
visual and acoustic encounters, one using only fixed spatial features and dynamic oceanographic
variables and one that also included Beaufort scale as a predictor. The visual models that included
Beaufort scale retained this variable in the resulting best fit models, whereas the acoustic models
did not, confirming this variable’s effect on visual detection probability. The visual and acoustic
models differed markedly for both small beaked whales and Baird’s beaked whales in the predictor
variables retained and the regions of high encounter rate prediction. Model results indicate Baird’s
beaked whale habitat preferences may be distinctive from other beaked whale species. This study
demonstrates the effectiveness of using acoustic data to develop habitat models and improves
current understanding of beaked whale distribution and habitat use in the CCE. Our results can be
used to better inform management and conservation efforts for beaked whales.
Remote Sensing of Large Herring Shoals by Baleen Whales
Dong Hoon Yi, Peter Tyack, Nicholas Makris
Presenter: Dong Hoon Yi
We show that low-frequency acoustic remote sensing of large herring shoals by certain baleen
whales such as humpbacks is physically feasible in continental shelf environments. From measured
Atlantic herring shoal characteristics, humpback whale call parameters, and oceanographic
parameters in the Gulf of Maine, we determine physical sensing limitations on long-range acoustic
sensing by humpback whales, including limitations on range resolution, cross-range resolution,
maximum detection range over seafloor reverberation and noise. We show that it is physically
possible for humpback whales to detect large herring shoals at ranges of roughly 2-20 km
depending on the depth of herring shoals under typical environmental conditions during peak
herring spawning in the vicinity of Georges Bank. We find the maximum detection range should be
smallest for the deepest herring shoals and greatest for the shallowest shoals due to depthdependent target strength variations of the herring associated with the swimbladder. These
detection ranges may have affected herring behavior, who have been shown to form large prespawning shoals in deeper waters where they are less vulnerable, before moving to shallow
spawning grounds where they are more vulnerable.
Modelling the impact of ocean environment on automatic aural classification
of marine mammals.
Carolyn M. Binder, Paul C. Hines
Presenter: Carolyn M. Binder
Passive acoustic monitoring (PAM) is widely in use to study marine mammals in their underwater
habitats. Since marine mammals can be found in all ocean basins, their habitats cover diverse
underwater environments. Properties of the ocean environment, such as sound speed profile,
bathymetry, and sediment properties can be markedly different between these diverse habitats,
70
UA2014
leading to differences in how a marine mammal’s vocalization is altered by propagation effects.
This distortion of vocalizations may impact the accuracy of PAM systems. Thus, to develop a PAM
system capable of operating in numerous environments one must understand how propagation
effects impact these systems.
Previous effort has shown that a prototype aural classifier developed at Defence R&D Canada could
successfully discriminate several cetacean species’ vocalizations in a relatively limited data set. The
aural classifier was found to be an effective PAM tool because it employs perceptual signal features,
which model features used by the human auditory system. The current work used the OASES (Ocean
Acoustics and Seismic Exploration Synthesis) pulse propagation model to examine the robustness
of the classifier under various environmental conditions. Preliminary results from transmitting
cetacean vocalizations over several ranges in a simulated underwater environment are discussed.
The modelled environment used to obtain these results was based on environmental data collected
during propagation trials. Aural classification accuracy was compared for signals propagated over
different ranges and provided a preliminary measure for the robustness of the perceptual features
to propagation effects.
Classifying Humpback Whale Individuals from their Nocturnal Feeding-Related
Vocalizations
Wei Huang, Fan Wu, Nicholas Makris, Purnima Ratilal
Presenter: Purnima Ratilal
A large number of humpback whale vocalizations, comprising of both songs and non-song calls,
were passively recorded on a high-resolution towed horizontal receiver array during a field
experiment in the Gulf of Maine near Georges Bank in the immediate vicinity of the Atlantic herring
spawning ground from September to October 2006. The non-song calls were highly nocturnal and
dominated by trains of "meows", which are downsweep chirps lasting roughly 1.4 s in the 300 to
600 Hz frequency range, related to night-time foraging activity. Statistical temporal-spectral
analysis of the downsweep chirps from a localized whale group indicate that these "meows" can be
classified into six or seven distinct types that occur repeatedly over the nighttime obervation
interval. These meows may be characteristic of different humpback individuals, similar to human
vocalizations. Since the "meows" are feeding-related calls for night-time communication or prey
echolocation, they may originate from both adults and juveniles of any gender; whereas songs are
uttered primarily by adult males. The meows may then provide an approach for passive detection,
localization and classification of humpback whale individuals regardless of sex and maturity, and
be especially useful for night-time and/or long range monitoring and enumeration of this species.
71
UA2014
Session 2c
Classification of Beaked Whale Signal Recorded in Atlantic
Odile Gerard
Presenter: Odile Gerard
Beaked whales are a group of more than twenty genetically confirmed species; they are very elusive
and were among the least known species until a few years ago. Because of their sensitivity to sonar,
an increased research effort dedicated to these species started ten years ago. Fourteen different
types of beaked whale echolocation clicks are known, nine of which attributed to a species and the
other five not yet assigned. All of the signals are upsweep frequency modulated and seem to be
species specific.
In 2010, NATO Undersea Research Centre (NURC) conducted a sea trial in Eastern Atlantic Ocean,
Southwest of Portugal. Three different types of beaked whale signals were recorded. Two of them
have been assigned to a species
The third type has similarities with Cuvier’s beaked whale signals but also exhibits some
discrepancies with what is typically found in the literature. The main signal characteristics (mean
spectrum, concatenated spectrogram, Inter-Click Interval histogram…, ) are compared with those
coming from Cuvier’s beaked signals recorded with the same acoustic device in the Mediterranean.
The results of the comparison are presented.
Acoustic broadband backscattering and classification of fish
Zhang Chun, Xu Feng, Zhang Qiao
Presenter: Zhang Chun
In recent years, using of active broadband acoustic technology on the research of fishery resources,
especially on the classification of different species of fish among mixed fish schools, has gained a
lot of attentions of the researchers. Compared with the narrowband and multi-frequency signal,
the broadband signal can get continuous frequency response on the whole bandwidth. The works
propose a wideband solution to classifying the live fish. A in-situ experiment in Qingdao, P. R. China
was implemented, and broadband acoustic echoes were collected on free-swimming from the sea
bass, the yellow croaker and the alepocephalus bicolor, which live in the sea-cages, and acquired
the broadband backscattering echo of 30kHz~200kHz. By analyzing the continuous frequency
responses of the broadband echoes which backscatter from three kinds of fish, the feature vector
of the spectral structures and the statistical spectral responses are extracted for classifier. A genetic
neural network was trained on broadband echoes from each species, Tests of the trained network
suggested an overall expected correct classification rate of 80-90% on all data collections. The
results show that the broadband echo characteristics have obvious advantages as a tool for species
and size classification of fish by contrast with traditional single- and multi-frequency acoustic
echoes. In this papers, we also analyze the environment factors that affect the classification of the
fish broadband echoes.
72
UA2014
Session 4: Acoustic Tomography in Shallow Seas, Benthic and Terrestrial
Waters
Session 4a
Location: Lecture Room A, Monday 23rd June 2014, 14:15 – 15:35
Passive acoustic tomography of fluid-mud processes on the Amazon
continental shelf
Jean-Pierre Hermand, Qunyan Ren
Presenter: Jean-Pierre Hermand
A passive acoustic remote sensing technique has been developed for fluid mud tomography through
observing range- and frequency-dependent features of the vertical impedance due to a moving ship
noise (Hermand & Ren, 2013). This paper discusses the first measurements that were carried out
during CANOGA (Canal Norte Geo-Acoustics) research expedition with two small boats offshore at
the mouth of the Amazon River in Brazil, June 2012 (Hermand & Vinzon, 2012). The first boat was
used as a noise sound source and to deploy an upward-looking bottom-mounted acoustic Doppler
current profiler and a multi-instrument frame for vertically profiling salinity, temperature,
concentration of particles in suspension and fluid mud density. The second boat deployed a
proprietary vector sensor to record the sound field generated by the first boat moving in straight
line at short range and constant speed. Data from several runs were processed by a global
optimization algorithm to invert for the density and layer thickness of fluid mud and water depth.
The results are in general agreement with the direct measurement data and expected fluid-mud
processes. Furthermore, the remote sensing approach provides new insights as it does not perturb
the fluid-mud rheological behavior. The proposed technique is possible to operate under strong
tidal currents and is sufficiently fast to survey wide areas. [Work supported by FNRS, CNPq, WBI,
CAPES.]
73
UA2014
Horizontal Ocean Current Tomography with Iterative Model Weighting
Constraint
Chen-Fen Huang, Naokazu Taniguchi, Jin-Yuan Liu
Presenter: Chen-Fen Huang
In an inverse problem of ocean acoustic current mapping, prior information about the unknown
current fields is usually required to improve the model parameters due to finite number of the
acoustic rays. An iterative inversion method is applied to tomographic reconstruction of horizontal
ocean current field to improve the estimate of current fields while prior information of the current
distribution is absent. The method solves for the model parameters iteratively with an updated
model weighting constraint from the previous estimate. The performance of the method is
evaluated via numerical simulations using current fields generated from various sizes of
checkerboard stream functions. Compared to a damped least square method, the iterative method
considerably improves the current mapping. The reconstructed current field using the iterative
method yields a fractional error less than 0.2 for most of the checkerboard stream functions. The
iterative method reduces the influence of model prior information on horizontal ocean current fields
and increases the accuracy of the current mapping results in the coastal seas.
Estimation of temperature information based on ocean ambient noise
measurement
Ju Lin, Wang Huan, Wang Fengbao
Presenter: Ju Lin
Traditionally the ocean ambient noise recorded in the ocean has only been studied for
understanding of noise levels, power spectral density, vertical and horizontal directionality, etc. As
a result, ambient noise is treated solely as the noise that a desired underwater acoustic system
must overcome. On the other hand, ocean ambient noise also reflects significant and abundant
information relating to the physical characteristics of the ocean where the noise propagate. In the
recent years, a passive ocean acoustic tomography method related to ocean ambient noise has
been developed. It’s been certified through both theory and experiments that the long-time, twopoint correlation of ocean ambient noise represent the deterministic time-domain Green’s
function(TDGF) between those two points. It is difficult to extract the time-of-arrival correlation
information quickly and accurately due to the random nature of the noise field and the spatialtemporal variability of the environment. In this study, a robust Cross-correlation of Ambient Noise
(CAN) technique, which combines empirical mode decomposition and beam forming method
together, is proposed. On November 15, 2012, a ambient noise measurement experiment was
performed near the mouth of JiaoZhou Bay, Qindao, China, a ten-hydrophone array and thermistor
chain were deployed synchronously, The noise field is dominated by shipping noise by analyzing
the noise observation because the experiment location was close to the shipping traffic channel of
Qingdao Port. Among the frequency band which is lower than 1kHz, the vertical correlation
coefficient is higher because of the effect of shipping generated noise. By applying the proposed
74
UA2014
CAN method, the direct arrival time structure of TDGF is obtained, the peak become more sharp
and the SNR of correlation peak are improved comparing with those of traditional CAN method,
then the estimation of sound speed are performed, and the inversion results are in good agreement
with the simultaneous temperature measurements. (work supported by NSFC No.41176033)
A Coastal Acoustic Tomography Experiment for Tidal Current Measurement in
the South of Jiaozhou Bay
Xiao-Hu Zhu, Xiaopeng Fan, Ju Lin, Chuanzheng Zhang
Presenter: Xiao-Hu Zhu
A coastal acoustic tomography experiment for measuring the tidal current in the mouth of Jiaozhou
Bay on the western coast of the Yellow Sea was carried out with three acoustic stations in
September, 2010. The carrier of frequency 5 kHz, modulated with the 10th order M-sequence was
transmitted every 3 minutes, and the reciprocal sound transmission was successfully among the
three stations. During the sound transmission experiments, twenty-eight repeat shipboard Acoustic
Doppler Current Profiler (ADCP) surveys were also performed along the sound transmission line to
get a comparison with the reciprocal sound transmission data. Based on the result of ray simulation,
the received signals are divided into two groups which travel through the top 15 m layer and full
depth water layer, respectively. An inversion with regularization is applied to estimate current
velocity in the upper and lower layer using the differential travel times from the two ray groups.
The average current velocities along the vertical section in the upper and lower layer, determined
by the inverse analysis using the travel time differences, were in good agreement with the ADCP
results. The root-mean-square differences between two measurements in the upper and lower layer
are 4.6 cm/s and 5.8 cm/s, respectively.
75
UA2014
Session 4b
Location: Lecture Room A, Monday 23rd June 2014, 16:30 – 17:30
Acoustic investigations of unsteady salinity intrusion in a diversion channel
Kiyosi Kawanisi, Mahdi Razaz
Presenter: Kiyosi Kawanisi
Long-term continuous measurements of tidal current and sound speed/salinity have been
conducted in a tidal diversion channel using Fluvial Acoustic Tomography System (FATS) with a
couple of 30 kHz broad-band transducers. The FATS was located around 8.7 km upstream far from
the mouth. The reciprocal sound transmission that was performed between the two acoustic
stations, located on both sides of the channel, enabled us to measure range-averaged sound speed
and water velocity along a ray path. The channel is a shallow tidal-forced river with a maximum
tidal range of 4 m at the mouth. The tides are primarily semidiurnal, but mixed with a diurnal
component. The freshwater runoff into the diversion channel was regulated by the array of sluice
gates, located 270 m upstream of the observation site. Although only one sluice gate was usually
opened slightly, all sluice gates were completely opened during flood events. The saline water was
flushed out by the gate operation for flood events. Thus, the salinity intrusion in the channel
presented significant unsteady nature. Since the salinity varied in the span of 0 to 25 owing to the
tides, the sound speed was significantly influenced by the salinity. The tidal velocity amplitude Ut
and the outflow velocity Ur associated the river discharge controlled the salinity intrusion. The
recovery of salinity intrusion after the gates were set at the normal condition (slightly opened) was
hardly found when 1/(Ut Ur) was smaller than around 60 m^-2s^2. The recovering time of the
salinity ranged from 9.5 days to 27 days.
Vertical profiling of temperature and velocity from the quite limited data set of
coastal acoustic tomography
Arata Kaneko, Chuanzheng Zhang, Xiaohua Zhu, Noriaki Gohda
Presenter: Arata Kaneko
The vertical profile reconstruction of temperature and velocity is applied to two-station coastal
acoustic tomography data, obtained in September 2012 in the Akinada of the Seto Inland Sea,
Japan. Only two arrival peaks are identified in the correlation waveforms of received acoustic data
because of the mean floor depth 35 m and the station-to-station distance 13.769km. The ray paths
corresponding to the two arrival peaks with a travel-time difference of 0.005s are determined by
the range-independent ray simulation based on the observed sound speed data. The path-averaged
temperature and velocity along the two ray paths are converted into the range-averaged
temperature and velocity for the five horizontal layers (0-5m, 5-10m, 10-15m, 15-20m and 20-50m)
by the inversion with three-point regularization, accompanied by the Lagrange multiplier. The twoday low-pass filtered temperature reconstructed is in good agreement with the monthly observed
temperature. The vertical profiles of tidally oscillatory flow are also reconstructed during half the
76
UA2014
M2 period. It is found from the power spectral analysis that the 3-day oscillation of velocity occurs
even in the homogenized water. The regularization method is proposed as one adaptable for the
quite limited number of data set in the coastal acoustic tomography.
Tomographic mapping of coastal upwelling generated in Hiroshima Bay, Japan
Chuanzheng Zhang, Arata Kaneko, Xiaohua Zhu, Noriaki Gohda
Presenter: Chuanzheng Zhang
A coastal acoustic tomography (CAT) experiment of range (4-9) km was performed with four sound
transmission and reception stations (acoustic stations), surrounding the northern part of Hiroshima
Bay, Japan. The good data set of one-way travel time was acquired along the five transmission lines
although significant data lack was produced by a number of oyster rafts distributed widely over the
bay. The coastal upwelling generated along the northern shore of Hiroshima Bay by the northerly
wind derived from typhoons passing the east side of the bay is the main target of this experiment.
The station-to-station distances are corrected to attain the sufficient accuracy of sound speed
(temperature) in such a way that the sound speed determined from the travel time data is equated
to that calculated by a couple of CTD (conductivity-temperature-depth) data on each transmission
line. The inversion (grid method) accompanied by twice the moving average, is applied to
reconstruct the horizontal distribution of temperature averaged for the upper 8 m. The initiation,
growth and decay processes of coastal upwelling, are mapped with the accuracy of 0.1?. The sea
level depression of about 0.1m due to the coastal upwelling is also discussed.
77
UA2014
Session 5: Acoustics in Polar Environments
Session 5a
Location: Lecture Room C, Thursday 26th June 2014, 8:30 – 10:30
Acoustic Communications Experiments in the Fram Strait 2013
Lee Freitag, Peter Koski, Hanne Sagen, Gaute Hope, Kevin Heaney
Presenter: Lee Freitag
During the September 2013 Under-Ice research cruise aboard the K/V Svalbard, a number of
experiments supporting research in Arctic acoustic communications were performed. The
experiments focused on 900 Hz to support communications over ranges of 100 km, but also included
transmissions intended to provide insight on propagation effects at frequencies of 80 to 600 Hz. A
combination of frequency-modulated sweeps and m-sequences were used for the propagation
studies, while coded phase-shift keying (PSK) was used to make measurements of the reliability of
communications signals.
Coincident with the acoustics measurements were XCTD and XBT profiles made near the
propagation paths, which varied from less than 20 km to more than 100 km. The measured soundspeed profiles show a strong surface channel which facilitated the excellent results that were
achieved at 900 Hz for communications.
In this paper we present initial results of both the low and high-frequency propagation
measurements, along with propagation modeling that attempts to correlate the observed channel
response with the modeled results.
Measurements of the Ambient Noise Field in an Arctic, Glacial Fjord
Grant B. Deane, Jaroslaw Tegowski, Oskar Glowacki Mateusz Moskalik, Philippe
Blondel.
Presenter: Jaroslaw Tegowski
increasing attention as sensitive indicators of shifts in climate. Recently, there has been interest in
using measurements of the underwater ambient noise in the fjords of marine-terminating glaciers
as a new tool to monitor and study glacier dynamics. There already exist many useful and wellestablished measurement systems for studying glaciers. However, ambient noise oceanography –
the use of naturally generated underwater acoustic signals to study source mechanisms and the
environment through which the noise propagates – presents some attractive advantages. Sound
can propagate long distances underwater and systems to monitor underwater ambient noise for
up to a year or more are readily available and relatively inexpensive. Accurate monitoring of glacier
dynamics using underwater ambient noise is predicated on a knowledge of the mechanisms
producing the noise in the fjord and the relationship of those mechanisms to their source spectrum
and generation statistics. Here are presented measurements of the underwater ambient noise
78
UA2014
made close to Hans Glacier in Hornsund fjord, Svalbard during the summer of 2013. Building on an
earlier campaign in 2009, the objective of the field study was to see if low frequency (20 Hz – 3000
Hz) and high frequency (3000 Hz – 20000 Hz) noise within the fjord could be associated with distinct
and separate sources. The measurement system consisted of two broadband hydrophones
separated by 43 cm along a horizontal axis and oriented with a magnetic compass and tilt sensor.
This arrangement enabled the creation of maps of the ambient noise directionality at different
locations within the bay containing Hans Glacier, and tested the idea that different physical
mechanisms are associated with different frequency bands of the noise. Melting ice, ice-wave
interactions and glacier calving events were all observed to contribute to the underwater noise field
and the ambient noise field directionality was found to be a strong function of frequency. This work
has been conducted under project No. UMO-2011/03/B/ST10/04275, National Science Centre,
Poland and supported by the Ocean Acoustics Division of the Office of Naval Research, USA, Grant
No. N00014-11-1-0158.
Acoustic propagation in the Marginal Ice Zone and the implications for
navigation of underwater vehicles
Gaute Hope, Hanne Sagen, Dag Tollefsen, Hans-Christian Tengesdal
Presenter: Gaute Hope
When gliders and AUVs venture into the Arctic below the sea-ice, radio communication and GPS
signals are shadowed by the perennial sea-ice cover. Acoustic signals are currently the only practical
solution for navigation in ice-covered oceans.
Navigation systems with a sparse network of fixed transponder positions, providing signals for
navigation and positioning of gliders and AUVs, require detailed knowledge about navigation
range, precision and transmission loss.
The propagation of acoustic signals is dependent on the ice-ocean environment. A pronounced
surface channel is present in ice-covered regions, trapping the acoustic energy above a cut-off
frequency. The trapped acoustic energy is greatly influenced by the sea-ice through frequency
dependent reflection and scattering. In the Marignal Ice Zone the surface channel is gradually
reduced and highly variable in space and time. The complex conditions causes multiple, sometimes
overlapping, arrivals of a signal, which has to be accounted for at the receiver system.
In 2010-2012 Sonobuoys were deployed in three aircraft missions along several lines from the open
ocean and into the Marginal Ice Zone in the Fram Strait. The Sonobuoys recorded transmitted
tomographic and RAFOS signals from a fixed transponder at increasing distances. We have
compared the signal strength and arrival times at the Sonobuoys with the ice-ocean environment
to estimate the environmental effect on the acoustic propagation parameters and the
consequences for navigation performance.
Preliminary results from analysis of observations and modelling will be presented.
79
UA2014
Zooplankton distribution studies combining acoustical and optical
observations
Lukasz Hoppe, Joanna Szczucka, Emilia Trudnowska
Presenter: Lukasz Hoppe
Concurrent acoustical and optical measurements have a great potential to describe zooplankton
assemblages over large temporal and spatial scales. It is difficult to assess thorough information
on zooplankton distribution with traditional methods (e.g. nets), that provides only discrete and
sparse information on biomass and community structure of zooplankton, therefore use of
alternative methods, should be taken into consideration. Acoustic echosounding allows fast, nonintrusive, and relatively cheap environmental studies with high temporal and spatial resolution.
Laser Optical Plankton Counter (LOPC) proved to be well suited to multi-scale studies of zooplankton
communities. LOPC delivers real-time information on zooplankton size spectra and abundance.
Presented results are based on the data collected on the West Spitsbergen Shelf in the summer
seasons of 2012 and 2013. The high frequency 420 kHz acoustics was supplemented by LOPC
measurements along the transects, additionally net sampling, that provides information on
zooplankton taxonomy, was taken at fixed points. Size spectra measured by LOPC were used as
input parameters in “high-pass” model of sound scattering on fluid-like particles. Model output
values of acoustic backscattering strength were compared to values obtained by echosounding. In
most cases there is a good agreement between measured and modeled values, except conditions
of very low zooplankton abundance and events of fish presence. Zooplankton size structure
measured by LOPC is helpful in validating and refinement of “high-pass” acoustic model for specific
set of scatterers. This gives a possibility to determine the theoretical backscattering strength of
zooplankton aggregations. Implementing two complementary methods allows to obtain fast and
thorough information on zooplankton patches and fills the gap in comprehensive studies of the
Three-dimensional Source Localization using an Ice-mounted Geophone
Stan E. Dosso
This paper develops an approach to three-dimensional (3D) localization of a transient ocean
acoustic source using a single three-component geophone mounted on the surface of Arctic sea ice.
Source bearing is estimated by maximizing the radial signal power as a function of horizontal look
angle, applying seismic polarization filters to suppress shear waves with horizontal (transverse)
particle motion which otherwise degrade the estimate. The inherent 180-degree ambiguity in
bearing is resolved by requiring outgoing (prograde) particle motion in the radial-vertical plane.
Source range and depth estimates and uncertainties are computed by Bayesian inversion of the
arrival-time differences of the water-borne acoustic wave and ice seismic waves, including the
horizontally-polarized shear wave and longitudinal plate wave. The 3D localization is applied to
geophone recordings of impulsive sources (imploding glass light bulbs) deployed in the water
column at a series of ranges (200 to 1000 m) and bearings (0 to 90 degrees) for three sites in the
80
UA2014
Lincoln Sea north of Elsmere Island, Canada, characterized by smooth annual ice, rough/ridged
annual ice, and thick multi-year ice. Good bearing estimates are obtained in all cases. Range-depth
localization is successful for ranges over which ice seismic arrivals could be reliably detected,
approximately 200 m on rough annual ice, 500 m on smooth annual ice, and 800 m on multi-year
ice. The effects of uncertainty in ice properties (thickness and wave speeds) on localization
performance are quantified by marginalizing over unknown environmental parameters.
An acoustical study of gas bubbles escaping from melting growlers
Jaroslaw Tegowski, Grant Deane, Philippe Blondel, Oskar Glowacki, Mateusz Moskalik
According to the most recent studies, a steady and consistent increase in the underwater noise level
is observed in the Arctic fjords. This phenomenon is mainly caused by intense melting of tidewater
glaciers and other associated effects. An important component of the underwater sound budget is
the noise generated by gas bubbles escaping from melting icebergs and growlers. In July 2013 we
carried out a comprehensive study of the Hans glacier in Hornsund fjord, Spitsbergen with logistic
support of the Polish Polar Station located in Isbjornhamna. One of the main points of the study
was to make synchronous video and sound recordings of gas bubbles coming out from melting
growlers. The research was conducted in the shore area and also in a small tank, in which sounds
were registered using the HTI-96 hydrophone and WildLife Acoustics SM2+ recorder. The
hydrophone was placed a few centimeters from blocks of ice in order to record sounds generated
by erupting single bubbles or chains of bubbles. Simultaneously, a synchronized video
documentation was made as well from a distance of several centimeters. Analysis of both
registrations allowed to define precisely the link between the type of bubbles (e.g. single, chains,
several parallel, different sizes, etc.) and the emission of sounds. Analyses of transient events were
conducted using wavelet technique and spectral analysis in full spectrum and 1/3-octave frequency
bands. The potential mechanisms of sound generation are discussed.
81
UA2014
Session 5b
Location: Lecture Room C, Thursday 26th June 2014, 10:45 - 11:45
The soundscape of the Fram Strait Marginal Ice Zone
Hanne Sagen, Dag Tollefsen, Hans C. Tengesdal
A series of acoustic experiments were conducted in the Marginal Ice Zone (MIZ) of the Fram Strait
in the years 2010-12 under the Waves-in-ice Forecasting for Arctic Operators (WIFAR) project led
by NERSC. The focus of this paper is results from ambient noise measurements with fields of
sonobuoys deployed in the MIZ from the open ocean to compact ice under varying environmental
conditions. Noise spectra (10 Hz – 1 kHz) are presented and categorized by environmental
parameters that include sea state, wind force and direction, ice concentration, and ocean swell.
Scatter plot representations of noise data are explored as a tool to infer local ice conditions. The
noise fields also included components due to marine mammals and distant seismic exploration.
Under sea-ice acoustic noise and propagation measurements in Tethys Bay
(Ross Sea, Antarctica)
Andrea Bordone, Roberto Bozzano, Sara Pensieri, Paola Picco, Elisabetta Schiano,
Underwater acoustic measurements have been recently carried out in Tethys Bay (Ross Sea,
Antarctica) during the XXIX Italian Antarctic Expedition to investigate the environmental noise and
to support acoustic propagation studies in the area. Tethys Bay is a small, deep cove close to the
Antarctic Italian base Mario Zucchelli Station (Baia Terra Nova -74°42’ S e 164°07’ E), covered with
sea-ice for most of the year. During the period of the experiment (November 2013) the pack-ice had
an almost constant thickness of about 2.2 m, so that the measurements were performed deploying
the instruments into the sea from holes having 1.3 m diameter drilled in the pack ice. They were
located along the bay axis at a distance of about 500 m each other. The sea depth was around 200
m except for the hole close to the coast, were it was only 25 m. An hydrophone RESON TC 4032 was
located in the outermost hole and measurements were collected at 0, 20 and 45 m depth. The
measurements were repeated each time moving the acoustic source, a transceiver transmitting FSK
pulses at 11 kHz, in the other three holes. During the experiment, sound speed profiles, sea
temperature and salinity, currents, as well as the main meteorological parameters were
continuously measured. The acquired passive acoustic measurements evidenced that the signal was
generally dominated by different sounds from seals. Finally, the collected data-set and results from
preliminary analysis of sound intensity attenuation is presented. The matching between the
measured data and data obtained through numerical ray-tracing models of the under-ice acoustic
propagation is discussed, pointing out the physical parameters that primarily impact on the
attenuation.
82
UA2014
Sources of long-term ambient ocean sound near the Antarctic Peninsula
R. P. Dziak, H. Matsumoto, D. R. Bohnenstiehl, M. Park, W S Lee, H. Klinck, M.J. Fowler,
T-K Lau, J.H. Haxel, D. K. Mellinger, K. M. Stafford
Presenter: R.P. Dziak
Hydrophone arrays (250-1000 Hz) were deployed within the Bransfield Strait and Scotia Sea
(Antarctic Peninsula) from 2005-2009 to record sources of ambient ocean sound. Icequakes, which
are broadband, short duration signals derived from fracturing of large free-floating icebergs or
ocean front icesheets, are a prominent feature of the ocean soundscape here. Icequake activity
peaks during austral summer and is minimum during winter, likely following freeze-thaw cycles.
Iceberg grounding and rapid disintegration also releases significant acoustic energy, equivalent to
large scale geophysical events (~10^6 Joules). Overall ambient sound levels appear to be ~5-10 dB
higher in the open, deep-ocean of the Scotia Sea compared to the relatively shallow Bransfield
Strait. Noise levels become lowest during the low annual temperatures of the austral winter, likely
due to freezing of regional sea ice of all scales. Ambient noise levels are highest during austral
spring and summer presumably due to melting and cracking ice. Vocalizations of blue
(Balaenoptera musculus) and fin (B physalus) whales also dominate the long-term spectra records
in the 28, 15-28 and 85 Hz bands Blue whale call energy is a maximum during austral summer-fall
in the Drake Passage and Bransfield when ambient noise levels are a maximum and sea-ice cover
is a minimum. Fin whale vocalizations were also most common during austral summer-early fall
months in both the Bransfield and Scotia Sea. The long-term noise spectra do not show significant
anthropogenic sources (ships and air-guns), likely due low coastal populations and the difficult
marine conditions of the Southern Ocean.
83
UA2014
Session 6: Acoustics of Bubbles, Oil and Gas
Session 6a
Location: Lecture Room C, Monday 23rd June 2014, 14:15 - 15:55
A technique to measure the real surface tension on a bubble wall
Tim Leighton, Mengyang Zhu, Peter Birkin
Presenter: Tim Leighton
The surface tension of water is a key parameter for assessing the degree of contamination of
harbours, ports and open waters. However standard methods of measuring surface tension do so
at the flat air/water interface at the top of a body of water, whereas in for many important
processes, this is not the manifestation of surface tension that is most important. For biogenic
decomposition, air/sea gas exchange, the production of aerosols (all three of which are known to
be important for climate on a global scale), as well as the behaviour of ship wakes and their effects
(in distributing contaminants in the water column, affecting the local sound field or the ship’s
acoustic signature), and in the harvesting and transportation of petrochemicals, it is the surface
tension on the wall of bubbles within the water column that matters. This paper explores a
technique that can measure the surface tension on a bubble wall, and compares it with the surface
tension measured at the air/water interface. Any difference would mean that modelling of the
above effects, based on measurements of surface tension on the flat air/water interface, would
contain systematic errors with global implications.
Passive acoustic quantication of gas releases
Benoit J. P. Berges, Timothy G. Leighton, Paul R. White
Presenter: Benoît J. P. Bergès
The assessment of gas leakages from anthropogenic and natural sources is becoming increasingly
important. This includes the detection of gas leaks and quantification of the gas flux. This has
application within oceanography (natural methane seeps) and the oil and gas industry (leaks from
undersea gas pipelines, carbon capture and storage facilities). Gas escaping underwater can result
in the formation of gas bubbles, this leads to specific acoustic pressure fluctuations (sound) which
can be analysed using passive acoustic systems. Such a technique offers the advantage of lower
power requirements for long term monitoring. It is common practice for researchers to identify
single bubble injection events from time histories or time frequency representations of hydrophone
data, and infer bubble sizes from the centre frequency of the emission. Such technique is well suited
for gas release presenting low flow rate, involving solitary bubble release. However, for larger
events, with overlapping of bubble acoustic emissions, the inability to discriminate each individual
bubble injection events makes this approach inappropriate. In this study, an inverse method to
quantify such release is used. The model is first described and its accuracy at different flow rate
84 Program and Book of Abstracts
UA2014
regimes is tested against experimental data in a test tank. This compares direct measurements to
estimated flow rates from the acoustic traces of a single hydrophone. Then results from hydrophone
data collected at sea during a controlled gas release experiment (QICS) are presented.
Acoustical classification of the shallow sediment gaseous structures in the
Southern Baltic Sea
Piotr Majewski, Zygmunt Klusek
Presenter: Piotr Majewski
The common occurrence of surface gas saturated sediments and gas outflows from the sea floor
have been recognized relatively recently in characteristic regions of whole world. So far in the area
of the Polish Exclusive Economic Zone of the Southern Baltic Sea only a few trials were taken to
determine acoustically the distribution of gas saturated sediments.
In the paper, a short introduction to present-day acoustic inventory of shallow gas occurrence and
its expression in the Southern Baltic Sea is presented.
Signals of various frequencies and beamforming were employed during surveys conducted in 20092013. Simultaneous usage of several single beam echosounders applying CW signals with diverse
frequency bands from 12 up to 200 kHz allowed to distinguish different forms of gas. Simple
classification methods basing on variety of echo envelope parameters were applied to distinguish
different forms of shallow sediment gaseous structures. Several examples of echo images and effect
of their classification, obtained for different frequency bands and associated with different forms
of gas existence in sediments is presented.
Comparison of theories for acoustic wave propagation in gassy marine
sediments
H. Dogan, T. G. Leighton, P. R. White
Presenter: H. Dogan
More than three decades ago, Anderson and Hampton [1, 2] (abbreviated as A&H hereafter)
presented the theories for wave propagation in gassy water, saturated sediments and gassy
sediments in their two-part review, which has been cited by many researchers in the geoacoustics
and underwater acoustics areas. They gave an empirical formulation based on the theory of Spitzer
[3] for the wave propagation in gassy water by adapting that for a viscoelastic, lossy medium,
though without providing a detailed derivation. Following Leighton [4], this paper presents a theory
based on non-stationary nonlinear dynamics of spherical gas bubbles and extends that 2007 paper
to include liquid compressibility and thermal damping effects. The paper then show how that
nonlinear formulation can be reduced to the linear limit, and from this it derives the expressions for
the damping coefficients, the scattering cross section, the speed of sound and the attenuation, and
compares with the A&H theory. The current formulation has certain advantages over A&H theory
such as implementing an energy conservation based nonlinear model for the gas pressure inside
the bubble, having no sign ambiguity for the speed of sound formula (which is important when
estimating the bubble void fraction in marine sediments) and correcting the ambiguity on the
85
UA2014
expression for scattering cross section, as identified in the recent work of Ainslie and Leighton [5].
Moreover, the theory presented here forms a basis for a nonlinear, time-dependent acoustic
estimation model for gas bubble distributions in viscoelastic mediums since it avoids the commonly
encountered assumptions on the bubble dynamics such as linearity, steady-state behaviour and
monochromaticity.
Attenuation of low frequency underwater noise using arrays of air-filled
resonators
Mark S. Wochner, Kevin M. Lee, Preston S. Wilson
Presenter: Mark S. Wochner
This paper investigates the acoustic behavior of underwater air-filled resonators to be used in an
underwater noise abatement system. In this case Helmholtz resonators are used, which are
produced in the form of inverted air-filled cavities with a combination of rigid and elastic members.
These resonators are intended to be fastened to a framework to form a stationary array
surrounding a noise source such as a pile driving operation or seismic surveying array. Previous
work has demonstrated the potential of using arrays of large encapsulated bubbles that can
predictably attenuate sound levels over any desired frequency band and which have shown levels
of reduction up to 50 dB [Lee and Wilson, Proceedings of Meeting on Acoustics 19, 075048 (2013)].
Open water measurements of underwater sound attenuation using resonators were obtained
during a set of lake experiments, where a low-frequency electromechanical sound source was
surrounded by different arrays of resonators. The results indicate that air-filled resonators are a
viable alternative to using encapsulated bubbles for low frequency underwater noise mitigation.
86
UA2014
Session 6b
Numerical modelling of a bubble curtain
Tobias Bohne, Cristina Diaz-Cereceda, Tanja Griessmann, Raimund Rolfes
Presenter: Tobias Bohne
The offshore wind energy plays an important role in the energy concept of the German Federal
Government. The pile driving during the construction of the foundation causes high sound pressure
levels in the sea. This noise emission can be potentially dangerous for the marine life, especially
mammals. The application of different sound mitigation systems, for example big bubble curtains,
promises a reduction of the pressure level. The lack of information about the physical effects makes
it difficult to evaluate these systems and their potential for optimization.
In this paper a method based on the wave equation for modelling a bubble curtain in detail is
presented. First, the method is used to analyse a simple bubble curtain with equidistant bubbles of
the same size. In the next step the interaction between two bubbles is considered.
First results show that the bubble oscillation and the interaction between bubbles have to be taken
into account for the typical excitation wavelength associated to pile driving.
The role of air bubbles in acoustic surface loss confirmed by historical data
(1949-2005) about attenuation excess of sound in oceanic surface channels
Xavier Cristol
Surface sound channels, associated with mixing layers, are frequently observed in ocean
environments. Acoustic rays may remain confined by a barocline sound-speed gradient in the
immediate neighborhood of the surface and are impacted cumulatively by superficial phenomena
that would remain negligible in configurations involving rare encounters with the sea surface. Over
the 1-25kHz frequency range, the most visible effect is a strong attenuation excess when compared
with attenuation due to only viscosity and chemical relaxation in bubble-free seawater. Our
presentation gives results from two topics related with this question: 1/ a synthetic review of
publicly available experimental data concerning attenuation excess in surface channels; 2/ an
attempt to get some global physical understanding and a numerical prediction of this excess,
relying on models for specific features of the surface neighborhood (surface waves and swells,
stable populations of micro-air-bubbles).
Firstly, we summarize and synthetically review all published results, most notably 3 large US
campaigns of the 1945-1975 eras: AMOS data collected and analyzed by Marsh and Schulkin in
1949-53 [1], complementary NRL measurement analyzed by Saxton-Baker (1950-55); later
campaigns conducted by Hall for validating AMOS and Saxton-Baker (Hall, 1972-76 [5]). These data
investigate the frequency interval 2.kHz-20.kHz. Smaller British and US British and US campaigns of
the 60’s and 70’s (Kibblewhite & Denham, 1965 [3]; Mellen & Browning, 1978 [4], et al.) provide
87
UA2014
further information, particularly for lower frequencies (down to a few 10 Hz’s). This body of work,
when presented in a common representation format, gives a very consistent, convincing picture of
sound attenuation in surface channels, as a function of frequency and sea state.
Secondly, we try to quantify the physical phenomena mostly responsible for the observed
attenuation excess. For this purpose, we compared the predictions of numerical models, including
scattering from the rough moving sea surface and the effects of air bubbles (Hall model for microbubble population). Micro-bubble effects are sufficient for explaining the observed trends of the
acoustic loss as a function of frequency, sea state and surface channel depth. We also consider the
lesser but slightly noticeable impacts of surface roughness and large air bubbles in plumes
associated with whitecaps.
88
UA2014
Session 7: Acoustics of marine renewable energy developments
Session 7a
Location: Lecture Room D, Wednesday 25th June 2014, 10:45 - 12:45
Designing practical on-site calibration protocols for acoustic systems; Key
elements and pitfalls
Michael Butler, Jenny Norris, Paul Lepper
Presenter: Michael Butler
Although acoustic systems are increasingly being used for environmental and noise surveys of
marine energy devices, there are currently no standard protocols for the on-site full bandwidth
calibration of these systems. Reports often include little or no information on the methods of
calibration used before, during or after surveys. Clearly, without proper calibration, the sound levels
may be far from accurate.
Hydrophone calibrations from internationally recognised test centres, such as NPL, allow providers
to reference their systems to international standards. Marine renewable energy devices, however,
are often deployed in remote areas and it is not always practical or cost-effective to send every
acoustic system to be independently tested on every deployment. On-site referencing of multiple
units to a single standardised system would help improve calibration traceability. Although this
may at first appear relatively simple, producing an accurate, full-spectrum calibration, particularly
in real-world test sites, is surprisingly difficult.
This paper will present recent work from the European Marine Energy Centre (EMEC) focussing
upon improving the calibration protocols for our in-house acoustic systems. It will present the basic
elements required to set up an accurate calibration system, good practice for calibration during
acoustic surveys and a number of pitfalls that should be avoided in the calibration process. This will
be illustrated with recent data from calibrations of EMEC’s Drifting Acoustic Recorder & Tracker
(DART) units. Finally, the paper will discuss the fine balance between a cost-effective calibration
process and the limitations that this will inevitably bring.
An environmental survey around the Narec Offshore Anemometry Hub (NOAH)
– a comparison between acoustic measurement instruments.
Peter Dobbins, Federica Pace, Irene Vollmy, Silvana Neves, Sophie Nedelec
Presenter: Peter Dobbins
In July 2013 a survey was conducted covering the area around the Narec Offshore Anemometry
Hub (NOAH), 3km offshore of Blyth in North East England. The site will eventually host an array of
12 experimental wind turbines and the object of this baseline survey was to provide an initial
dataset for analysis and also enhance understanding of instrument operation and sensitivity. The
survey was carried out by members of the Bio-Acoustic Research Consortium (BARC) using the
UA2014
Newcastle University research vessel RV Princess Royal. A number of acoustic (pressure)
sensors/recorders were deployed, some of which were built in-house by consortium members and
some of which were Commercial Off The Shelf (COTS) devices, along with a particle velocity sensor
as well as some grabs and trawls for collecting various species for lab-based behavioural analysis.
A visual survey was also undertaken along a zig-zag transect for comparative analysis of detection
rates. This paper will describe a comparison between the acoustic measurements obtained with
two COTS instruments: a Songmeter SM2M (Wildlife Acoustics) and SSQ906G LOFAR sonobuoys
(Ultra Electronics), along with results obtained with the particle velocity sensor. The analysis covers
a total duration of 2.5 hours of simultaneous recordings from the SM2M and sonobuoys. In
summary, some high frequency clicks were recorded in the SM2M files which could not be detected
by the sonobuoys. There were some “clacking” noises associated with RV Princess Royal. Initially
there appeared to be more higher-frequency masking noise in the sonobuoys recordings. However,
these devices apply a pre-whitening filter before the data are transmitted and it is anticipated that
applying an inverse filter will restore the normal ambient noise low frequency emphasis. The paper
will present examples of sounds recorded with both instruments, along with power spectra and
other analyses and will explore the reasons for the differences.
Methods for measurement of long term radiated noise from a wave energy
system
Paul A. Lepper, Stephen P. Robinson
Presenter: Paul Lepper
Over the last few decades we have seen a significant growth of offshore renewable development.
Offshore wind power is well established at a commercial scale, however more recently we are seen
emerging industries for both wave and tidal generators. Partial and full scale development systems
have been developed using a wide diversity of technologies over the last few years. For a number
of these technologies the development of large scale projects is currently underway with
commercial scale operation likely in the next few years.
These developments require assessment of potential environmental impact across a wide diversity
of areas including underwater noise generated by these systems both during construction and
operation. To date there exist no standard measurement methodologies for assessment of radiated
noise from these systems available to the industry.
Moreover, there are a number of significant challenges with regard to measuring the radiated noise
from both wave and tidal energy devices, including operation in harsh, highly dynamic
environments, measurement and
platform self-noise, complex acoustic interactions
(anthropogenic, environmental, biological, etc.) and strong temporal variations that need to be
captured to characterize these source. The physical diversity of the emerging renewable energy
system technologies also poses significant challenges to the development of standardized
measurement methodologies.
90
UA2014
This paper discusses of some of these challenges and provides a description of attempts to develop
a generic measurement methodology for assessment of radiated underwater noise from wave
energy systems. Data is presented for measurements conducted on a full scale Pelamis wave energy
converter under a variety of sea-state.
Environmental inversion with an autonomous hydrophone in a wave energy
device deployment site
Cristiano Soares, Erica Cruz, Friedrich Zabel, Andre Moura
Presenter: Cristiano Soares
This paper presents environmental inversion results of acoustic data in shallow water in Peniche
(Portugal), collected in September 2013, during the Simple Underwater Renewable Generation of
Electricity (SURGE) Project - a FP7 European collaborative demonstration project aiming at building
a grid connected wave energy converter of type WaveRoller.
A single autonomous hydrophone was moored at the position foreseen for the Waveroller
deployment. Computer generated acoustic signals were transmitted over a 3 km oceanic transect
with a step of 300 m. Incoherent transmission loss (TL) for 1/3 octave frequencies in the band from
318 to 1270~Hz is used in an acoustic inversion procedure for the estimation of geoacoustic
parameters of a two-layer seafloor. The acoustic inversion is posed as an optimisation problem
aiming at minimising the root mean square error (RMSE) between field TL and replica TL, using a
genetic algorithm. This procedure is repeated a number of times in order obtain a posterior
distributions for each unknown parameter, and the solution of the inversion is obtained by taking
the maximum of the a posteriori distribution of each unknown parameters.
The RMSE between the field TL and acoustic model TL obtained for the solution across the acoustic
transect varies between 1.5 and 2.6 dB, depending on the frequency. As a procedure to validate the
obtained model, the RMSE between field TL and model TL is calculated for an alternative frequency
band (from 1600 to 8064~Hz), in order to check the amount model mismatch for frequencies not
entering the inversion procedure. In that case, the RMSE varies between 2.5 and 4.6~dB. This
increment in the RMSE can be considered relatively small, allowing the obtained physical model to
be considered meaningful, and therefore adequate for noise modelling purposes over an eventual
impact area.
Field deployments of a self-contained subsea platform for acoustic monitoring
of the environment around marine renewable energy structures
Benjamin Williamson, Philippe Blondel, James Waggitt, Paul Bell, Beth Scott
The drive towards sustainable energy has seen rapid development of marine renewable energy
devices, and current efforts are focusing on wave and tidal stream energy. The NERC/Defra
collaboration FLOWBEC-4D (Flow, Water column & Benthic Ecology 4D) is addressing the lack of
knowledge of the environmental and ecological effects of installing and operating large arrays of
these devices. The FLOWBEC sonar platform combines a number of instruments to record
91
UA2014
information at a range of physical and multi-trophic levels at a resolution of several measurements
per second, for durations of 2 weeks to capture an entire spring-neap tidal cycle. An upward-facing
multifrequency Simrad EK60 echosounder (38, 120 and 200 kHz) is synchronised with an upwardfacing Imagenex Delta T multibeam sonar (120° x 20° beamwidth, 260 kHz) aligned with the tidal
flow. An ADV is used for local current measurements and a fluorometer is used to measure
chlorophyll (as a proxy for plankton) as well as turbidity. The platform is self-contained, facilitating
rapid deployment and recovery in high-energy sites. Five 2-week deployments were completed in
2012 and 2013 at wave and tidal energy sites, both in the presence and absence of renewable
energy structures. These surveys were conducted at the European Marine Energy Centre, Orkney
(Scotland). Using multifrequency target identification and multibeam target tracking, the depth
preference and interactions of birds, fish schools and marine mammals with renewable energy
structures can be tracked, together with dive profiles and predator-prey interactions.
Measurements from the subsea platform are complemented by 3D hydrodynamic model data,
concurrent shore-based marine X-band radar and shore-based seabird observations. These
datasets offer insights into how fish, seabirds and marine mammals successfully forage within
dynamic marine habitats, how marine energy devices might alter the behaviour of such wildlife and
whether individuals face collision risks with tidal stream turbines.
Underwater Sounds from Drillships and Support Vessels During Exploratory
Drilling Offshore Alaska
Melanie E. Austin
Underwater sound levels were measured during drilling by Shell of two exploratory wells offshore
Alaska in 2012. Two drillships were used to drill top holes of exploratory wells, one in the Chukchi
Sea and one in the Beaufort Sea. A dedicated acoustic monitoring program was performed to
measure sounds as a function of range and direction from each of the drillships during different
stages of the operations and from each support vessel involved with the project. Drilling sounds
were characterized during excavation of a mudline cellar and while drilling. Sounds from ancillary
activities including anchor handling and ice management were also measured. The greatest sound
levels were measured during anchor setting and during excavation of the mudline cellars. Measured
vessels included supply ships, anchor handling vessels, ice breakers, and an oil spill response fleet.
Vessel sound signatures were found to be directional with higher received levels obtained from
abeam the vessels relative to forward and aft of the vessels. The greatest sound levels were
generated by the anchor handling vessels. Subsequent sound level measurements of the ice breaker
vessels on dynamic positioning were collected in the Chukchi Sea in 2013 to better characterize
sounds from this vessel activity. Underwater sound characterization results for each of the
aforementioned sources will be presented.
92
UA2014
Session 7b
Underwater sound due to a subsea high speed turbo-compressor
Bas Binnerts,Pieter Van Beek,Sander Von Benda-Beckmann, Erik Nennie
Presenter: Bas Binnerts
In the oil & gas industry there is a trend towards more subsea activities. To improve gas recovery
from existing and new fields at greater depths, the produced gas will be compressed, processed and
transported via subsea templates and underwater networks (pipelines, flexible risers, etc.). Besides
the huge consequences for the subsea installation itself (reliability, maintenance, etc.), it also has
consequences for underwater wildlife through the underwater source vibrations leading to sound
radiation. Regulations aimed at managing the impact of underwater sound on marine life have
been put in place by different nations. Many offshore operations require an assessment of the
potential impact of underwater noise on the environment, which requires knowledge of the sound
transmitted by the subsea components.
Until now very little is known about the underwater source mechanisms, the acoustic strength of
these underwater networks, the coupling of the emitted source sound to the surrounding medium
and the impact of the sound on the underwater wildlife. The dynamic behavior of networks for
compressing and transporting gas, and the translation into emitted noise into air are rather well
understood. However, due to the presence of the water the dynamic behavior from such subsea
installation is very different than in air. To predict the dynamic behavior, the presence of the water
cannot be neglected and has to be taken into account.
This paper presents a simplified model for a subsea high speed turbo-compressor coupled to the
KrakenC normal mode propagation model. With this combined model the noise at remote locations
can be predicted and compared with the ambient noise and other anthropogenic noise sources such
as for instance shipping, dredging and wind farm operation noise.
New methods in impact pile driving noise attenuation
John T. Dardis Ii, Per G. Reinhall
Presenter: John T. Dardis Ii
Under water noise from impact piling reaches peak sound pressure levels on the order of ~103 Pa
at range 3000 m, ~104 Pa at range 60 m and ~105 Pa at range 10 m. These peak pressures have
deleterious effects on underwater fauna resulting in regulatory agencies placing limitations on
offshore impact pile driving.
In this paper, we review current noise attenuation methods and discuss their limited success. Our
inspection of the mechanics behind impact pile driving finds the rise time in the pile’s propagating
stress wave responsible for peak acoustic pressures. This knowledge guides us to alternate methods
to achieve noise reduction.
93
UA2014
Of the possible methods, two stand out as economically feasible, structurally sound and high
implementable. It is these two methods, the double pile and the zero Poisson’s ratio pile, that are
presented in detail. Both finite element analysis and sub-scale testing promise greater than a 20
dB reduction in peak sound pressure levels and full scale testing is in preparation. Noise reduction
at these levels will have a profound positive effect on the industry of marine piling.
Soil vibration due to offshore pile driving and induced underwater noise
Katja Reimann, Jurgen Grabe
Presenter: Katja Reimann
The topic of hydro sound protection to marine mammals is frequently discussed in Germany due to
the aimed expansion of offshore wind energy in the North Sea. As a result of the impact pile driving
considerable hydro sound pressure levels are emitted into the sea. The German Standards on
underwater noise demand a limit of sound exposure level of no more than 160 dB in a distance of
750 m from the pile driving location. Therefore different noise mitigation techniques are under
development to decrease the hydro sound level significantly. Nevertheless these limits are not met
often.
All of these measures like bubble curtains or cofferdams have in common that they act on the direct
transmission path between pile surface and water. During the pile driving the major part of the
impact energy is transmitted into the soil which results in wave propagation. The influence of the
subsoil on the emitted hydro sound is not clear by now.
The German research project BORA concerns hydro sound emissions and the prognosis with
numerical simulations. To a better understanding of the whole physical process and all the
influences on the hydro sound under offshore conditions, detailed numerical models are set up. To
validate these models extensive offshore measurement campaigns are conducted in 2012 and 2013
in this project.
This paper gives an overview of the measured data of the pile dynamics, the soil vibration and the
hydro sound pressure to follow up the secondary transmission path from the pile through the soil
into the water. The data of the pile dynamics and the vibration of the soil are evaluated to estimate
the possible influence on the hydro sound.
Model results for offshore piling acoustics featuring an elastic sediment with a
depth-dependent shear wave speed
Michael Wood, Victor Humphrey
Presenter: Michael Wood
The noise from marine piling operations has come under increased scrutiny due to the accelerated
development of offshore wind farms. Due to the high levels associated with piling noise there is
concern that this may adversely affect marine life. For this reason improved methods are being
developed to predict these levels. This paper describes a finite-element model of the near-field
94
UA2014
acoustics of a pile in situ. Pile parameters are in line with those specified for the COMPILE
benchmark model. Two cases are considered: one with a fluid sediment, and one with an elastic
sediment. The shear wave speed in the elastic sediment case is dependent with depth based on a
power law; this provides a very low shear wave speed at the interface between the water and the
sediment. The compression wave speed in the sediment is kept constant with depth. The results
from the comparison show differences in the noise generated due to the two sediments. In the fluid
sediment case the nature of the received acoustic signal is typically tonal in nature with a frequency
determined principally by the sound speed and length of the pile. Conversely, by including the elastic
seabed the signal takes on more broadband characteristics. In addition, the inclusion of shear
propagation with the power law speed dependence gives rise to the slow moving ground roll wave
that has been recorded in piling events before. The results here highlight the importance of taking
the shear propagation into account in the sediment as the character of the piling pulse is greatly
affected. Moreover, the interface wave, that is known to be present in measurements, is only
modelled when an elastic sediment is considered.
The evanescent pressure waves associated with ground roll waves from seabed
impacts
Hazelwood R, Macey P
Presenter: Dick Hazelwood
Disturbance of the seabed will radiate waves in both media. Whilst most studies of biological
impact have looked at pressure waves in the water, the majority of aquatic life is more sensitive to
motion, not having evolved pressure sensors.
More detail will be presented on the interaction of the rolling motion of the seabed ,the ground roll,
and the nature of the localised pressure varaitions close to the seabed. These provide a way to
measure the waves without the uncertainties due to the mounting of vector sensors such as
geophones
Effective Reduction of Offshore Piling Noise
Karl-Heinz Elmer
Presenter: Karl-Heinz Elmer
High underwater noise emissions from offshore piling are potentially harmful to marine life and can
reach dangerous levels in a large area. An increasing number of erected and pro¬spec¬ted offshore
wind turbines needs effective noise reducing methods to achieve the German BSH standard level of
160 dB SEL at a distance of 750 m from pile driving.
The innovative method of hydro sound dampers (HSD) uses curtains of robust air filled elastic
balloons with high underwater noise reduction effects and special PE-foam elements with high
dissipative effects from material damping, to reduce impact noise. The resonance frequency of the
elements, the optimum damping rate for impact noise, the distribution and the effective frequency
range can be fully controlled.
95
UA2014
The theoretical background of hydro sound dampers is explained. Numerical simulations of
radiated noise from pile driving, simulations of noise mitigations using hydro sound dampers, and
test results of measured HSD noise mitigations are presented.
Offshore tests have already shown the high potential of HSD-noise mitigation even in the lower
frequency range of today’s large hydraulic hammers between 50 and 350 Hz. In the first half year
of 2014 a new HSD-noise mitigation system is applied to the monopiles of an offshore wind farm in
the North Sea. To reduce the mobilization time the complete HSD-system is hanging below the
hydro hammer. It is expected that this first serial offshore application will demonstrate a new
effective way to reduce offshore piling noise. The first measured results of noise reductions will be
presented and discussed.
96
UA2014
Session 7c
Underwater noise assessment of wave energy devices
Erica Cruz, Ines Machado, Teresa Simas
Presenter: Erica Cruz
Environmental Impact Assessment (EIA) of ocean energy projects is not only a legal requirement
but also an assurance of the projects sustainability, a promoter of public acceptance and a benefit
for industry, making the project more attractive to investors and governments who traditionally
have seen environmental concerns as a barrier.
Noise is one of the environmental descriptors of high relevance and concern in any project or
anthropogenic activity, ocean energy projects are not an exception. The increasing number of wave
farms or demonstration areas becomes an opportunity to learn, analyze and evaluate potential
environmental impacts including the underwater noise impact on the marine environment.
In general, a best practice to assess the underwater noise of a given source involves its
characterization, the characterization of the baseline acoustic environment, the transmission path
and the characterization of the receptor. Considering the underwater noise variability it is necessary
to analyse long data sets of measurements which might not be available or might not be possible
to acquire in a short period of time.
As ocean energy is still in a demonstration phase, the project’s deployment period is usually
uncertain, sometimes for short periods of time, making it difficult to coordinate monitoring
programmes in order to accomplish authoritie’s requirements in terms of EIA process deadlines.
This study aims to review three monitoring plans that were or are being applied to the underwater
noise assessment of the wave energy projects in Portugal and Spain in order to develop a
methodology that fits the permitting requirements without compromise acoustic data quality.
Underwater sound levels at a wave energy device testing facility in Falmouth
Bay, UK
Joanne K. Garrett, Matthew J. Witt
Presenter: Joanne Garrett
There is a paucity of evidence on the noise produced from in situ wave energy converters (WECs)
during all stages of their deployment, operation and decommissioning. The aim of this research is
to gather empirical data to address this knowledge gap and to inform the consenting process.
A WEC has been trialled at the Falmouth Bay Test Site (FaBTest), in Cornwall, UK since March 2012.
The area supports considerable commercial shipping and recreational boating along with diverse
marine fauna, including bottlenose dolphins and harbour porpoises.
A passive acoustic monitoring device, recording broadband sound in the effective frequency range
10 Hz to 32 or 48 kHz, for half an hour in every hour, has been deployed at the FaBTest site since
March 2012. Underwater sound monitoring covered a two week baseline period, a five day
97
UA2014
installation period, testing periods when the WEC was producing power and when the device was
not producing power but was in situ.
The median sound level during the baseline period ranged from 60-80 dB re 1 µPa in the frequency
range 0.01-10 kHz. It is likely that the considerable shipping present at the site affects the sound
levels.
Sound levels were, on average, higher during installation activity compared to periods of no
installation activity in the frequency range 10-5000 Hz with a median increase of 8.2 dB re 1 µPa
(interquartile range = 6.7 dB).
Average sound levels were found to be louder at times when the WEC was producing power
compared to times when the device was in situ and not producing power in the frequency range
10-1000 Hz with a prominent peak in the frequency range 57-63 Hz.
From the long term monitoring of the site it has been identified that the sound levels are highly
variable, and it is difficult to determine the effect of the wave energy converter in such a variable
ambient noise environment.
Cabled observatory enabled acoustic monitoring of hydrothermal discharge
Karen Bemis, Guangyu Xu, Darrell Jackson, Peter Rona, Russ Light
Presenter: Karen Bemis
New technological developments have recently enabled long-term monitoring through cabled
seafloor observatories. In order to increase understanding of the complex interactions of
earthquakes, volcanic eruptions, ocean waves and the hydrothermal systems, the Cabled
Observatory Vent Imaging System (https://sites.google.com/a/uw.edu/covis/ ), known as COVIS,
has been designed for long-term monitoring via remote sensing of multiple types of hydrothermal
discharge. The imaging and Doppler modes use backscatter from the plume itself to sense the
boundaries of the plume and the vertical flow rates within the plume. The diffuse mode uses the
decorrelation effect of turbulence in front of a solid object to detect areas of outflow on the
seafloor. COVIS is currently installed at the NEPTUNE observatory's Endeavour site
(http://www.oceannetworks.ca/installations/observatories/neptune-ne-pacific ); COVIS images
the plumes rising from Grotto Vent on the Endeavour Segment of the Juan de Fuca Ridge off the
western coast of North America. COVIS's observations of plume bending and volume flux highlight
the potential for long-term monitoring via remote sensing in this Marine Protected Area. Recent
studies have lead to the inversion of volume flux observations for heat flux in the discharging
plumes and to the recognition of the transient affects of atmospheric storms on plume behavior.
98
UA2014
Applications
Organizer: Alessandra Tesei
The SMO antenna: status and first results
S. Viola, G. Riccobene, F. Simeone
Presenter: S. Viola
Submarine Multidisciplinary Observatory (SMO), deployed on March 2013 at about 100 km off
Sicilian coasts, is a cabled underwater acoustic antenna composed of 10 large bandwidth (10 Hz –
70 kHz) hydrophones, installed a different depths, from 3150 m to 3350 m. The goal of SMO
(https://web.infn.it/smo) is a long term monitoring of the underwater acoustic environment for
studies on astrophysical, biological, geophysical fields. Acoustic signals acquired from sensors are
continuously sampled off-shore and transmitted through optical link to the shore station for realtime analysis and recording. Thanks to an innovative data acquisition system based on underwater
time stamping, all acoustic sensors are synchronized and phased with the absolute GPS time. First
results and future extension towards a very large acoustic array (hundreds of sensors) will be
presented.
99
UA2014
Processing strategies for evaluating the ship radiated noise using an
underwater vertical array
Tomaso Gaggero, Mirko Bassetti, Erica Firenze, Alessandra Tesei, Andrea Trucco
Presenter: Tomaso Gaggero
The problem of the assessment of the acoustic impact of ships on the marine environment is gaining
an important role in the international community pushing the shipping sector to face with such a
problem. A key aspect in the evaluation of the effects of ships’ noise is represented by the source
characterization and therefore by the onsite measurements of the noise emitted. Measurements of
noise at sea can be carried out following different approaches: while in the navy field fixed acoustic
ranges are used, in the civil field a more flexible and portable system can be needed and an array
of hydrophones deployed by a supply vessel represent a solution. An example of such a system was
presented at UA2013 [1] and consists of an array of three digital hydrophones, each equipped with
a depth sensor. In the present paper a post processing software developed in Matlab® is presented
to elaborate the outputs of the above mentioned system. The inputs of the processing software are
the GPS signals of both the supply and the target vessel, and the volt signals from the hydrophones.
The results after the software processing not only cover the needs of the ANSI/ASA S12.642009/Part1 standard but it is also able to provide a wider spectrum of information on the acoustic
emissions of the vessel such as both vertical and horizontal directivity. Finally the possibility of using
the array for a spatial filtering technique, in order to improve the SNR, is presented.
[1] Alberto Figoli, Alessandra Tesei, Ruggero Dambra, Mirko Bassetti, 2013. A DIGITAL VERTICAL
ARRAY WITH VARIABLE GEOMETRY FOR ACCURATE MEASUREMENT OF UNDERWATER SHIPRADIATED NOISE, 1st international conference and exhibition on Underwater Acoustics (UA2013).
100
UA2014
Development of ultra sideband transducer
Toyoki Sasakura
Presenter: Toyoki Sasakura
The bandwidth of most conventional transducer using echo-sounder is narrowband which relative
bandwidth (bandwidth/center frequency) is about 0.1 values, less than one octave bandwidth. The
recent demand of broadband transducers has been increased in the field of underwater acoustic
detection to improving signal to noise ratio and range resolution. Broadband transducers have also
been required in the dolphin vocalization research works and/or in the high resolution echo
sounding research works.. Until now scientists studying dolphin vocalizations have used lowfrequency (10kHz~50kHz) and narrowband (<20 kHz) underwater speakers, which can reproduce
only a part of vocalizations dolphins use (over several tens Hz to 150 kHz), for playback experiments
to dolphins. If a broadband transducer could be developed, it would enable to reproduce a variety
of dolphin vocalizations and enhance understanding of them. We had developed the transducer
covering 20 kHz – 150 kHz in 2011 for echo-sounder. However this transducer do not have enough
bandwidth for our purpose of playback of various dolphin vocalizations. Therefore we have applied
the creative technique and tried to design ultra wideband transducer that has four octaves
bandwidth. The design and characteristics of 4-octave transducer were described. And also the
application of ultra wideband transducer is introduced.
Basic Study of Rhomboidal Acoustic Lens Constructed with Phononic Crystal
Takenobu Tsuchiya, Tetsuo Anada, Nobuyuki Endoh, Sayuri Matsumoto, Kazuyoshi
Mori
Presenter: Takenobu Tsuchiya
A high-resolution acoustic imaging system is an important aid in turbid water where an optical
camera system fails. Reentry, phononic crystal (PhC) is synthetic materials that are formed by
periodic variation of the acoustic properties of the material. Planate ultrasonic lens constructed
with PhC that have a negative refractive index for ultrasound waves was designed by Dr. Zhang. In
previous study, we fabricated a prototype of a planar acoustic lens made from stainless steel rods.
However, the focal gain of the fabricated planate acoustic with PhC was lower than conventional
lens, because the planar acoustic lens with negative refractive index allowed focusing of a sound
when a spherical wave was incident to the plate. To fabricate a high gain lens constructed with PhC,
we design the a rhomboidal lens by focusing plane wave. Therefore, it is necessary to measure the
dependence of negative refractive index at the plate constructed with PhC by incident angle of
plane wave. In this paper, to determine the dependence of refractive index at the plate constructed
with PhC by incident angle of plane wave, we measured the direction of sound propagation getting
through the plate which was rotated from 0 degree to 30 degrees. And, we design a rhomboidal
acoustic lens.
In a basic experiment, we measured the dependence of negative refractive index of the plate
constructed with PhC by incident angle of plane wave was measured. And, we simulated basic
properties of rhomboidal acoustic lens with PhC by numerical analysis method. As a result, it is
UA2014
clearly shown that the refractive index of the plate is about -0.5 when incident angle was varied
from 10 degrees to 25 degrees. The rhomboidal PhC lens has a higher gain about 7 dB than the
planar PhC acoustic lens at f = 700 kHz and it has a very narrow beam.
102
UA2014
Session 9: Calibration of Sonar and Hydrophones
Organizer: Bo Lövgren and Stefan Schael
Location: Lecture Room C, Wednesday 25th June 2014, 10:45 - 12:25
Low Frequency Tank Calibration by comparison
Bo Lovgren
Presenter: Bo Lövgren
In order to check the performance of Low Frequency Sonar Systems (1 kHz and below) there is a
need for Calibrated Sources that can be arranged at different distances and in different
environments. For that purpose Saab has developed a Signature Generation Module that fits in an
AUV, and makes these performance checks easy to perform with low personal effort. In order to
calibrate the source itself, it would be feasible to use a medium-sized test tank, but because of the
small dimensions in relation to the wave-length there will be errors due to multi-path.
This paper describes a method where the Source Level of one Calibrated Source is first determined
in absolute terms in free-water, where after it is checked in the indoor Test Tank using a calibrated
hydrophone. The difference in Estimated Source Level is then used for characterization of the Test
Tank. This means that subsequent Calibrated Sources of the same design will only need to be
checked in the Test Tank, in order to Estimate the True Source Level.
The paper shows an example of such a calibration scheme, together with the corresponding results
and a discussion around possible sources of error.
Adulteration of underwater acoustic measurements
Stefan Schael
Presenter: Stefan Schael
The awareness of underwater threat will be always a constant companion of the NAVY. All efforts
were and will be done to analyze and minimize the unknown threat.
The underwater emission generated by different components and hydrodynamic behavior are
evaluated by a hydro acoustic measurement system. Procedures and recommendations for
silencing the underwater signature are dependent on the accuracy of the range facilities. The
influences caused by the environmental area, uncertainties of the steady state of the ship and the
technical range system have to be taken into account.
The German subordinated Technical Center for Ships, Naval Weapons Maritime Technology and
Research (WTD71) of the Federal Office of Bundeswehr Equipment, Information Technology and InService Support operates a fixed underwater signature range in shallow water and in cooperation
with Norway and the Netherlands a fixed underwater acoustic range in deep water. In non-routine
cases, a mobile measurement system will be deployed in shallow water areas. The development
and installation, including the quality acceptance tests of those measurement systems, are carried
out by WTD 71, which is accredited to perform the calibration of Hydrophones. Periodical
inspections and calibrations assure the high demands on the quality.
103
UA2014
The quality of the electronic measurement chain depends on the calibration of the sensors and the
installed components. A short overview of the measurement chain will present the components in
use. The procedure of the system calibration will be shown as well as the comparison of different
international mobile range systems, measuring the same sound source at the same location.
The calibration of hydroacoustic channel of mobile measurement module
Krystian Buszman, Ignacy Gloza, Rafal Jozwiak, Karol Listewnik
Presenter: Krystian Buszman
In recent years, there was noticed an increased interest in research how the same systems or
different systems with the same declared parameters for measuring ship signatures at different sea
regions for different environmental and hydro-meteorological conditions. It particularly relates to
measuring the acoustic signature of the ship. The article describes a detailed procedure for
calibrating hydroacoustic measurement channel of mobile module made by the Polish Naval
Academy. The calibration chain consists of: hydroacoustic transducer with elements of electronics
and acoustic channel with analog-to-digital converter and signal processing procedures using time
domain and one-third octave (OTO) analysis. Calibration takes into account test insert voltage by
specifying a sinusoidal signal with constant amplitude with center frequencies in the band OTO
from 3 Hz to 25 kHz. The results of the calibration tests are described in the article. Performing this
calibration procedure it was necessary to compare the results obtained with the use of acoustic
measurement systems with other participants in the study. These studies were carried out under
the European Defence Agency project SIRAMIS which will provide knowledge on ship signatures
collected using the systems of the various participants in different sea areas under different
environmental conditions.
Calibrating hydrophones at very low frequencies
Tom Dakin
Presenter: Tom Dakin
Low frequency hydrophones are a cost effective way to augment seismometers. However facilities
for calibrating hydrophones at very low frequencies are not readily available. This paper describes
the design and initial results of the Ocean Networks Canada VLF hydrophone calibration system.
Array Shape Estimation Using Measurements of Heading and Depth Sensors
Xiuting Yang, Min Li, Yanhong Hu
Presenter: Xiuting Yang, Min Li, Yanhong Hu
Array shape usually deforms when the towed ship makes manouvre, and it leads to the
performance degradation of towed line array sonar. One important way to solve this problem is to
estimate the array shape realtime and compensate the time delay caused by array deformation. In
this paper, a method based on Paidoussis equation using measurements of heading/ depth sensors
has been proposed to estimate the array shape by two steps: firstly, the array shape is predicted by
104
UA2014
the conventional Paidoussis equation by numerical approach; secondly, the measurements of
heading/depth sensors have been used to check and modify the accuracy of array shape estimation.
Simulation and sea trial results confirm the efficiency of this method.
105
UA2014
Session 10: Comprehensive Nuclear-Test-Ban Treaty Monitoring
Organizer: Georgios Haralabus and Mario Zampolli
Location: Lecture Room A, Tuesday 24th June 2014, 8:30 - 10:10
Results from ocean currents and acoustic propagation modelling studies in
support of the installation of CTBTO Hydroacoustic station HA04, Crozet
Islands, France
Mario Zampolli, Georgios Haralabus, Jeremy P. Stanley, Lucie Pautet, Mark K. Prior,
Patrick Marsaleix, Florent Lyard, Kevin D. Heaney, Richard Campbell
Presenter: Mario Zampolli, Georgios Haralabus
The Crozet Islands (French Southern and Antarctic Territories), the location in which CTBTO
International Monitoring System Hydroacoustic Station HA04 is to be installed, is a particularly
challenging oceanographic environment in the south-western Indian Ocean. Strong currents and
internal tides which can extend to depths of about 1000 m, as well as a very complex bathymetry,
make it challenging to find ocean bottom deployment sites for the six hydrophone moorings that
offer a good balance between hydrophone acoustic coverage and mooring sustainability. In order
to fully understand these crucial issues, the CTBTO initiated two modelling projects with the aim of
developing a consolidated environmental and acoustic picture to facilitate the identification of the
most suitable locations for the hydrophone installations. A 3-D time-domain oceanographic
modelling study showed that the interplay between known circulation currents, which go around
the islands in a counter-clockwise direction, and internal tides can generate local currents extending
over the entire water column which can affect the mooring causing hydrophone flow noise,
strumming and potentially fatigue. An acoustic study used a 3-D parabolic equation basin-scale
propagation model. This showed the importance of 3-D effects on global coverage estimation, and
made it possible to narrow down the selection of hydrophone locations and depths within the
constraints posed by the environment.
Using correlation matrices to identify temporal characteristics of ambient
noise
Stephen M. Nichols, David L. Bradley
Presenter: Stephen M. Nichols
In the deep ocean sound channel, biologic, anthropogenic and geophysical noise mechanisms
combine to produce a very dynamic ambient noise field. As a complement to traditional spectral
analysis techniques, correlation matrices can be used to characterize the ambient noise field by
identifying ambient source mechanisms in frequency bands in which the acoustic levels tend to
change together. Previous studies have investigated the frequency structure revealed by this
approach. This study examines the use of correlation matrices as a method of identifying temporal
characteristics of underwater ambient noise sources. The benefits and limitations of this technique
106
UA2014
will be demonstrated using ambient noise data recorded by the Comprehensive Nuclear-Test Ban
Treaty Organization’s (CTBTO) hydroacoustic monitoring system.
Antarctic’s Siren Call: The Sound of Icebergs
Haru Matsumoto, Robert.P. Dziak, Delwayne Bohnenstiehl, Jean Tournadre, Joe Haxel,
T-K Lau, Matt Fowler
Presenter: Haru Matsumoto
While the steady increase in global shipping traffic has been identified as a primary cause of rising
ocean noise level, in the southern hemisphere the disintegration of large icebergs was found to be
a significant noise source that influences the soundscape of ocean basins. Two large icebergs, B15a
and C19a, calved off the Ross Ice Shelf in early 2000s and drifted eastward to the warmer South
Pacific Ocean in late 2007. For the next 1.5 years, while these icebergs were rapidly melting, they
not only affected water circulation and marine ecosystem in their vicinity but also influenced the
low-frequency ambient noise level of the South Pacific Basin. From 2008 to early 2009, the
disintegration of B15a and C19a continuously projected loud low frequency sounds into the water
column. The sounds propagated efficiently to lower latitudes, influencing the soundscape of the
entire South Pacific basin. The icebergs’ sounds were recorded at Juan Fernandez Islands (34oS,
79oW) by deep-water hydrophones maintained by Comprehensive Nuclear-Test-Ban Treaty
Organization (CTBTO). The sounds also propagated across the equator (~10,000 km from icebergs)
and were recorded by a deep-water hydrophone at 8oN, 110oW maintained by Pacific Marine
Environmental Laboratory at National Oceanic Atmospheric Administration and Oregon State
University. In 30-36 Hz range, the noise level was ~6 dB and ~3 dB higher than baseline years at
respective location. Spectrogram shows that at CTBTO hydrophone, icebergs’ sounds dominated
frequency range below 100 Hz in which baleen whales vocalize. Some large icebergs in the Southern
Ocean have life spans over a decade. We discuss that icebergs calved off Antarctica can collectively
generate a considerable amount of sound energy, which then propagates across ocean basins,
influencing the ocean acoustic environment for the duration of the iceberg’s disintegration.
Evanescent wave coupling in a geophysical system: Airborne acoustic signals
from the Mw 8.1 Macquarie Ridge earthquake
Evers, L.G., D. Brown, K. Heaney, J.D. Assink, P.S.M. Smets, M. Snellen
Presenter: Läslo Evers
Atmospheric low frequency sound, i.e., infrasound, from underwater events has not been
considered thus far, due to the high impedance contrast of the water-air interface making it almost
fully reflective. Here, we report on atmospheric infrasound from a large underwater earthquake
(Mw 8.1) near the Macquarie Ridge, which was recorded at 1,325 km from the epicenter at IMS
station IS05. Seismic waves coupled to hydro-acoustic waves at the ocean floor, after which the
energy entered the SOund Fixing And Ranging (SOFAR) channel and was detected on a IMS
hydrophone array (H01W). The energy was diffracted by a sea mount and an oceanic ridge, which
acted as a secondary source, into the water column followed by coupling into the atmosphere. The
107
UA2014
latter results from evanescent wave coupling and the attendant anomalous transparency of the sea
surface for very low frequency acoustic waves. These findings have implications for the verification
of the CTBT, where underwater events might also be monitored based on their atmospheric acoustic
signature.
Basin Scale Time-Domain Modelling for CTBTO Tracking
The Peregrine 3D model has been applied to predicting coverage of the proposed International
Monitoring Station (IMS) in Crozet, as well as to address issues of the observation of hydro-acoustic
signals in the physical 2D acoustic shadow of seamounts and ridges (Heaney, Campbell, Snellen,
JASA 2013). In this paper, we apply the broadband Peregrine model to determine travel time-tables
from basin scale 3D acoustics. Two specific cases are investigated. The first is the computation of
acoustic arrival times for the Perth-Bermuda propagation, which was explained as 2 adiabatic
mode paths in 1993 (Heaney, Kuperman and McDonald, JASA 1993). The second example is the
generation of travel time-tables for hydro-acoustic signals from Ascension, looking southwest
through Drake Passage. The comparison of these 3D travel time tables with 2D, in-plane mode
one group speed travel time tables will be used to evaluate localization accuracy of the current
Comprehensive Test Ban Treaty Organization system.
108
UA2014
Session 11: Distributed Networked Systems for Surveillance
Organizer: Frank Ehlers and Arne Schulz
Adaptive Bayesian behaviors for AUV surveillance networks
Ryan Goldhahn, Paolo Braca, Kevin Lepage
Presenter: Ryan Goldhahn
Autonomous underwater vehicles (AUVs) present a low-cost alternative and/or supplement to
existing underwater surveillance networks. They may however be inferior to conventional assets in
speed, endurance, and sensor payload. Multi-sensor data fusion techniques and collaborative
autonomous behaviors are thus desired to fill the resulting performance gap. In this work, a rangedependent acoustic model for the predicted probability of target detection is combined with the
detections observed on all available platforms in a Bayesian framework to compute a posterior
distribution on target position for each ping. This posterior is then used to by the AUVs to
collectively optimize their future actions based on a mission-driven measure of network
performance. A Bernoulli filter is used to jointly estimate both the target state and whether zero
or one target is present. Simulation results are presented quantifying the performance increase
using these adaptive behaviors over traditional pre-planned trajectories. Real-data results are also
presented for adaptive behaviors running on two AUVs during the COLLAB13 trial conducted by the
NATO Centre for Maritime Research and Experimentation (CMRE). Work supported by NATO Allied
Command Transformation (ACT).
Problems in globally optimizing underwater surveillance networks in
communications limited environments
Kevin Lepage, Juri Sildam
The use of unmanned systems for underwater surveillance presents many challenges. For
multistatic ASW, these challenges are equally divided into the sensing challenges, typified by the
high rate of false alarms for a given detection threshold, and by the severely limited
communications environment, that generally challenges the ability to successfully pursue
collaborative strategies and to seek data fusion gain. In this talk the experience of CMRE deploying
its multistatic ASW autonomous security network demonstrator is described and the factors limiting
its performance are addressed. Strategies and promising research directions for improving the
performance of underwater surveillance networks are outlined.
109
UA2014
Parameter estimation for non-cooperative multistatic sonar
Martin Michaelis, Martina Brotje, Frank Ehlers
Presenter: Martin Michaelis
The main task of multistatic active sonar is the localization and tracking of objects of interest
(targets). Therefore, a precise knowledge of the position of the acoustic sources as well as of its
own sonar sensor position and heading is mandatory for each receiver in the multistatic active
sonar system. In this paper we consider a receiver consisting of an antenna array, which is towed
by an autonomous underwater vehicle (AUV). If the AUV is exploiting non-cooperative sources
neither the position of the source nor the time of a transmission is known in advance. Only rough
estimates can be gathered at the signal processing stage. In this case it is necessary to improve the
available knowledge over time. We assume here that the positions of some targets are known
according to a given uncertainty. These targets can be fixed objects like wrecks or small islands. In
comparison to solely processing the direct blast, additional information is obtained by evaluating
the reflections from known targets. In this paper we discuss the potential of using these reflections
for improving the estimates of the system parameters, in particular the positions of the acoustic
sources and the times of transmission. We present an algorithm for automatic parameter
estimation, which is based on the multihypothesis tracking (MHT) filter, and discuss results for
simulated data.
Optimal Area Coverage in Autonomous Sensor Networks
Mark R. Balthasar, Sara Al-Sayed, Stefan Leier, Abdelhak M. Zoubir
Presenter: Mark R. Balthasar
Autonomous sensor networks (ASNs) are a special kind of wireless sensor networks (WSNs) that
rely on collaborative in-network data processing instead of routing aggregated data to a fusion
center. Ideally, they are able to operate without human intervention. Hence, ASNs are particularly
useful in hazardous environments or areas that are inaccessible to humans.
Each ASN application starts with the problem of network formation. Instead of deploying the
network in a predefined layout, one should consider mobile sensor nodes that distribute themselves
autonomously in the region of interest (ROI), while avoiding collisions. Ultimately, the network
should reach a quiescent state in which blanket coverage of the ROI is attained.
In this paper, we consider two algorithms for autonomous network distribution with collision
avoidance, namely, Virtual Forces (VF) and Extended Virtual Spring Mesh (EVSM). VF considers the
motion of individual nodes as a result of repulsion forces from neighboring nodes as well as
obstacles, and an attraction force to neighbors in order to ensure swarm coherence. In addition, we
introduce an exposure force that pushes the nodes to blind spots such as the shadow regions of
obstacles. EVSM considers the network as a mesh of virtual springs that exert either an attractive
or repulsive force on the nodes. We improve the algorithm by adding a repulsion force for obstacle
avoidance. Our simulations show that a combination of VF and EVSM yields the best area coverage
in terms of speed and spread.
110
UA2014
Acoustic communication and localization in AUV cooperative surveys
Andrea Caiti, Francesco Di Corato, Davide Fenucci, Benedetto Allotta, Fabio Bartolini,
Riccardo Costanzi, Jonathan Gelli, Niccolo Monni, Marco Natalini, Luca Pugi,
Alessandro Ridolfi
Presenter: Andrea Caiti
The experimental results in acoustic communication and localization obtained with the “Typhoon”
Autonomous Underwater Vehicle (AUV) in the CommsNet13 field trial are presented. The
“Typhoon”s are a set of three AUVs developed by the Authors within the framework of the
“Thesaurus” project, funded by the Tuscany Region, aiming at developing techniques for systematic
cooperative autonomous exploration of marine archaeological areas by AUVs. The CommsNet13
experiment, which took place in September 2013 in the La Spezia Gulf, North Tyrrhenian Sea, was
organized and scientifically coordinated by the NATO S&T Org. Ctr. for Maritime Research and
Experimentation (CMRE); it included among its objectives the evaluation of on-board acoustic
Ultra-Short Base Line (USBL) systems for navigation and localization of AUVs. For the experimental
conditions encountered, the results obtained show that, by integrating on-board navigation
systems (Inertial measurement Units, Doppler Velocity Logs) with acoustic fixes the navigation error
is limited to ca. 20m worst case, and to ca. 10m for most of the time of the experimental run. With
our communication/localization system, designed for the team operation, latency was however a
non-negligible factor, as well as the packet loss % in networked communication. Acoustic channel
characterization, based on the environmental data measured at the experimental site, is also
presented, in order to separate the effect of the communication architecture from those of the
physical channel.
111
UA2014
Session 12: Experimental and modelling validation of target strength
measurements
Session 12a
representation
David Prowse
Presenter: David Prowse
Target Echo Strength (TES) is a fundamental component of the active sonar equation. Knowledge
of TES signatures is needed in order to understand the performance and design drivers of active
sonar detection systems, to develop and implement target detection and classification algorithms
and exploit TES signatures to either maximise target detection, or remain stealthy. The
requirements for target echo strength information, and its application to a range of underwater
functions are considered.
TES is often more than just a simple numeric value that provides a macroscopic representation of
target reflectivity; and its interpretation is driven by the particular application where it is to be used.
Various ways of representing TES (including Median Target Strength, Integration Target Strength,
Peak Target Strength and high fidelity time domain representations) are examined and the impact
on the requirements for a measurement system are considered.
Regardless of the various methods of representing TES signatures, the fundamental requirement
for the measurement is to ensure sufficient Signal to Noise Ratio such that the various TES
characteristics can be observed, isolated and measured so that their effect can be quantified,
investigated and understood. Factors that must be addressed to achieve this are discussed,
including:
- Noise & Reverberation control;
- Calibration;
- Experimental control;
- Understanding of the environment;
- Errors and confidence limits;
- Appreciation of application;
- Body of evidence.
These factors are considered and measurement options, including configurable 2D measurement
arrays and advanced processing, are proposed to enable reliable TES measurements.
ATLAS ELEKTRONIK UK are at the forefront of the research, development and supply of underwater
technology and have a deep understanding of how TES affects design, performance and operation
of active sonar systems.
112
UA2014
Model Tank Measurements and Using a Random Noise Feld to Determine the
Scattering Properties of an Object
Sandrine Rakotonarivo, Selda Yildiz, Philippe Roux, Earl Williams, W. A. Kuperman
Presenter: W. A. Kuperman
We present hydrophone measurements taken at the periphery of a model tank to demonstrate that
the ambient noise field in the model tank can be transformed into coherent radiation suitable for
probing the tank. This suggests that one might determine the structural properties of an object in
an analogous way. In particular, we derive a method to estimate the structural or surface
impedance matrix (or equivalently the inverse of the structural Green’s function) for an elastic body
by placing it in an encompassing, spatially random noise field and cross-correlating pressure and
normal-velocity measurements taken on its surface. A numerical experiment is then presented
demonstrating that the scattered field obtained from this noise correlation based structural
impedance agrees with standard scattering theory.
On the design and construction of drifting-mine test targets for sonar, radar
and electro-optical detection experiments
Henry Dol
Presenter: Henry Dol
The timely detection of small hazardous objects at the sea surface, such as drifting mines, is
challenging for ship-mounted sensor systems, both for underwater sensor systems like sonar and
above-water sensor systems like radar and electro-optics (lidar, infrared/visual cameras). This is
due to the low target echo strength and radar/lidar cross-section of partly submerged objects at
small grazing angles, which are also intermittently shielded by waves and their response is being
hidden in significant surface reverberation. In 2009-2010, the feasibility of ship-based detection by
state-of-the-art sensor technology was successfully assessed using specially designed drifting-mine
test targets. In this paper, we look at the test target requirements for different observation
technologies, and focus on the design and construction of the target objects. For illustration of the
suitability of the test targets and available sensor technology, some sea-trial results are included
for visual and IR detection above water, and for sonar detection under water.
Open water target strength measurements and validation
Keith Page, Charles Taylor, Ben Hodder
Presenter: Keith Page, Charles Taylor, Ben Hodder
Target strength measurements of structures larger than about 1m, particularly those much larger,
can be difficult to make in a tank environment. This is due to a number of constraints associated
with the object size; including far-field distances and reverberation. Therefore techniques are
required which allow the measurements to be made in open water. Measurements of underwater
acoustic target strength made in this environment have both advantages and disadvantages when
compared with tank measurements. The benefits of an increased volume of water include allowing
113
UA2014
large structures to be assessed with reduced reverberation, whilst the shortcomings are related to
deployment challenges, control of the experiment and issues with understanding the sound field.
An overview of the benefits and problems of various methods of open water target strength
measurement are discussed and outlined. Specifically addressing how to confirm or corroborate
the target strength obtained, by careful experimental design, monitoring of equipment and sound
fields, and making use of a number of TES measurement techniques.
114
UA2014
Session 12b
Finite element modeling of acoustic radiation force for elastic objects
Ahmad T. Abawi
The study of the acoustic radiation force produced by acoustic beams has been the topic of active
research in the last few years mainly due to its ability for particle trapping and non-contact
manipulations. However, efforts in modeling the radiation force have mainly focused on analytic
solutions and thus have been limited to spherical objects, where the acoustic radiation force has
been calculated for a plane wave as well as various types of beams. But up until recently, even
those efforts were limited to using on-axis beams, where the incident beam is along an axis that
goes through the center of the sphere and thus reduces the problem to an axially symmetric one.
In this work we use the finite element technique to compute the acoustic radiation force for an
arbitrary elastic object and for an arbitrary incident beam. One of the main objectives of this work
is to understand and interpret the data we collected at a recent experiment we conducted at the
Washington State University test tank facility, where a 2.5-inch PMMA sphere was ensonified by
an intense amplitude-modulated ultrasound beam focused at its surface.
Improved modeling accuracy of the Elastic Object Response by Inclusion of
Higher Order Re-Scattering
Marten .J.J. Nijhof, Aubrey L. Espana, Kevin L. Williams
Presenter: Marten Nijhof
Current methods of automatic target detection and classification using low- to mid- frequency
sonar are based on correlating the measured Target In the Environment Response (TIER) of
potential targets with the TIERs in a database of known objects. Obtaining experimental TIER
measurements of proud, partially and fully buried objects to populate the database is an expensive
undertaking. Using numerical predictions of the TIER instead of experimentally gathered data is a
potential efficient alternative. In addition, a numerical model also offers insights into the physics
involved as an added advantage.
Previously, an efficient numerical hybrid method based on Finite Element and Helmholtz Kirchhoff
Integral (HKI) models was developed for prediction of the TIER of arbitrarily positioned axially
symmetric objects. This hybrid model was successfully tested and validated for a number of cases.
However, the main limitations of this hybrid method (besides being restricted to axially symmetric
targets) are the neglection of higher order re-scattering between the target and the sediment, and
the inability to account for discontinuities or changes in the sound speed and density of the medium
surrounding the target surface (unless such changes are perpendicular to the axis of symmetry).
For certain target/environment combinations, comparison of measurement and model results
suggest that higher order re-scattering contributes significantly to some features of the acoustic
115
UA2014
response. The hybrid model was extended to include higher order re-scattering by successive
iterations of running the original model. As part of the approach a novel non-singular HKI method
was implemented and successfully extended to work with non-homogenous media. The resulting
hybrid model that includes higher order re-scattering was validated against the results of an
experimental test-setup for which higher order re-scattering effects are shown to be relevant.
Acoustic scattering from partially buried cylinders: Measurement validation
and interpretation using physical acoustics and finite element models
Aubrey L. Espana, Marten J. J. Nijhof, Kevin L. Williams, Daniel S. Plotnick, Philip L.
Marston
Presenter: Aubrey L. Espana
When attempting to detect and classify objects in an ocean environment, it has been shown that
the surrounding environment, specifically the target’s deployment within that environment, greatly
affects the measured acoustic response. While sea trials go a long ways towards extending the
development of better classification systems, they are costly and ultimately only investigate the
target at a handful of burial depths and ranges. There exists a need for robust models that can
predict the target response with the required level of fidelity to result in positive classification
outcomes. To accomplish this, however, reliable measurements and efficient models go hand-inhand, one validating the other or vice versa. A specific example of this is presented here by
examining the acoustic scattering from a solid cylinder in an ocean environment, partially buried
with its axis at an oblique angle relative to the sand sediment. Plots of the measured target strength
versus frequency and aspect angle reveal a number of interesting acoustic phenomena, some of
which are explained using physical acoustics models. To help validate these measurements and
offer further incite into the physical mechanisms involved, results from a numerical hybrid model
are presented. This model, based on Finite Element and Helmholtz Kirchhoff Integral methods, has
previously shown success in predicting the response of proud and partially buried targets with axes
parallel to the sand sediment. To help validate the use of the hybrid model in this non-symmetric
environment, a controlled tank experiment was conducted using a scaled version of the solid
cylinder, mounted next to an air-water interface. The comparison between the hybrid model and
tank measurements exposes potential areas where higher order re-scattering effects may become
important, and is the topic of the following talk by M. J. J. Nijhof.
Performance Modelling and Experimental Validation for a SedimentPenetrating SAS
A.J. Hunter, I. Mulders, M.J.J. Nijhof, B.A.J. Quesson, A.L.D. Beckers
We present a hybrid acoustic model for evaluating the performance of a broadband low frequency
sub-sediment imaging sonar and outline plans for its experimental validation. The hybrid model is
comprised of the OASES propagation model, a stochastic reverberation model based on Lambertian
scattering, and various elastic target scattering models including an analytical model for spherical
116
UA2014
shells and a finite-element model for axisymmetric objects. Processing gains are incorporated by
running simulated broadband data realisations through the processing chain of TNO’s MUD sonar,
which includes matched filtering and synthetic aperture sonar (SAS) processing. We show example
simulated results and attempt to predict the performance of our new MUD-2 sonar for future
validation in upcoming sea trials.
An efficient numerical target strength prediction model: validation against
analytic solutions
Laurent Fillinger, Marten J.J. Nijhof, Christ A.F. De Jong
Presenter: Laurent Fillinger, Marten J.J. Nijhof
A decade ago, TNO developed RASP (Rapid Acoustic Signature Prediction), a numerical model for
the prediction of the target strength of immersed underwater objects. The model is based on
Kirchhoff diffraction theory. It is currently being improved to model refraction, angle dependent
reflection and transmission (possibly through anisotropic materials) and multiple reflections and
transmissions. These improvements are validated against available analytic solutions for simple
shapes such as cylinders and spheres. Validation efforts will also be conducted against other
numerical models using complex shapes such as a model submarine [in the second Benchmarking
of Target Strength Simulation (BeTSSi II) workshop that will be held in September 2014]. The target
strength model and a comparison of the numerical results with analytic solutions are presented. In
addition, the effect and relative importance of the various improvements are evaluated.
117
UA2014
Session 13: Habitat Mapping: Procedures and Results
Session 13a
Location: Lecture Room B, Thursday 26th June 2014, 14:15 - 16:15
Towards Joint Use Of Side Scan Sonar And Sub-Bottom Profiler Data For The
Automatic Quantification Of Marine Habitats. Case Study: Lourdas Gulf,
Kefalonia Isl., Greece.
Fakiris Elias, Zoura Despina, Ferentinos George, Papatheodorou George
Presenter: Fakiris Elias
Numerous surveys have been and are still being performed by simultaneously using conventional
Side-Scan Sonars (SSSs) and Sub-Bottom Profilers (SBPs). This is in respect of gaining fast, wide
scale, three dimensional mappings of the seafloor and its substrate, extracting maximum value
from a single, time limited, survey. The combination of these two systems offer good knowledge of
both the stratigraphy and the bio-habitats of the seabed, aspects often linked to each other, since
substrate characteristics can, in some cases, be indicative for the flourishing of certain species.
However, a basic drawback is the inconsistency of mapping scales between the two systems, while
SSS produces high resolution backscatter maps (of much higher density than MBES ones) but SBP
produces substrata (and bathymetric) information of high vertical but very low horizontal density.
Although Multi-beam Echo Sounder tends to be the preferred tool nowadays towards gaining
knowledge about the seafloor, numerous surveys have been and are still being performed by
simultaneously using conventional Side-Scan Sonars (SSSs) and Sub-Bottom Profilers (SBPs). This is
in respect of gaining fast, wide scale, three dimensional imaging of the seafloor and its substrate,
extracting maximum value from a single and time limited survey. The combination of these two
systems offer good knowledge of both the stratigraphy and the habitats of the seabed, aspects
often linked to each other. However, a basic drawback is the inconsistency between their mapping
scales, while SSS produces high resolution backscatter maps (of much higher density than MBES
ones) while SBP produces substrata information of high vertical but very low horizontal density.
In this work, 100 kHz SSS and 3.5 kHz SBP data, collected simultaneously during a geophysical
survey at Lourdas gulf, Kefalonia Island, Greece, underwent post-processing and analysis, to extract
numerous statistical features from both the seafloor and its substrate, towards automatic seafloor
classification. The SSS records were mosaicked using Geocoder and the mosaic image was subjected
to textural analysis, using the SonarClass software tool, to extract a large number of features.
Unsupervised classification of the SSS features leaded to an accurate segmentation of the seafloor
into homogenous regions. The SBP images were processed using multi-scale elongated steerable
filters, to detect all seismic reflectors, and numerous features were extracted regarding the acoustic
transparency and density of the seismic reflectors (layering) as well as the rugosity of the seabed.
Supervised classification of the SBP features exhibited their high ability to discriminate between
118
UA2014
different known sea bed types, making them suitable for use as a substitute of traditional ground
discrimination systems. The combination of the SBP track classes to the SSS segmentation, leaded
to a full coverage - high detailed classification map of the study area.
Automatic classification of bedforms using phase differencing bathymetric
sonar
Diego Porpilho, Antonio Henrique Da Fontoura Klein, Rafael S. V. De Camargo, Michel
Franco Volpato Prado, Jarbas Bonetti, Andrew Short, Elias Fakiris.
Presenter: Diego Porpilho
This paper classifies bedforms and habitat using side scan images from a phase differencing
bathymetric sonar. The study area is the inner shelf, between 3 and 15 m depth, of Barra da Lagoa
– Mocambique beaches located on the northeast of Santa Catarina’s island, Brazil. The data was
collected with an EdgeTech® 4600 540 KHz interferometric system (phase differencing bathymetric
sonar) which outputs side scan sonar images and swath bathymetry, proving images that are 3 and
4 times the width of the water depth. The data covered an area of approximately 12 km² and was
collected using the softwares Hypack®2013 and Discover®, and processed with SonarWiz5® and
SonarClass® for side scan and Hypack® for bathymetry. The preliminary results show an inner shelf
dominated by finer sediments, but containing 0.5 to 0.7m lower elevation patches of coarse grain
rippled sediments, validated comparing automatic and manual classification of the images on the
SonarClass®, which uses textural parameters. The different bottom types were classified using
SonarClass® and also validated with ground-truthing station besides the bedforms in accordance
of literature classification.
Diurnal variation in sediment backscattering properties caused by
photosynthesis of microphytobenthos (Southern Baltic Sea)
Natalia Gorska, Ewa Kowalska-Duda, Filip Pniewski, Adam Latala, Jacek Marszal, Jan
Schmidt
Presenter: Ewa Kowalska-Duda
In approximately one third of the continental shelf area light reaching the sea bed is sufficient for
microphytobenthos photosynthesis. This process is responsible for oxygen supersaturation in
sediment pore water and formation of numerous bubbles, which can change the marine sediment
backscattering properties. The understanding of the impact of microphytobenthos photosynthesis
on the characteristics of the signals backscattered by the Baltic marine sediments is crucial in the
development of the hydroacoustical benthic habitat classification techniques in the Baltic Sea.
Therefore, it was our goal to characterize variations in the sediment backscattering properties
induced by microphytobenthos photosynthesis and altered by the physical (e.g., variation of
temperature and light conditions) and biological (e.g., benthos microalgal biomass or benthos
macrofauna activity) environmental factors.
The study included five multiday interdisciplinary laboratory experiments. Changing the light
conditions (L:D cycle) diurnal variability of the signal backscattered by the sandy sediments was
119
UA2014
analyzed. Hydroacoustical and oxygen concentration measurements were carried out under
controlled light, temperature and salinity conditions. Microphytobenthos photosynthesis was
monitored by measuring PAM fluorescence. In order to understand the influence of chosen
environmental variables on the seabed backscattering diurnal variations, caused by
microphytobenthos photosynthesis, each experiment differed in one key aspect.
The study demonstrated that the diurnal variation in the backscattering at sandy sediments (echo
energy was larger during daytime than at night) can be controlled by benthic microalgal
photosynthesis. The intensity of photosynthetically active radiation and the microphytobenthos
biomass should be considered as important factors governing the impact of photosynthesis on the
backscattering. It was shown that the effect of the macrozoobenthos activity (bioturbation) can be
more significant than the impact of the intensive photosynthesis process: in case of their
simultaneous performance the first process can masks the second.
Do fish and blue algae blooms coexist in space?
Malgorzata Godlewska, Katarzyna Izydorczyk, Shaowen Ye, Bronislaw Dlugoszewski,
Marta Cendrowska
Cyanobacterial blooms are recognized nowadays as a worldwide significant water quality problem,
particularly characteristic of shallow eutrophic lakes and reservoirs. It seems that the probability
for blooms of cyanobacteria will increase in the next decades due to the possible consequences of
climate change on freshwater ecosystems. Toxins produced by cyanobacteria threaten not only
aquatic animals, including fish in which they may accumulate, but also people, who may be exposed
to cyanotoxins both directly, by drinking water and indirectly, by consumption of contaminated fish.
Measurements of fish and blue algae distributions were performed simultaneously in shallow
drinking water Sulejow reservoir, Poland, to check for their coexistence in space and time. EK 60
Simrad echosounder with 200 kHz transducer looking horizontally, and 70 kHz transducer looking
vertically were used for recording fish, while phycocyanin fluorescence on-line detection method
was used to estimate cyanobacteria biomass. Maps of fish and cyanobacteria distributions sampled
every 20 sec were produced using GIS techniques. When looking at average values (by transect) for
fish abundance and cyanobacteria biomass the trends are opposite and it seems that fish can
actively avoid harmful algae by migrating to places, where their concentrations are smaller.
However, analysis of the detailed distribution maps shows that concentrations of fish and algae
sometime do match in space and time, posing the danger for fish to become contaminated.
Acoustic mapping of submerged macrophytes in selected lakes of the
Drawieński National Park
Marta Cendrowska , Wojciech Puchalski, Mikolaj Adamczyk, Malgorzata Godlewska
Submerged macrophytes are among the key elements in freshwater ecosystems, and they are often
considered good indicators of ecological status and water quality. However, traditional methods of
120
UA2014
their investigation are destructive, tedious, costly and time consuming. Hydroacoustic
measurements of submerged macrophytes were performed in 4 lakes of the Drawienski National
Park, NW Poland, using split beam echosounder Simrad EK60 with 200 kHz transducer. Data were
analysed with Sonar 5 Pro, macrophytes module. Maps of macrophyte cover and height were
produced for the lakes using GIS techniques. Although all lakes were clear-water ones, providing
good conditions for plant growth, the macrophyte density, their occurrence depths and health
condition differed greatly between the lakes. Simultaneously with acoustic survey, macrophyte
grab samples (using a rake attached to rope) were taken, which allowed for classification of
vegetation assemblies. Charophytes (different species: Chara tomentosa, Ch. filiformis, Ch. aspera,
Nitella opaca, Nitellopsis obtusa) were easily distinguishable hydroacoustically from other plants,
as Eurasian watermilfoil (Myriophyllum spicatum) or pondweeds (Potamogeton sp.). Since they
were also occupying different depths, they can be separated and assessed independently from
other types of vegetation. Many Charophyte species are considered as biological indicators of water
quality, thus hydroacoustic monitoring of them can be used to assess ecological status of inland
waters as required by the European Water Framework Directive, and to declare Special Areas of
Conservation under the Habitats Directive. Such rapid and non-invasive monitoring of submerged
vegetation becomes particularly important when facing huge differences in the community
composition and plant cover due to climate change and interannual fluctuations.
High-resolution multibeam mapping of habitats in the extremely shallow
waters of the Venice Lagoon
F. Foglini, F. Madricardo, P. Blondel, A. Kruss, M. Sigovini
Presenter: Federica Foglini
Recent advances in multibeam technology have opened new possibilities in the field of mapping
bottom morphologies, substrates and habitats. In particular, the substrate mapping by mean of
backscatter and water column multibeam data is relatively new in extremely shallow environments,
where multibeam systems have rarely been used.
In this study we present the first results of an extensive survey carried out in the very shallow Venice
lagoon channels using a Kongsberg EM-2040 DC multibeam system during 2013. Although
numerous biological studies have been carried out to study the habitats in the tidal flats and salt
marshes of the Venice lagoon, the channel habitats are still almost unexplored.
The new possible applications in habitat mapping given by the analysis of high-resolution
bathymetry, backscatter and water column data will be discussed. In particular, we will show the
first results of a survey carried out in a natural tidal channel of the northern part of the Venice
Lagoon (water depths from 15 m to less than 1 m). On the high resolution backscatter data
collected, we carried out a two-dimensional (2D) textural analysis with the TexAn software. We
then performed an unsupervised classification of the backscatter data. As a result, we were able to
identify different backscatter areas where several grab samples were collected for ground truthing.
With the help of this sampling we calibrated the textural analyses and obtained a classification of
the different kinds of substrate. The characterization of the substrate was then tested with a set of
121
UA2014
bottom photographs. In addition, the first results of combined bathymetric, backscatter and water
column data analysis for bottom vegetation detection in the shallow lagoonal channels will be also
shown.
122
UA2014
Session 13b
Development of Acoustic Colour technique using multi-frequency swath
acoustic backscatter
Garret Duffy, Yuri Rzhanov, Eli Rinde, Berit Horvei, Colin Brown
Presenter: To Be Decided
Swath acoustic data collected by multibeam or sidescan sonar systems has long been a recognised
tool to efficiently map the bathymetry and texture of large areas of seabed. Backscatter data may
be spatially classified based on either characteristic acoustic angular response measurements or
inverted backscatter model parameters that depend on seabed physical properties. However, the
inverted model parameter of roughness is intrinsically linked to the acoustic wavelength; only the
roughness spectrum with wavelength less than half of the acoustic wavelength affects the surficial
acoustic backscatter. Therefore, inferred surficial texture is intrinsically acoustic frequencydependent. This research aims to test the feasibility of using multi-frequency acoustic backscatter
to create acoustic classes based on a broadband acoustic response to natural roughness spectra
and therefore more accurate spatial delineation of facies.
On 18th and 19th September 2013, simultaneously-acquired EM2040D (300 kHz (CW pulse)),
EM710 (100 kHz (CW)) and EM302 (30 kHz (CW)) benthic acoustic data were collected using
Kongsberg Maritime’s test vessel, Simrad Echo, within the Ormo–F?rder Marine Protected Area in
Oslofjorden, Norway. 47 line-km of multi-frequency data were collected resulting in a survey
extending 2.9 km by 3.2 km. In addition to the latter overlapping survey lines, 5 non-overlapping,
2.9 km-long cross-lines were carried out to investigate anisotropic backscatter effects; the crosslines were run perpendicular to bathymetric contours to minimise variation of possible depthdependent benthic texture within a swath. These three datasets have been independently
automatically processed producing facies’ maps with delineation on a pixel level. This paper
discusses results of delineations and compares maps for datasets collected with different
frequencies.
MARTA: an AUV for Underwater Cultural Heritage
B. Allotta, F. Bartolini, R. Conti, R. Costanzi, J. Gelli, N. Monni, M. Natalini, L. Pugi, A.
Ridolfi
Presenter: Benedetto Allotta
MARTA, acronym for MArine Robotic Tool for Archaeology is a small-sized AUV developed in the
framework of the ARROWS project. The ARROWS project (start Sep 2011, end Aug 2015) is funded
by the European Commission in the framework of the FP7 call ENV-2012, challenge 6.2-6,
devoted to Development of advanced technologies and tools for mapping, diagnosing, excavating,
and securing underwater and coastal archaeological sites. MARTA will operate in an heterogeneous
123
UA2014
team of vehicles with a common mission to perform and a distributed and shared world model
updated based on non-synchronous information collected by each of the vehicles of the team. Each
of the vehicles will be equipped with acoustic communication means in order to be able to
communicate when submerged. The University of Florence is in charge for the design and
construction of the MARTA AUV, according to specifications written in compliance with the
requirements formulated by the Archaelogical Advisory Group, including archaeologists both from
inside and outside the ARROWS consortium. MARTA will be moderate in cost, with respect to
commercially available AUVs, and light enough in order to be deployed by two people from a small
boat. MARTA will operate at a maximum depth of 150m and, in addition to a pair of acoustic
modems for interveicular communication and USBL localization, it features two different payloads
i.e.: A pair or synchronised digital TV cameras with visible light as well as structured light (blue
laser) illuminators;
A Multibeam echosounder.
The paper will illustrate the vehicle design and the use of on-board acoustic instrumentation for
communication, localization, and sea-bottom imaging. Preliminary experimental data from the
field will be also presented.
High-resolution multibeam bathymetry applied to underwater research: a case
study from the Lagoon of Venice
F. Madricardo, F. Foglini, A. Kruss
Presenter: F. Madricardo
The Venice Lagoon that surrounds the historical city of Venice has been the site of human
settlements since Roman times, containing innumerable archaeological remains throughout it. The
shallowness of the lagoon has for a long time prevented the use of underwater acoustics that, in
general, can help to extensively and efficiently explore the bottom and sub-bottom for new
archaeological discoveries. However, the recent technological development of the multibeam
systems enables them to achieve very high performances also in very shallow waters.
In this study, we show the first results of a six month long multibeam survey that has been carried
out by means of a Kongsberg EM2040 DC multibeam system in the Venice Lagoon. All the channels
of the lagoon and the Canal Grande in the city of Venice were surveyed. The Digital Terrain Maps
(DTM) obtained from the multibeam data reach a resolution up to 5 cm giving a new insight in the
underwater archaeological research in the area.
124
UA2014
Session 14: Innovative Approaches for Characterizing Ocean Bottom Properties
Organizer: Martin Siderius, Sergio Jesus, Peter Nielsen, Jean-Pierre Hermand and Ross
Chapman
Location: Lecture Room A, Monday 23rd June 2014, 10:45 - 12:45
Experimental studies on passive bottom loss estimation from a compact array
mounted on an autonomous underwater vehicle
Lanfranco Muzi, Martin Siderius, Peter Nielsen
Presenter: Lanfranco Muzi
The seabed bottom loss (BL) is an important quantity for predicting transmission loss in the ocean.
A recent passive technique for BL estimation as a function of frequency and grazing angle exploits
marine ambient noise as an acoustic source. Conventional beamforming of the noise field by a
vertical line array of hydrophones is a fundamental step in this technique, and the beamformer
resolution in grazing angle affects the quality of the estimated BL. The technique has so far been
applied successfully by using moored or drifting arrays with a length of several meters. However,
nowadays a much simpler and cost-effective bottom-survey system can be conceived: Compact,
sub-meter-length arrays and a low-power-consumption data-acquisition payload can be mounted
on autonomous underwater vehicles (AUV), which could then map seabed reflection properties by
sampling the ambient noise field as they move under the surface. The GLASS 12 and GLASS 13
experiments represent the first attempt to investigate the performance of such compact arrays for
BL estimation. In this presentation, data collected by the AUV-mounted, 5-element, 0.4m long array
are processed to compute the BL, which is then used with an inversion method to estimate the
geoacoustic properties of the bottom. Rather than accurate determination of the seabed
parameters, the study aims at assessing the sensitivity of this array to different bottom types, as
well as illustrating the specific challenges posed by the system. Results from the two sites will be
presented along with those using alternative methods (e.g., core data) for comparison.
Estimating seabed scattering mechanisms via Bayesian model selection
Gavin Steininger, Stan Dosso, Charles Holland, Jan Detmer
Presenter: Gavin Steininger
This paper presents a quantitative and objective approach to classify the dominant seabed acoustic
scattering mechanism(s) and estimate seabed scattering and geoacoustic parameters from
measured backscatter data. The classification system is based on trans-dimensional sampling and
the deviance information criterion. Scattering is modeled using first-order perturbation theory as
due to one of three mechanisms: interface scattering from a rough seafloor, volume scattering from
a heterogeneous sediment layer, or mixed scattering combining both interface and volume
scattering. The classification system is applied to six simulated test cases where it correctly
identifies the true dominant scattering mechanism as having greater support from the data in five
cases; the remaining case is indecisive. The approach is also applied to measured backscatter data
125
UA2014
from the Malta Plateau where volume scattering is determined as the dominant scattering
mechanism. This conclusion and the scattering/geoacoustic parameters estimated in the inversion
are consistent with properties from previous inversions and with core measurements from the site.
[Work supported by ONR.]
Attenuation of sound in mud sediments
Allan D. Pierce, William L. Siegmann, Joseph O. Fayton
Presenter: Allan D. Pierce
A physical mechanism is proposed for the attenuation of sound in mud sediments. Mud is
envisioned as being largely made up of clay platelets, most of whom carry a net negative charge
because of isomorphous substitution. The platelets tend to repel each other, and a tendency for
gravitational settling is by itself unable to cause the formation of a sediment in water, and a
colloidal suspension would, at first sight seem more likely. However, short range chemical
(quantum mechanical) forces can cause edges of individual platelets to bond with the faces of
other platelets. This is responsible for the card-house structure of mud and the fact that the
porosity of mud is of the order of 90%. The presence of such bonds, nevertheless, is subject to
the usual rules of statistical thermodynamics, and bonds are continuously being broken and
reformed. At any given given temperature, there is a set fraction of platelets that are bonded
on a given edge to the face of another platelet, and this fraction is in accord with a Boltzmann
distribution. A sound wave carries temperature and pressure fluctuations which disturb this
equilibrium distribution. When a bond is reformed, there is a new energy release that does not
go into the sound wave, so there is a net loss of energy in the sound wave. The proposed
mechanism is similar to the relaxation effects that account for the attenuation of sound in air
(vibrational relaxation of oxygen and nitrogen) and in sea water (chemical relaxation involving
magnesium sulfate)
Recursive Bayesian synthetic aperture geoacoustic inversion in the presence of
motion dynamics
Bien Aik Tan, Peter Gerstoft, Caglar Yardim, William S. Hodgkiss
A low signal to noise ratio (SNR), single source/receiver, broadband, frequency-coherent matchedfield inversion procedure recently has been proposed. It exploits coherently repeated transmissions
to improve estimation of the geoacoustic parameters. The long observation time improves the SNR
and creates a synthetic aperture due to relative source-receiver motion. To model constant velocity
source/receiver horizontal motion, waveguide Doppler theory for normal modes is necessary. However, the inversion performance degrades when source/receiver acceleration exists. Furthermore,
processing a train of pulses all-at-once does not take advantage of the natural incremental acquisition of new data along with the ability to assess the temporal evolution of parameter uncertainty.
Here, a recursive Bayesian estimation approach is developed that coherently processes the data
pulse-by-pulse and incrementally updates estimates of parameter uncertainty. It also approximates
126
UA2014
source/receiver acceleration by assuming piecewise constant but linearly changing source/receiver
velocities. When the source/receiver acceleration exists, it is demonstrated that modeling acceleration is critical for correct inversion. The method is demonstrated in simulation and real data
analysis of low SNR, 100–900 Hz LFM pulses from the Shallow Water 2006 experiment.
Bayesian geoacoustic inversion of airgun modal dispersion using a single
hydrophone in the Chukchi Sea
Graham Warner, Stan Dosso, David Hannay, Jan Dettmer
Presenter: Graham Warner
This paper presents results of a trans-dimensional Bayesian geoacoustic inversion algorithm
applied to the dispersive properties of water-column modes in a shallow water waveguide. Mode
arrival times at a single hydrophone were simulated and measured for multiple impulsive acoustic
sources with unknown pulse times and uncertain source-receiver ranges. Measurements were
collected using an ocean bottom hydrophone for a ship-towed airgun array during a shallow
hazards seismic survey in the Chukchi Sea, Alaska. A warping dispersion analysis was used to
determine mode arrival times for source-receiver ranges of 3 to 20 km. Frequency-dependent
arrival times were inverted for the number of fluid sediment layers, geoacoustic parameters (layer
thickness, sound speed, and density), and water sound-speed profile. Trans-dimensional Bayesian
sampling is performed using a reversible-jump Markov chain Monte Carlo algorithm with parallel
tempering to improve sampling efficiency over the unknown number of sediment layers. Marginal
posterior distributions indicate parameter uncertainties and quantify the resolving power of the
data. Dispersion inversion results are compared to trans-dimensional inversion of received level
versus frequency and range, and to an analytic solution for the seafloor sound speed from
headwave arrival times at close ranges.
Nonlinear inversion of ship radiated noise in shallow water
Steven E. Crocker, Peter L. Nielsen, James H. Miller, Martin Siderius
The Centre for Maritime Research and Experimentation conducted the Glider Acoustic Sensing of
Sediments experiment in the Mediterranean Sea in the summer of 2012. Among the objectives of
the sea trial was to employ a new autonomous underwater vehicle to collect acoustic data to invert
for geoacoustic properties of the seafloor sediments. Acoustic data were collected with the vehicle
bottom moored in 18 meters of water while the R/V Alliance made a close pass at five meters per
second. Data from a single hydrophone channel were inverted for the compression wave speed and
attenuation in the seafloor sediments. The inversion process operated on broadband received levels
recorded during the pass-by of the Alliance and the ambient noise spectrum while the Alliance
loitered four kilometres distant. Parameter estimates were generated using an evolutionary
algorithm to minimize an objective function that was developed for this study and a forward model
based on acoustic normal modes. The model vectors generated during the optimization were postprocessed using a maximum-likelihood approach to arrive at the parameter estimates and
127
UA2014
uncertainties. Good agreement was found between the compression wave speed estimated by the
inversion and measurements performed on two geophysical cores taken in close proximity to the
bottom moored vehicle. The broadband source spectrum of the Alliance was also estimated across
more than six octaves of frequency.
128
UA2014
Session 15: Modeling Sonar Performance in Uncertain Environments
Organizer: Georgios Haralabus and Chris Strode
Location: Lecture Room A, Friday 27th June 2014, 8:30 - 10:30
Mitigation Methods and Techniques for Enhancing Sonar Operational
Confidence
Mark Gammon, Marcel Lefrancois, Chris Strode
Presenter: Chris Strode
There are numerous challenges to effectively model sonar performance given the variability in the
ocean acoustic parameters, and the uncertainties with respect to factors such as target strength.
As such, in most cases it is well recognized by the operational community that it is nearly impossible
to accurately predict the performance of a sonar system for the full range of operational settings.
Given this situation, numerous initiatives both technical and operational are required to
harmoniously provide a level of confidence in sonar prediction under practical conditions. Some
initiatives and tools that are being utilized include the NATO Multi-Static Tactical Planning Aid
(MSTPA) and the Environmental Modelling Manager (EMM). The latter is embedded in a
developmental version of the Canadian Maritime Acoustic Processing System (MAPS). Other
initiatives and procedural guidelines will be presented with the caveat that some may have a
greater or lesser impact on operator confidence under realistic operational conditions.
Modelling multistatic sonobuoy fields in uncertain environments
Andrew Holden
Presenter: Andrew Holden
A multistatic field consisting of active and passive sensors can be deployed with the aim of detecting
submerged platforms transiting through the area. The ODIN engagement tool has recently been
extended with a range independent bistatic reverberation model to allow multi-static scenarios to
be modelled. This study uses the ODIN model to look at one possible small sensor field which is
tested to find paths where the platform remains undetected. In practice, knowledge of the
environment is poor. This study looks at how a design set for one sound speed profile performs for
a small number of other sound speed profiles.
Propagation of Acoustic Waves through a Spatially Fluctuating Medium:
Theoretical Study of the Physical Phenomena.
G. Real, X. Cristol , J.-P. Sessarego, D. Fattaccioli
The authors focus on the effects of phenomena, such as linear internal waves, on underwater
acoustic propagation. Linear internal waves are responsible for fluctuations of the depthdependent sound speed profile and, hence, induce distortions of the resulting acoustic pressure field
and degradation of the associated sonar performances. The main goal of this study is to develop a
129
UA2014
scaled experiment configuration able to provide some results representative of this kind of
distortions.
To do so, a theoretical study of the phenomenon has been carried out: we extended the existing
governing equations (Tatarskii) from 2D to 3D and obtained an expression for the standard
parabolic equation applied to the Fourier transform of the moments of order 2 and 4. We were also
able to express analytically the number of eigenrays linking source to receiver throughout a
fluctuating medium.
Various simulation programs were developed and used for the following purposes:
1) Validating or discarding some relationships given by Flatte through his classical dimensionless
analysis (ΛΦ plane) ;
2) Tracing rays through an acoustic lens featuring a plane face and a randomly rough face to
model refraction of the rays and to evaluate parameters like average the mean number of eigenrays
and the phase difference between extreme micro paths that are involved in the definitions of
Flatte’s dimensionless numbers;
3) Propagating an acoustic wave through the same object in order to anticipate for the shape of
the distorted pressure field, including diffraction effects.
Acoustic lenses featuring these shapes were otherwise realized and tested experimentally, as
reported in a companion paper (session “Tank Experiments”).
We were able both theoretically and experimentally to induce acoustic scattering that mimics, at
reduced scale and frequencies around 1MHz, the vertical and horizontal correlation properties and
the corresponding array performance that would be observed by sensor arrays at sea at the order
of 1kHz, after propagation through a linear internal wave field, or reflection on a rough sea surface.
Passive Sonar Performance Characterization and Transmission Loss
Measurement Using a Calibrated Mobile Acoustic Source
V. Premus, P. Abbot, M. Helfrick, T. Paluszkiewicz
Presenter: V. Premus
A system and methodology for rigorously measuring the performance of a passive sonar detection,
classification, and localization (DCL) system using a calibrated mobile acoustic source is presented.
The source, an unmanned undersea vehicle (UUV) equipped with a projector of known directional
and frequency response (omni-direction over the band 400 Hz-2 kHz), is capable of transmitting a
pre-programmed waveform and dead-reckoning in accordance with a pre-programmed geometry.
Through the use of a linear frequency-modulated (LFM) tracking sweep, a high-precision clock, and
two GPS capable sonobuoys, the system employs a long baseline tracking algorithm to provide,
real-time, ground truth reconstruction of the mobile source trajectory independent of the sensor
under test. For a given passive acoustic receiver and broadband processor, it will be shown how
performance metrics such as range-dependent probability of detection (PD) and false alarm rate
(FAR) can be calculated concurrently with the measurement of in-situ transmission loss to quantify
system performance and ultimately provide a measured figure-of-merit (FOM) for a desired system
operating point. PD can also be decomposed as a function of signal-to-noise ratio (SNR) to provide
130
UA2014
a measurement of the system’s recognition differential (NRD). Other important metrics such as
array gain (AG) and signal gain degradation (SGD), and their impact on sonar equation
reconciliation, will also be discussed. Data from a recent field test conducted on the New Jersey
shelf break will be used to illustrate the methodology and interpret detection performance in the
presence of an oceanographic anomaly, a cold pool duct, which was generated from the meander
of a Gulf Stream eddy observed in August, 2011.
Measurement and Model Forecast Comparison of Acoustic Signal-Excess
Fluctuations
Yong-Min Jiang, Christopher Strode
Presenter: Yong-Min Jiang, Christopher Strode
In September 2012, mid-frequency acoustic signal excess measurements were conducted in the Gulf
of Lions to study the impact of geophysical and water column environmental knowledge on sonar
performance prediction. This paper presents a measurement and model comparison of the acoustic
signal excess variations by using different categories of Rapid Environmental Assessment inputs.
Specifically, five different categories of ocean environmental information have been used, including
1) climatology, 2) a high resolution regional ocean model, 3) a high resolution regional ocean model
initialized by survey ship data collection, 4) high resolution regional ocean model updated by
assimilating gliders’ data and 5) grey ship on-scene for an operation. The bathymetric data used
were either from climatology or collected by a multi-beam system for the scientific validation. The
comparison was performed on the data collected along both range dependent and range
independent tracks, seven source-target ranges for each track, two source depths at each sourcetarget range and three different receiver depths. The impacts of varying degrees of fidelity of
environmental information, as well as the conditions/configurations for setting up the acoustic
model, on signal-excess predictions are discussed.
Adding error bounds to energy flux estimates of trasmission loss
Kevin D. Lepage
Presenter: Kevin D. Lepage
The utility of sonar performance prediction tools may be significantly enhance if these tools give
operators some indication of the reliability of their estimates based on the uncertainties of the
environmental parameters being fed to the models. Previous work has been presented by the
author where the energy flux method for predicting the energy intensity in shallow water acoustic
propagation has been extended to treat uncertainty in the boundary conditions or in the index of
refraction. In this paper this work is revisited, with comparisons given between the energy flux
variance provided by the theory and that obtained through Monte-Carlo simulations using a
coupled mode code.
131
UA2014
Session 16: Outer Continental Shelf, Shelfbreak and Canyon Acoustics
Organizer: Jim Lynch
Location: Lecture Room D, Monday 23rd June 2014, 16:30 - 18:10
Acoustical diagnostics of processes on the shelf of the Black Sea
Valeriy Goncharov, Boris Kuryanov, Andrey Serebryany
The target of our work is to study the possibilities of acoustic tomography on the sea shelves where
surface and bottom reflections are important. In October 2010 we carried out a special acoustical
experiment on the shelf of the Black Sea not far from Gelendzhik. In the experiment compound
phase-manipulated signals were transmitted and the time responses of the medium were measured
using three bottom-moored receiving and transmitting transducers separated by 1-km distance.
The variations of the vertical structure of the sound velocity and temperature field were restored
by combined inversion method which was developed in Institute of Oceanology, RAS. The results of
reconstructing the sound and current velocities, as well as internal wave parameters were
compared with the data of independent measurements made by ADCP, thermistor chain, CTD
sounds and other standard oceanographic instruments.
A Normal Mode Approach to Modelling Airgun Signals in Australian Coastal
Waters
Matthew W Koessler, Alexander N Gavrilov, Alec J Duncan
Presenter: Matthew W Koessler
A significant portion of the Australian offshore environment is characterised by sediments of
significant elasticity. Moreover, thin and thick layers in the bottom can significantly affect the
underwater sound propagation when considering the transmission loss of low to mid frequency
signals. Seismic surveys conducted along Australia’s continental shelf are an opportunity to test
shallow water sound propagation modelling results against measured data. The following work
investigates the ability of adiabatic normal mode theory to model the transmission loss of
broadband airgun signals over a range-dependent bathymetry with a layered elastic bottom. The
adiabatic mode results presented here are compared to a recent study (Duncan et al., 2013), which
considered both measured data and modelled transmission loss from a seismo-acoustic parabolic
equation model. Over a 2-100 Hz frequency band, the optimal sound transmission was observed
around the modal cut-off frequencies of the sea surface-seafloor waveguide. At the lower end of
the frequency band, propagation in this environment was dominated by a few one or two low order
modes. The benefits and drawbacks of using an adiabatic normal mode approach are discussed in
the context of both airgun signal modelling and the wider field of underwater sound propagation
modelling. Ultimately, the aim of this work is to develop a better physical understanding of sound
propagation in range dependent environments when the influence of elastic layers in the bottom
cannot be ignored.
132
UA2014
Observations of Horizontal Coupling in the Monterey Bay Canyon
Kevin B. Smith, James M. Upshaw, Timothy Kubisak
Presenter: Kevin B. Smith
Data has been collected on acoustic vector sensors mounted on autonomous underwater gliders in
the Monterey Bay during 2012-2013. In this work, we show results of intensity processing to
estimate bearing to impulsive sources of interest. These sources included small explosive shots
deployed by local fisherman, and humpback whale vocalizations. While the highly impulsive shot
data produced unambiguous bearing estimations, the longer duration whale vocalizations showed
a fairly wide spread in bearing. The causes of the ambiguity in bearing estimation are investigated
in the context of the highly variable bathymetry of the Monterey Bay Canyon, as well as the
coherent multipath interference in the longer duration calls.
Frequency shift of broadband acoustic signals and its relation to the nonlinear
internal waves in shallow water
Broadband acoustic signal propagation in shallow water with strong nonlinear internal waves
(IWs), goes through a measurable frequency shift that is directly related to the thermocline
displacement. This phenomenon is more complicated when the geometry of the IW front is not
simply a straight line, but variable (usually curved) in range. Strong azimuthal dependence of the
acoustic field to the nonlinear IW direction of propagation can further influence this situation. In
this paper we first demonstrate several robust features that we have observed and explained in
shallow water acoustic waveguide where a nonlinear IW is approaching an acoustic source-receiver
track with an oblique angle. Acoustic field then is modeled by a three-dimensional (3D) Parabolic
Equation, along with horizontal rays and vertical modes using a realistic environmental input from
a simultaneously measured 3D nonlinear IW field. Data and model comparison results indicate a
direct relationship between the propagation parameters of the broadband acoustic signals and the
3D sound speed variability due to strong time-varying nonlinear IWs in shallow water waveguide.
This behavior can be used as to monitor some parameters of the internal waves via acoustics.
The “Integrated Ocean Dynamics and Acoustics” (IODA) hybrid modeling effort
Tf Duda, Yt Lin, Ae Newhall, Kr Helfrich, Wg Zhang, M Badiey, Pfj Lermusiaux, Jf Lynch
Presenter: YT Lin
Regional ocean models have long been integrated with acoustic propagation and scattering
models, including work in the 1990s by Robinson and Lee. However, the dynamics in these models
is not inclusive enough to represent submesoscale features that are now know to be very important
acoustically. The features include internal waves, thermohaline intrusions, and details of fronts.
Regional models in theory predict internal tides in many locations, but the nonlinear steepening
these waves and their conversion to short nonlinear waves is improperly modeled, because
133
UA2014
computationally prohibitive nonhydrostatic pressure is needed. To include the small-scale internal
waves of tidal origin, a nested hybrid model is under development. The approach is to extract longwavelength internal tide wave information from tidally forced regional models, use ray methods or
mapping methods to determine internal-tide propagation patterns, and then solve twodimensional high-resolution nonhydrostatic wave models to “fill-in” the internal wave details. The
resulting predicted three-dimensional environment is then input to a fully three-dimensional
parabolic equation acoustic code. The output from the nested ocean model in hindcast mode is to
be compared to field data from the Shallow Water 2006 (SW06) experiment to test and ground
truth the ocean model.
134
UA2014
Session 17: Radiated Noise from Ships and Production Platforms
Session 17a
Location: Lecture Room B, Wednesday 25th June 2014, 14:15 - 16:15
General characteristics of shipping underwater noise
G. Grelowska, E. Kozaczka, S. Kozaczka, W. Szymczak
Presenter: G. Grelowska
Underwater noise control is very hot topic at present. It is connected with role of the underwater
noise as a form of underwater pollution. The increase of the marine transport and the use of the
huge merchant ships have the significant impact on rapidly increase of the underwater noise.
In the paper the basic sources of underwater noise generated by the ship will be analyzed and also
the propagation of it will be taking into account. The main attention will be concentrated on the
sound generated by the ship propeller.
Analysis of underwater acoustic noise measured at the ship bow during sea
trials
Measurements of underwater acoustic noise radiated by ships are typically performed by an
experimental set-up positioned at a distance by the target. To determine the radiated noise level,
at 1 meter from the ship acoustic center, the transmission loss should be computed and great
attention should be paid in setting the parameters of the adopted propagation model. The chance
to assess the underwater noise at the ship bow represents a great opportunity to investigate the
radiated signals in an alternative way, working in a particular location, at short distance from the
acoustic sources of the vessel, whatever and wherever they are. In addition, this measurement
allows to determine the sonar self-noise. This paper presents an analysis of the self-noise observed
at the bow of a modern ship during sea trials, by the means of both broadband levels magnitude
and spectral analysis. The self-noise has been obtained by selecting specific records among a large
set of noise measurements with the aim of evaluating a number of different situations in which the
ship can operate. In particular, it has been analysed the effect on the noise spectrum structure of
the propulsion mode (in terms of auxiliary machineries activity status and propellers rotational
speed) and of the ship conduct (in terms of speed and heading). Nevertheless a widespread and
heterogeneous set of acoustic sources influences the measure of noise, some interesting remarks
regarding the relation between the self-noise and the above-mentioned factors are proposed.
135
UA2014
Sound generation by the ship propeller in the subcavitation range
E. Kozaczka, G. Grelowska, W. Szymczak, S. Kozaczka
Presenter: W. Szymczak
This paper deals with both theoretical and experimental investigation connected with the sound
generation by the ship propeller. The main acoustical characteristic that describing features of the
acoustic field are the time history, power spectrograms and sound intensity. Some experimental
results of investigation will be shown in the functional form – acoustic pressure - advance coefficient
that were obtained in the water tunnel. The simple theory of sound generation by the ship propeller
working in the non uniform stream of water will be shown.
Measurement System to Assess Underwater Noise from Vessels and Marine
Activities
Jesse Spence, Andrew Patterson, Ray Fischer
Presenter: Jesse Spence
Underwater noise produced by vessels and platforms can impact mission goals and marine life. For
example, excessive noise produced by fisheries research vessels may alert fish and other aquatic
animals to the ship’s presence, thereby affecting stock assessments and research objectives.
Similarly, vessels relying on underwater acoustic communications systems, sonar operations, and
other forms of acoustic exploration require noise from the vessel to not interfere with operations.
Vessels and platforms operating in environmentally sensitive areas may also require lower levels of
underwater noise to meet desired or imposed criteria; requirements for limiting underwater noise
radiation are being applied to new vessel and platform constructions by owners. Regulatory bodies
such as the International Maritime Organization and others are also considering regulations on
underwater noise for shipping, at a minimum, to combat the rise in noise levels in the world’s
oceans.
A highly portable underwater noise measurement system has been developed to assist shipyards,
vessel designers, and owners to assess the noise that is produced by new and existing builds. The
system can be used to determine compliance with vessel specifications and/or regulations and to
identify causes of noise components within the radiated noise spectrum. Measurements with this
system comply with the latest regulatory measurement techniques such as ANSI/ASA S12.64-2009
as well as other standards. The system can be deployed from the test vessel using a crane or Aframe, thus eliminating the need for additional vessel support. Measurements can be performed
at any convenient location, with the main limiting factor being environmental conditions (e.g. wave
height and water depth).
This paper presents an overview of the system design and capabilities, as well as example
measurements from testing of a research vessel with a low underwater noise signature.
Comparisons of measurements made with the portable measurement system to measurements
from a permanently moored test array are provided, demonstrating the accuracy of the approach.
136
UA2014
Integrated tool for the acoustic assessment and monitoring of marine
activities and operations
Andre Moura, Erica Cruz, Diogo Lopes, Raul Hospital-Bravo, Josep Sarrate, Pedro Diez
Presenter: André Moura, Raúl Hospital-Bravo
A combined tool to predict and monitor the sound produced by offshore facilities is being developed.
The aim is to provide a top quality, integrated solution to pre-asses, monitor and mitigate the
acoustic impact of offshore energy parks, or other marine operations, while allowing an
optimization of costs. The system consists of a monitoring tool (buoy), which is able to
autonomously collect and send in-situ acoustic data to shore, and a numerical simulation tool,
whose results can be used during the design phase of the facility to predict the acoustic impact and
during the operation phase to complement the information collected by the hydrophones.
The monitoring tool comprises a buoy, or an array of buoys, equipped with hydrophones and other
environmental sensors, self-sufficient power and the data recording and transmission components.
The buoy design focuses on maximizing the autonomy of the tool, improving the reliability of the
data stream and generating the information needed to optimize the frequency and success of
maintenance operations therefore reducing the overall cost. The acoustic data can also be used to
detect anomalies in the mechanical components of the energy devices, by identifying deviations in
their acoustic signature.
The simulation tool computes 2D and 3D sound maps produced by multiple noise point sources, for
medium-sized sea domains. The underwater propagation phenomenon is modeled through the
Helmholtz equation, which considers the wave physics, and includes the seawater absorption.
Non-uniform sound speed vertical profiles and adjustable bottom reflection properties are also
included in the model, making it more realistic than other state-of-the-art alternatives. The
numerical method includes a set of plane waves working as base functions, providing a superior
accuracy with an affordable computational cost. For optimal results the simulation tool will be
calibrated in-situ through acoustic data provided by the monitoring tool.
SIRAMIS : Preliminary Analysis of Acoustic and Seismic Ship Signatures
L. Fillinger, A. Mantouka, C. De Jong, I. Gloza, A. Sanchez, E. Moya, S. Schael, T.
Lennartsson, G. Petit, R. Fardal, H. Hasenpflug, A.L.D. Beckers
SIRAMIS is a project coordinated by the European Defense Agency standing for Signature Response
Analysis on Multi Influence Mines. As most of the international trade is carried out through marine
routes, it is important to evaluate the vulnerability of the merchant vessel fleet to sea mines in order
to be able to limit the potential exposure to this threat.
In this project, the participating nations pool their measurement and analysis capabilities to
improve their knowledge on the underwater signatures of merchant vessels and understanding of
the near field ship signature interaction in relevant and realistic scenarios.
The project involves a series of recording campaigns performed near shipping lanes in the national
waters of the participants, using various multi-influence measurement systems. The data analysis
UA2014
will help to separate the effect of the differences between the measurement systems and the
environments from the features specific to the measured ships. A further analysis will investigate
the relationship between the merchant vessels signatures and their characteristics.
This paper will present the initial results of the acoustic and seismic signature analysis.
138
UA2014
Session 17b
Prediction of propeller radiated noise by onboard measurement
Jong-Hoon Jeon, Won-Ho Joo
Presenter: Jong-Hoon Jeon
Propeller radiated noise from the ship is regarded as a main noise source to disturb the ocean
ecosystem. In order to preserve the oceanic life, MEPC (Marine Environmental Protection
Committee of IMO) has been drafting the underwater noise regulation since 2008. Accordingly, how
to predict propeller radiated noise exactly is an important issue. Conventional method directly
measures the propeller radiated noise using hydrophones located in the sea. This method requires
much time and cost for the installation and maintenance of the measurement system. Furthermore,
it is difficult to measure the propeller radiated noise under the bad weather condition.
This paper proposes an indirect prediction method of propeller radiated noise by onboard
measurement. The method can largely save the time and cost compared with the conventional
direct method. The proposed method firstly measures the noise and vibration transmission
coefficients of hull structure in the dock to estimate how much propeller radiated noise is
transmitted into the ship. Secondly, the noise and vibration inside the ship is measured during the
sea trial, followed by the ambient noise separation. Finally underwater propeller noise is predicted
by using the transmission coefficients and measured onboard noise and vibration signals. The
feasibility of proposed method was verified through the cavitation tunnel experiments. The method
was applied to the prediction of propeller cavitation noise during sea trials for two kinds of ships.
From the results, it is expected that the proposed method may enable to predict propeller radiated
noise with ease, thus to contribute the propeller noise control.
Source level estimates of small cargo ships from measurements in a fjord
S. Coward, H Dong, D Tollefsen
This paper presents estimates of source levels of small cargo ships from measurements conducted
in a shallow-water environment of the Oslofjord (Norway). Data were recorded on a hydrophone
on the Networked Intelligent Underwater Sensors (NILUS) unit placed on the seabed. Noise due to
commercial ships in a nearby shipping lane was measured and corrected to monopole source levels
with use of the RAM propagation model. An environmental model was constructed based on input
from a prior survey of the measurement area. A vertically distributed source model was used in
modeling. Information on ship range and draft was taken from ship AIS data. Estimates of source
level spectra (30 Hz – 1.4 kHz) and broadband radiated noise levels are presented and compared
with values from the literature.
139
UA2014
Pressure signals of low frequency generated by the ship's hull.
J. Bielanski
Presenter: J. Bielanski
Using the boundary methods for modeling the flow around the small motor boats and pontoons,
will be calculated dimensionless characteristic of hydrodynamic pressure field (HPF) in the lower
hemisphere. HPF will be presented at the deep water depth (H> = 10L, L length of motorboats) and
shallow water with a depth comparable to the length of the vessel (H ~ L).
140
UA2014
Session 18: Sensitivity of underwater acoustic observables
Organizer: Emmanuel Skarsoulis
Multiple-Source Localization and Environmental Sensitivity
Stan E. Dosso, Michael J. Wilmut
This paper develops a highly efficient multiple-source localization algorithm and considers the
sensitivity of this approach to uncertainties in ocean environmental parameters. The algorithm is
based on Bayesian matched-field localization and treats the number of sources, source locations
(range and depth), complex source strengths (amplitude and phase), and noise variance as
unknown parameters. The number of sources is estimated by minimizing the Bayesian information
criterion, while source locations are sampled using Gibbs sampling. Source strengths and noise
variances are sampled using an implicit maximum-likelihood formulation. This approach requires
solving a large number of matrix inversions (100,000s), which can be carried out in parallel on a
graphics processing unit (GPU). This provides a highly efficient algorithm which can simultaneously
localize a large number of sources (e.g., 5-10) in real time. Like all matched-field methods,
localization performance is degraded by uncertainty in ocean environmental parameters
(environmental mismatch). While this can be addressed by including environmental parameters in
an augmented inversion, the sampling of environmental parameters cannot be parallelized in a
straightforward manner and the efficiency is lost. However, the sensitivity to environmental
uncertainty can be mitigated by processing multiple frequencies and/or time samples (snapshots),
which can be carried out efficiently in parallel on the GPU although the number of parameters
increases to include source strengths and variances for each frequency and snapshot. This paper
examines the trade-offs involved in this problem.
Sound-speed tomography using angle sensitivity-kernels in an ultrasonic
waveguide
Florian Aulanier, Hacheme Ayasso, Philippe Roux, Barbara Nicolas, Jerome Mars.
Presenter: Barbara Nicolas
Classical ocean acoustic tomography uses travel-times as observables to retrieve sound-speed
perturbations from acoustic measurements. In this study, inversions of new observables (emission
and reception angles) are performed to retrieve sound-speed perturbations.
Short-range (~1 km) acoustic propagation of broadband mid-frequency signals (~2 kHz-central
frequency and 1.5 kHz bandwidth) in shallow-water waveguides (~50 m) shows strongly interfering
multiple acoustic paths. The use of source-receiver arrays and double-beamforming array
processing allows separating most of the acoustic arrivals and associating them to the different
acoustic paths. Then, it is possible to relate linearly small sound-speed perturbations to small
141
UA2014
variations of travel-times, emission and reception angles with the time-angle sensitivity kernel (TA-SK) theory.
Two methods are tested to invert angle variations using the appropriate sensitivity kernels in order
to estimate sound-speed perturbations produced by a thermal plume during a small-scale
laboratory experiment. Both methods are based on a Bayesian approach with smoothness prior for
heat map. The first estimates the conditional Maximum a posteriori (MAP) given the smoothness
parameter while the second calculate the joint MAP for the heat map and the smoothness
parameter under positivity constraint. Time-series of sound-speed perturbation estimates of both
methods exhibit the convection of a hot-water particle in agreement with thermocouple
measurements. However, joint MAP shows better results in retrieving positive sound-speed
perturbations at the vicinity of the heater, where it is not obvious in conditional MAP results.
Furthermore, investigations on the validity domain of the sensitivity kernel approach have been
undertaken. In particular, experimental data show that the linear assumption of the first-order Born
approximation may no longer be valid depending on the size and strength of the thermal plume.
Travel-time sensitivity kernels in a shallow water environment
E.K. Skarsoulis, B.D. Cornuelle, M.A. Dzieciuch
Presenter: E.K. Skarsoulis
Travel-time sensitivity kernels of peak and phase arrivals in a Pekeris shallow-water waveguide are
considered. Peak arrivals are defined as the maxima of the acoustic pressure envelope at the
receiver, whereas phase arrivals are defined as the maxima of the received acoustic pressure itself.
The supports of the sensitivity kernels are concentrated about the corresponding eigenrays
connecting source and receiver. While the two-dimensional (2D, range-depth) kernels of peak and
phase arrival times are comparable, the three-dimensional (3D, range-depth-azimuth) kernels
exhibit differences in shape and magnitude, which point to differences in the way that the two
observables sample changes in the water mass: the sensitivity of phase arrival times exhibits axial
symmetry about the eigenray, whereas that of the phase arrival times persists in the horizontal
cross-range dimension, i.e. sideways from the eigenray. Phase arrivals can resolve the sensitivity
kernel components of overlapping peak arrivals. Otherwise the separation of peak arrivals and the
corresponding sensitivity kernels would require a very high bandwidth. [Work supported by ONR].
A hybrid approach for ocean acoustic tomography based on statistical
characterization of the acoustic signal and the identification of modal arrivals.
Michael Taroudakis, Costas Smaragdakis
Presenter: Michael Taroudakis
A hybrid approach is presented for problems of ocean acoustic tomography, based on the statistical
characterization (SC) of the acoustic signal. The statistical characterization is used for the
estimation of a reference solution to the inverse problem of estimating the sound speed profile in
the water column using a Genetic Algorithm. By applying first order perturbation approach,
variations of the sound speed profile are associated with modal travel time variations. This
142
UA2014
relationship provides the framework for the development of an iterative scheme which converges
when the reference environment is close to the actual one and provides a fine tuning of the results
obtained by the original method. The performance of the method is demonstrated by means of
simulated experiments in range-independent environments.
Non-Perturbative Evaluations of Time Sensibility Kernels using Alternative
Definitions of Propagation Delay
Xavier Cristol
The Time Sensitivity Kernel (TSK) classically quantifies the observability of sound speed fluctuations
from time delays; it is defined as the Frechet derivative of propagation delay on variations of sound
speed, and is generally derived using the perturbative approximations of Born, Rytov or KellerRytov. An intellectually bothering, but universally admitted property of the TSK is the so-called
“doughnut banana paradox”: in 3D, the TSK is zero just along the eigenray connecting source and
receiver, and features maximum weight in a tube around this ray. Recent claims attribute this
paradox to Born or Rytov approximations as only first order terms, to be “corrected” by better
evaluation techniques (Djebbi & Alkhalifha, 74th EAGE, Copenhagen, 2012).
For investigating this question, we evaluate exactly the TSK in free space and in stratified channels.
Firstly, we used three different “operational” definitions of travel time:
1) location of maximum correlation with transmitted waveform (definition pertinent e.g. for active
tomography configuration using known wide-band waveforms);
2) centre of mass of the squared pulse envelop (pertinent for passive configuration ignoring the
form of transmitted waveform, assumed featuring a flat energetic envelop);
3) “instantaneous time” following Djebbi and Alkhalifa definition (the imaginary part of the
derivative of the phase on angular frequency, divided by the phase itself), which is pertinent for a
waveform with a flat spectrum over a part of its support.
Secondly, Frechet derivatives of time delays defined according to these alternative definitions are
evaluated, with no further approximation, using analytical or modal expressions of the Green
function.
The different three definitions of time delay produce highly similar, but different expressions for the
TSK, depending not only on mean frequency, but also on the various parameters of the waveform:
pulse length, bandwidth... In single eigen-ray configurations, all these TSK feature the “doughnut
banana” structure, with cancellation of TSK along the mean ray. Our analysis based on exact
evaluation of the TSK confirms the common opinion based on perturbative expressions of TSK.
143
UA2014
Session 19: Sonar Performance Modeling and Verification: Applications to Active
and Passive Sonar
Session 19a
Location: Lecture Room B, Monday 23rd June 2014, 14:15 - 15:55
Overview of the reverberation component of TREX13
Brian T. Hefner, Dajun Tang
In the spring of 2013, a shallow water reverberation experiment was conducted to measure
contemporaneous acoustic and sufficient environmental data so detailed model/data comparisons
could be achieved and important environmental factors could be identified for different
applications. The Target and Reverberation Experiment (TREX13) was sponsored by the US Office
of Naval Research and the Strategic Environmental Research and Development Program. It was
conducted from April to June of 2013 off the coast of Panama City Beach, Florida, in collaboration
with multiple institutions and involving three research vessels: The R/V Sharp, R/V Walton Smith,
and the Canadian Force Auxiliary Vessel Quest. From a SONAR viewpoint, reverberation consists of
two-way propagation and a single backscatter. Therefore, reverberation, transmission loss, and
bottom backscatter were repeatedly measured over a time period of several weeks in the frequency
band of 2-10 kHz, along with extensive environmental measurements. To reduce the area over
which environmental measurements were needed, the reverberation was measured using a
horizontal line array mounted 1 m above the seafloor in 19 m of water. The reverberation,
transmission loss, and bottom backscatter were measured along a single beam of the array out to
a distance of 7 km. Discussed will be planning and execution of the field experiments, strategies
and steps for data analysis, and modeling efforts.
Correlation of reverberation with bottom sand waves along the TREX
reverberation track
Dale D Ellis, Sean P Pecknold, John R Preston, Jie Yang
Presenter: Dale Ellis
The Target and Reverberation EXperiment (TREX13) took place in the Gulf of Mexico just off the
coast of Panama City, Florida. The reverberation experiments were conducted, weather permitting,
between 22 April and 16 May. Both source and receiver, ITC2015 and FORA, were fixed in location
and were 1.8 and 2.1 m above the seabed respectively. Of particular interest here are various pulses
between 1800 Hz and 3600 Hz, with the latter frequency being near the design frequency of FORA.
During TREX13, reverberation data were taken during all hours of the day, allowing study of
reverberation variation over time and sea surface conditions. In addition to the time and weather
dependence, the directional dependence of reverberation level (RL) could be determined using the
UA2014
FORA triplet array. The focus of the experiment was on RL returning from a track of relatively
uniform water depth of 20 m, extending about 10 km to the southeast of the source and receiver.
The most interesting observation was the effect of the bottom sand waves or dunes, roughly 1 m
peak to trough and spaced about 300 m apart, on RL. This correlation between the two had been
noted in the pre-TREX experiment in 2012 [Ellis and Preston, UA2013] but not analysed in detail.
Predictions from an adiabatic normal mode reverberation model [Ellis et al., ISURC 2008] were used
to compare with measurements, using the detailed bathymetry, but otherwise with inputs
independent of range. As expected, the model predicts a peak in the reverberation at the peak of
the sand dunes. However, counterintuitively, the peaks from the data are, more often than not,
anti-correlated with the peaks of the bathymetry, i.e. high RL correlated with the troughs of the
sand dunes. Clearly, some mechanisms other than depth effects are responsible for the changes in
RL. Extensive bathymetric and bottom measurements have been made along this track; these are
being investigated by other researchers to facilitate understanding of the reverberation
mechanisms.
[Supported by US Office of Naval Research, Ocean Acoustics Code]
A fast algorithm for the computation of incoherent propagation loss for
variable water depth: a validation study
H. Ozkan Sertlek, Michael A. Ainslie
Accurate and fast estimation of propagation loss (PL) is needed for simulations of sonar or acoustic
communication performance, and for environmental risk assessment. Accurate calculation of PL in
range dependent and lossy waveguides can require computationally expensive wave theory
techniques . In this work, a new propagation model is introduced for a range-dependent isovelocity
waveguide. The proposed model is based on a combination of Weston’s average intensity approach
and adiabatic normal mode theory. Thus, mode effects arising at both low frequencies and long
ranges are automatically considered. Using this approach, the depth and frequency dependence of
PL can be calculated analytically in terms of the Faddeeva function without tracing rays or
calculating normal mode eigenvalues (except low order modes). For the validation, PL results are
compared with KrakenC and Bellhop by using test cases from Weston Memorial Workshop 2010.
Comparison with results computed independently using a full wave model demonstrate the
accuracy of the model applied to WMW 2010. Thus, the proposed solution can be an alternative
solution for large scale problems such as broadband calculations in sound mapping without
requiring long computational times.
145
UA2014
Sonar Performance Modeling and Verification: Applications to Active and
passive Sonar
Xavier Cristol, Bruno Chalindar
Presenter: Xavier Cristol, Bruno Chalindar
The surface reflection loss, generally expressed as a decrease of acoustic intensity in dB per surface
bounce, is of a crucial importance in sonar performance assessment; we review and evaluate the
most classical formulas for this surface loss, widely used in the field of underwater acoustics. We
classify and analyse the different existing algorithms, for frequencies spanning from a few 100 Hz
up to some 25 kHz:
1) Theoretical formulas relying on the effective loss due to surface roughness only, like the Marsh,
Schulkin and Kneale formula (1962), with corrections by Kuo (1988); such formulas include oldfashioned models for sea surface roughness spectra (Pierson-Neumann, Pierson-Moskovitz) and
need to be modernized with more accurate spectra, like JONSWAP. Effects due to other phenomena
like loss due to near surface air bubbles are by principles discarded in such algorithms.
2) The strange case of the formulas referred to Beckmann and Spizzichino, that have nothing to
do with these authors, is considered; we follow and clarify the story of these formulas down to their
very first origin in unclassified documents: an empirical formula established in 1962 by Marsh and
Schulkin, from a set of experimental data with no information concerning dependence on incident
angle.
3) Empirical formula of loss in configurations where the sound field is confined near the surface:
surface channels associated with mixing layers in deep waters (AMOS experiments in the 1940’s –
1950’s; Saxton & Baker measurements, 1955, etc.); shallow sea channels with hard sandy or
gravelly bottom (Weston and Ching data of the 1960’s and formulas, 1989).
We use and compare the different formulas in a set of typical scenarios involving passive and active
sonars in typical waters. Global recommendations for the choice of algorithm and their applicability
domain are given for the low frequency Sonar domain.
Analysis of sonar detection performance in South China Sea for ASW using
ASORPS
Chiao-Ming Peng, Yuan-Ying Chang, Yin-Ying Fang, Chung-Wu Wang, Chi-Fang Chen
Presenter: Chiao-Ming Peng
This paper deals with analysis of sonar detection performance and the temporal and space
variations in South China Sea. In this area, the underwater environment is complex with water
depth from 100m to 2000m, internal waves occurring daily, and typhoons impacting annually.
Underwater target detection is the key issue to antisubmarine warfare. The tactics depends heavily
on sonar range prediction. The uncertainty in ocean environment causes inaccurate target
detection. A sonar range prediction model (Advanced Sonar Range Prediction System, ASORPS) is
used and validated by simulation and experiment.
146
UA2014
In ASORPS, we first use the range-dependent acoustic model with the hydrographic output of the
Taiwan Coastal Ocean Nowcast and Forecast System (TCONFS) to map the sound speed profile
distribution, geography and bottom character database to compute reverberation level,
transmission loss and ambient noise levels, then calculate detection coverage and probability.
Acoustic data measured with two 4-hydrophone arrays moored on 400m depth sea bed and
receiving 5 kHz signal from an active source in a circular track of 5 km in diameter around Don-Sa
Island in 2013 summer are used to compare with the simulation.
This paper also presents the simulation results of the sonar detection performance in South China
sea of different seasons, and with effect of internal waves in different locations.
147
UA2014
Session 19b
Location: Lecture Room B, Monday 23rd June 2014, 16:30 - 17:30
Sediment acoustics: The need for improvement
Nicholas P. Chotiros, Marcia J. Isakson
Presenter: Nicholas P. Chotiros
The high frequency environmental acoustics sediment model published in the High-Frequency
Ocean Environmental Acoustic Models Handbook (APL-UW 9407), which has been widely adopted
by underwater acousticians and sonar modelers, is examined in the light of recent sediment
acoustic models and measurements. Due to multiple scattering effects the sound speeds and
attenuations for the larger grain sizes (phi<-1) need to be updated. The sediment densities for the
middle range of grain sizes (1<phi<5) are underestimated due to poro-elastic effects. Sound speed
dispersion and frequency dependence of attenuation are over simplified. A poro-elastic model with
extensions to account for grain contact physics is proposed. For practical applications, an efficient
parameterization of the poro-elastic model allows the number of adjustable parameters to be
reduced to a level comparable with that of simpler fluid and elastic models, while retaining all its
physical advantages. [Work supported by the Office of Naval Research, Ocean Acoustics Program].
Dynamic Self-Organizing algorithm for unsupervised segmentation of sidescan
sonar images
Ahmed Nait-Chabane, Benoit Zerr, Gilles Le Chenadec
Presenter: Ahmed Nait-Chabane
This paper deals with the dynamic neuronal approach for segmentation of textured seafloors from
sidescan sonar imagery. The classical approach developed in our previous work [1] based on
classical Self-organizing maps (SOFM) approach gives good results. The result of the classification
is a set of sediment clusters representing the different kinds of seabed.
However, this classical approach gives satisfying results only when a comprehensive training set is
available. If the training set lacks a particular kind of seabed, it will be unknown for the classifier
and the classification will be reduced to the closest known sediment cluster. As it is not always
feasible to know the entire seabed types before the training phase, a dynamic algorithm solution
capable of incremental learning has been developed.
The Dynamic Self-organizing maps (DSOM) algorithm used in this work is an extension version of
classical SOFM algorithm developed by Kohonen. It is based on growing neuronal map size during
the learning processes. Therefore, the size of the map is small in the beginning but increase
dynamically using control vigilance threshold.
To assess the consistency of clusters sediment representation, the proposed approach is tested on
large data sets acquired in the same region but also on data sets from two different geographical
regions.
148
UA2014
Performance analysis of single receiver Matched-Mode processing for source
localization
Yann Le Gall, Francois-Xavier Socheleau, Julien Bonnel
Presenter: Yann Le Gall
Matched-Mode processing (MMP) is an alternative to Matched-Field processing (MFP) that
explicitly uses the modal description of the acoustic field to estimate source position and/or oceanic
environmental parameters. MMP involves two steps, first the modes are separated, and then they
are compared to mode replicas in order to infer the desired parameter values. If the source emits a
short time broadband signal, modal separation can be achieved with a single hydrophone using a
time-frequency analysis tool called warping. Within this framework, a method that allows to predict
MMP performance is provided: the Ziv-Zakai bound is derived, which gives a bound on the
achievable performance that is much tighter than the Cramer-Rao bound. The results are then used
to perform quantitative performance analyses of the single receiver MMP source localization
problem. Particularly, performance with respect to noise is examined for incoherent MMP.
149
UA2014
Session 20: Sonar Signal and Information Processing
Session 20a
Location: Lecture Room A, Wednesday 25th June 2014, 10:45 - 12:45
on Performance as Implemented with OASES
Arthur B. Baggeroer, Henrik Schmidt
Presenter: Arthur Baggeroer
We consider the minimum mean square resolution limits for estimating the coefficients of an EOF
expansion of sound speed profile representing the haloclene and pynocline layers in the Arctic using
ocean acoustic tomography. The Cramer-Rao Bound (CRB) sets a greatest lower bound for an
unbiased estimate of these coefficients. The geometry modeled is a point source transmitting to a
vertical line array placed within the surface sound channel. The dependence of these limits upon
source/receiver separation, receive array length and number of sensors as well as source center
frequency and bandwidth. The signal model is wide band coherent with a complex Gaussian gain
which transmits N pulses observed in a surface Kuperman-Ingenito surface noise model The CRB
implies the optimal processor is a whitened matched filter followed by square law detection and
envelope averaging over the N pulses. This is a first step towards bounding the performance with
a parameter grid for horizontal variability.
Source motion parameter estimation using direct and multipath arrivals at a
pair of hydrophones
Kam W. Lo, Brian G. Ferguson
The signal emitted by an acoustic source moving on the sea surface or under water arrives at a
hydrophone located above the sea floor via a direct path and one or more multipaths. By exploiting
the direct and multipath arrivals at a pair of hydrophones, it is possible to estimate the complete
set of five motion parameters of the source as it travels past the hydrophone pair in a straight line
at constant speed and constant height above the sea floor, provided that it is known a priori on
which side of the hydrophone pair the source’s closest point of approach is located. When the
source is a surface vessel and the two hydrophones are positioned close to the sea floor in a shallow
water environment, the cross-correlogram of the hydrophone pair shows three dominant, distinct
tracks. The middle track, with positive amplitude, represents the temporal variation of the
differential time of arrival (DTOA) between the direct path signals at the two hydrophones. The
upper or lower track, with negative amplitude, represents the temporal variation of the DTOA
between the direct path signal at one hydrophone and the bottom-surface-reflected path signal at
the other hydrophone. A model for these three tracks is derived, which is a function of time and the
UA2014
five motion parameters of the source. A least-squares fit of this model to the observed tracks
provides estimates of all five motion parameters of the surface vessel. The effectiveness of this
source motion parameter estimation method is demonstrated using real data recorded from a
bottom-mounted hydrophone array in a shallow water experiment for the transits of two different
vessels.
Compressive sensing in acoustics and seismology
Peter Gerstoft, Caglar Yardim, Angeliki Xenaki, Christoph Mecklenbrauker
Compressive sensing (CS) is a technique for finding sparse signal representations to
underdetermined linear measurement equations. I will describe why compressive sensing has found
to be useful and then focus on a few applications in acoustics and seismics: 1) high-resolution
beamforming. 2) In passive fathometry, compressive sensing is used to invert for the number of
sediment layer interfaces and their depths. 3) Earthquake location.
Sparsity and super-resolution in sound source localization with sensor arrays
Angeliki Xenaki, Peter Gerstoft, Klaus Mosegaard
Presenter: Angeliki Xenaki
Sound source localization with sensor arrays involves the estimation of the direction-of-arrival
(DOA) from a limited number of observations. Compressive sensing (CS) is a method for solving such
undetermined problems which achieves simultaneously sparsity, thus super-resolution, and
computational speed. We formulate the DOA estimation as a sparse signal reconstruction problem
and show that methods which exploit sparsity have superior performance compared to traditional
methods for DOA estimation. To demonstrate the high-resolution capabilities and the robustness
of CS and other sparsity promoting optimization techniques in DOA estimation, the methods are
applied to experimental data from underwater acoustic measurements in the challenging scenario
of source tracking from single snapshot data.
DoA Estimation Algorithm as Applied to Wideband Processing
Gleb Sidelnikov
Presenter: Gleb Sidelnikov
In some hydroacoustic applications, direction of arrival (DoA) of several wideband signal sources
should be estimated, with the angular separation between the signals less than the width of the
antenna array directional characteristic. In this case the Rayleigh criterion resolution cannot be
applied, and superresolution algorithms should be used. There are many superresolution
algorithms based on various approaches such as solving polynomial equations (Prony, root-MUSIC),
estimating the spatial spectrum of signal power (Capon, Johnson). A superresolution algorithm for
narrowband signals has been proposed, based on the solution of observation equation. This paper
extends the algorithm to wideband signals.
151
UA2014
According to the proposed algorithm, DoAs of wideband sources are the roots of the same equation.
DoA estimation is stable if the product of the spectral analysis bandwidth by the observation time
exceeds the number of sources by more than one. With a linear equispaced antenna array, it is
sufficient that half of the corresponding value is exceeded for stable DoA estimation. A particular
solution is separated from the general solution to minimize the DoA estimation variance. If the
observation time is increased, the mathematical expectations of DoA estimation tend to zero, and
the variance asymptotically tends to the minimum possible value determined by Rao-Cramer
inequality. The algorithm does not require much computational resources. The presented results
from mathematical simulation validate the theoretical conclusions.
Estimating source spectra from recordings made in a reverberant underwater
channel
Kay L. Gemba, Eva-Marie Nosal
Presenter: Kay L. Gemba
The performance of a detector can be optimized by understanding the spectral characteristics of
the signal of interest. However, anechoic underwater recording environments required to make
ideal recordings are generally not available or are cost-prohibitive. This talk presents a practical
solution for working in an imperfect recording environment; the source spectra is obtained by
equalizing the recording with the correctly scaled inverse of the channel impulse response (IR). An
experiment was conducted at the University of Hawai’i diving well (depth of 18ft) using various
known source signals such as sweeps, tonals and band-limited signals (band-width of 10 kHz).
Inversion of IR is accomplished in a least-squares sense and dereverberated tonals and band-limited
signals are compared to their original signatures. Transfer functions and impedance mismatch of
the playback equipment were unknown and transducer source levels were not available to scale
the IR. However, frequency dependent scaling of the IR was possible since the combined amplitude
response of the playback and recording system was known. In particular, scaling is accomplished
by using a filter bank in the time domain. The IR is first filtered into several band limited versions,
each with similar amplitude response. Next, each band-limited version is scaled before the entire
set is re-assembled prior to inversion. This procedure removes the skew due to the amplitude
response and properly scales the IR prior to inversion. Experimental results are compared to
synthetic results of an image-source model of the diving well to quantify scaling and
dereverberation performance. While the dereverberation method was previously applied to SCUBA
signals, it is generally applicable to other point-sources if the impulse response of the recording
environment can be obtained separately.
152
UA2014
Session 20b
Multivariate Distributions of Clutter Levels for Automated Classifiers
Bruce Newhall, Anna Slowikowski
Presenter: Bruce Newhall
Active sonar automated classifiers begin by identifying potential targets in a set of acoustic data.
These candidates must pass a detection test of having sufficient level above the detection threshold.
The levels at these points also must be a local maximum. Once candidate target locations have
been selected a small snippet of data is excised centered about each candidate point. The behavior
of the time series within these snippets is used to classify signals of interest from clutter. The
distribution of data levels within the snippets may be required to define an optimal classifier. A
general formula for the conditional distribution of levels is derived given that a point is within the
snippet. The conditional distributions are a function of the distance of the point from the central
peak, and depend on the bivariate distribution of pairs of points prior to snippet selection. We
compare these analytic distributions with data taken from the NATO Base 04 exercise. Before
snippets are formed, the prior data intensities are well modeled with generalized Pareto
distribution marginals and a bivariate Gumbel copula.
Seafloor classification using statistical modeling of wavelet subbands
N.-E. Lasmar, A. Baussard, G. Le Chenadec
Presenter: N.-E. Lasmar
This paper deals with the classification of textured seafloor images recorded by sidescan sonar.
In this study, the statistical modeling of these textured images uses multi-scale and multiorientation representations. We propose to characterize textured images by parametric
probabilistic models of wavelet coefficients. For the supervised classification task, our approach is
based on the Bayesian framework. The interest of the use of a probabilistic model, in this
framework, consists in minimizing misclassification rate provided that the model fit the observed
statistics.
The aim of this paper is twofold. Firstly, we study the observed distributions of wavelet subbands.
In this way, the generalized Gaussian (GG) distribution, which is a well-established model to
characterize the marginal distributions of the wavelet subbands, is considered. To take into account
the joint statistics of wavelet coefficients, we consider also the Gaussian copula based multivariate
generalized Gaussian (GC-MGG) model. The goodness-of-fit is studied and validated using the
classical Kolmogorov-Smirnov test for the GG model and an appropriate multivariate test for the
GC-MGG model. Secondly, a supervised learning context is adopted for the classification stage by
using a probabilistic k-Nearest Neighbors classifier. Each textured image will be represented by its
153
UA2014
GG or GC-MGG estimated parameters and given a collection of training images, the Kullback-Leibler
divergence is used to estimate the similarity between a test image and seafloor classes.
A database of small images (typically 200x200 pixels) has been compiled from Klein 5500 data
recorded during the Battlespace Preparation campaign carried out by the SACLANT Undersea
Research Center in La Spezia, Italy. This database is composed of six classes: silt, sand, Posidonia,
vertical ripples, 45° ripples, and rock.
The obtained results on the database, using both models (GG and GC-MGG), are reported and
analyzed. They show the interest of this kind of classification method for sonar image classification.
Classifying continuous active sonar echoes for target recognition
Stefan M. Murphy, Paul C. Hines, Kevin Dunphy
Presenter: Stefan M. Murphy
Classification and tracking are two important techniques for enhancing active sonar performance.
Classification rejects unwanted clutter using echo analysis, and tracking provides a history of target
motion while rejecting clutter that doesn’t support realistic target motion. Continuous active sonar
(CAS) has been proposed as an alternative to conventional pulsed active sonar (PAS), largely in
order to provide tracking updates at a much higher rate than is possible with PAS. Unfortunately,
these faster updates come at the cost of reduced classification performance, at least for CAS that
uses linear frequency modulated waveforms. In this case, maximizing the update rate requires subband processing. Classification of echoes from these sub-bands is expected to be relatively poor,
since the full bandwidth is favoured for classification. An alternate processing scheme for CAS uses
full-band processing, which is typically used for PAS. This potentially maximizes classification
performance rather than providing faster updates as in the sub-band approach. A risk of this
scheme is the potential for complications in echo signals arising from coherence loss caused by the
long duration of CAS waveforms. One facet of a recent Canada-U.S. sea trial, TREX13, focused on
conducting experiments that allow direct comparison of the performance of CAS and PAS in shallow
water. In this paper, DRDC’s echo classification software was tested with sonar echoes from
TREX13. The software, which was originally developed for PAS applications, was used to evaluate
whether CAS echoes can be classified as accurately as PAS echoes.
Signal Change Detection Method Used for Mine-Like Objects Segmentation in
Sonar Images
Ivan Aleksi, Dieter Kraus, Zeljko Hocenski
Presenter: Ivan Aleksi
In this paper a Signal Change Detection (SCD) method is used and a novel thresholding (SCDT)
approach is presented. SCDT is used for segmenting underwater Mine-Like Objects in side-scan
sonar images. Generally, SCD is a statistical method used for detecting the time instances where
the amplitude levels of amplitude discrete signals change significantly. SCD approximates 1D real
signals by choosing for a predefined number of jumps that amplitude discrete signal that achieves
maximum similarity, i.e. fitting probability. Changes may occur at unknown time instances and
154
UA2014
unknown amplitude levels. In this paper, the row and/or column vectors of 2D Sonar intensity
images are considered as 1D signals. Two unknown time instances are detected with three
amplitude levels, which are calculated as mean values within the detected time indices.
Experimental results show that the SCDT approach achieves an efficient segmentation of sonar
images in the desired background, shadow (low intensity) and object (high intensity) region.
Underwater moving targets detection using the sequence of acoustic image
Teng Tingting, Sun Dajun, Lan Hualin, Mei Jidan
Presenter: Teng Tingting
The range-azimuth acoustic image can be obtained by the active sonar using the beamforming
technique. Since the complex underwater environment, not only the strength of the targets echoes,
but also the clutters and reverberation fluctuate from ping to ping. These problems make the
moving targets detection more difficult. The traditional Detection-Before-Tracking (DBT) technique
usually fails while the Signal-to-Clutter/Reverberation Ratio (SCNR) is low. In the paper, the
Tracking-Before-Detection (TBD) technique is applied for the moving targets detection. This method
is based on the energy accumulation and the soft decision detection. Along with the timesequenced accumulation, the target detection probability can be improved. Here the acoustic
image sequences were obtained by the echo signals processing using the near-filed focusing wideband beamforming algorithm, and then accumulated by the projection operator. The target
detection was executed using 2-dimentional CFAR algorithm. The numerical simulations and the
pool experiment were conducted for the performance validation. The results indicated that the
detection probability for the fluctuant target was heightened, and the tracks of the static targets,
clutters and reverberation were reduced during the time accumulation procedure in comparison
with the DBT technique.
Passive sonar denoising for diver detection in presence of snapping shrimp
L. Fillinger, A.J. Hunter, M.C. Clarijs
Since 2010, TNO has conducted a number of experiments on passive diver detection in Dutch
waters. We have demonstrated detection, localization and tracking of divers wearing open circuit
and closed circuit underwater breathing apparatus as well as boats. However, until now, we have
not conducted any such experiments in warm waters.
The soundscape in warm, coastal waters can be very different due to complex bathymetry profile,
the presence of breaking waves and biological noise.
In collaboration with Aruba Ports Authority, TNO conducted a series of measurements to evaluate
the feasibility of passive diver detection and harbor protection in such an environment. Although
the ambient noise level recorded is lower than in the Netherlands, it is dominated by snapping
shrimp noise. These snaps constitute a non-Gaussian background noise, for which conventional
signal processing techniques are not suited.
155
UA2014
In this paper, the effect of the snapping shrimp noise on the correlation, as well as the effect of denoising techniques applied in the temporal and in the delay domain are demonstrated.
156
UA2014
Session 20c
Target AOU Growth Containment using High Duty Cycle Sonar
Doug Grimmett
Unlike conventional Pulsed Active Sonar (PAS) which listens for echoes in between short-burst
transmissions, High Duty Cycle (HDC) sonar attempts to detect echoes amidst the continual
interference from source(s) transmitting with nearly 100% duty cycle. The potential advantage of
HDC is an increased number of continuous detection opportunities, leading to improved target
detection, localization, tracking, and classification. A common PAS and HDC sonar transmission
waveform is the linear frequency modulated signal (LFM), which provides good target range
estimation. The range accuracy of the waveform is inversely proportional to the processed signal
bandwidth. In traditional PAS sonar processing, the full signal bandwidth is transmitted over a
short time interval within the repeat cycle. It may be processed with a matched filter, resulting in
a single detection opportunity. In HDC sonar processing, the signal’s bandwidth is spread out over
the entire transmission repeat cycle. It may be split up into sub-bands by processing shorter time
blocks, resulting in multiple detection opportunities per waveform cycle. Though the probability of
detection and ranging accuracy may be lower for HDC sonar than for PAS (due to reductions in
source level and processed bandwidth), a distinct advantage is that there is less time lapse between
measurement scans. Such a rapid measurement update-rate effectively contains the growth of the
target’s area of uncertainty (AOU) within a kinematic tracker. This may improve target localization,
holding, maneuver detection, and false track rate. The value of this effect is demonstrated through
processing and analysis of HDC sonar data from the TREX’13 at-sea experiment.
Target Doppler Estimation and Range Bias Compensation using LFM High Duty
Cycle Sonar
Doug Grimmett
Unlike conventional Pulsed Active Sonar (PAS) which listens for echoes in between short-burst
transmissions, High Duty Cycle (HDC) sonar attempts to detect echoes amidst the continual
interference from source(s) transmitting with nearly 100% duty cycle. The potential advantage of
HDC is an increased number of continuous detection opportunities, leading to improved target
detection, localization, tracking, and classification. Continuous transmission waveforms may be of
several types. Frequency modulated waveforms (LFMs) provide good range measurements but not
target Doppler; continuous waveforms (CWs) provide good Doppler measurements but not target
range; and sophisticated broadband waveforms attempt to provide both good range and Doppler
measurements simultaneously. This paper describes an algorithm for Doppler estimation within a
single cycle of a continuous LFM HDC signal. Using this method, target Doppler can be estimated
157
UA2014
more quickly than the typical approach of estimating range-rate over multiple ping cycles. This
approach is possible when the HDC LFM signals are processed over short time intervals, which
provides a set of multiple measurements within each waveform cycle. This is achieved through
standard heterodyne and spectrogram processing. The Doppler estimate is provided at the
information processing stage, and therefore is available to the tracker for improved target state
estimation. Additionally, the obtained Doppler estimates can be used to correct the measurements’
range bias errors which are inherent in processing LFM signals. Bias errors are a more significant
problem for HDC sonar than for PAS because of the reduction in transmitted bandwidth per unit
time (frequency sweep rate). They may pose a particular problem in sensor fusion between
multistatic sonar sensors by preventing data association (gating). However, these bias errors can
be determined using the estimated Doppler and compensated for. The Doppler estimation and bias
correction algorithm is applied to data from the TREX’13 experiment, and shows the effectiveness
of the method.
A Doppler estimation technique is based on the signals with good correlation
properties: Experimental results
Alexander V. Burenin, Vladimir V. Bezotvetnykh, Michail V. Lebedev, Evgeny A.
Voytenko, Dmitry S. Strobykin
Presenter: Burenin Alexandr Victorovich
A Doppler estimation technique is presented. The technique is based on the application of the
composite signal packet. Signal package consists of signals with good correlation properties (e.g.
Barker codes, M-sequence). The proposed approach provides a reducing in computational load and
hardware cost in comparison with the method of ambiguity function. The experimental results
confirming the efficiency of the technique are presented.
Ontology Design for Cooperative Underwater Target Tracking
Frank Ehlers
Presenter: Frank Ehlers
Using information science terminology, cooperative target tracking for multistatic low frequency
active sonar systems can be seen as inherited from a more general ontology for a distributed goal
driven observation system. The term "distributed" refers to both, the spatial distance between
components of the observation system and to local decision making onboard the platforms involved
in the cooperative target tracking operation. The goal that drives the cooperative target tracking
platforms can be formulated for the non-cooperative target case as a minimization of the number
of behavioral choices left to the target. This formulation can be used as the main criterion for the
design of the ontology. The collection of operationally relevant instances of the ontology can be
used to construct the concept of operations for the tracking platforms. We will utilize this
terminology to describe the findings from the analysis of multistatic simulations and sea trials.
158
UA2014
Session 21: Soundscapes and Measuring Noise
Session 21a
Location: Lecture Room B, Tuesday 24th June 2014, 14:15 - 16:15
Applying the dynamic soundscape to estimates of signal detection
Jennifer L. Miksis-Olds, Julia A. Vernon, Kevin Heaney
Presenter: Jennifer L. Miksis-Olds
The field of underwater acoustics is currently struggling with how to define and apply the concept
of a soundscape, which originated for use in the terrestrial environment, to the underwater
environment. Sound (especially low frequency sound) travels greater distances underwater
compared to in air, so sources from 1000s of kilometers away have the potential to significantly
contribute to local soundscapes. Understanding the source contributions within a local soundscape
can therefore be complex. In applying our understanding of the soundscape to signal detection,
sound level trends, or noise impacts, it is first necessary to parameterize the acoustic environment.
This work examines the impact of the dynamic soundscape and selected sound level parameters on
estimates of signal detection area, also referred to as active acoustic space when considering
effective acoustic communication between vocalizing animals. The range of signal detection was
investigated at three site locations of the Comprehensive Nuclear Test-Ban Treaty Organization
International Monitoring System. Transmission loss to each hydrophone was computed using the
OASIS Peregrine parabolic equation model for a source within the upper 300m of the water column
to be consistent with the location of vocalizing baleen whales. Daily, monthly, and seasonal
soundscape measurements were incorporated into the sonar equation to estimate the variability in
signal detection area as a function of sound level and time.
Passive calibration of soundscapes
Thomas Folegot, Dominique Clorennec, Cedric Gervaise, Yann Stephan
Presenter: Thomas Folegot
Quonops©, a global ocean noise prediction platform developed and operated by Quiet-Oceans,
brings together real-time contextual information and environmental data into acoustic
propagation models to deliver a range of soundscapes. Deterministic or statistical maps of noise
are built according to the given set of ocean-meteorological conditions and maritime activities. In
Europe, this modeling approach contributes to fulfill the requirements of the European Marine
Strategy Framework Directive (2010).
Nonetheless, environmental uncertainty, especially weak or sparse data describing bottom
properties leads to difficulties.
A method of passive inversion of the properties of the bottom based on passive measurement of
the global noise radiated by the maritime traffic is proposed. This method is envisioned as a solution
UA2014
to calibrate basin-scale soundscapes that rely on a sparse network of long-term acoustic in-situ
monitoring.
The method has been successfully implemented and tested on data gathered during a passive
acoustic experiment which has been conducted in the Mediterranean by the French Hydrographic
and Oceanographic Office (SHOM).
Ship Noise mapping in the North Sea
M.E.G.D. Colin, M.A. Ainslie, A.A.F.M. Beeks, A. M. Von Benda-Beckmann, C. Booth , N.
Bouton, C. Harris, C.A.F. De Jong, H.O. Sertlek, H.W. Slabbekoorn
Presenter: M.E.G.D. Colin
There is an increasing concern that anthropogenic underwater sound may have a negative impact
on marine life. Governmental authorities are introducing regulations to address this problem. The
European Commission, for instance, has adopted the Marine Strategic Framework Directive
requiring EU Member States to achieve or maintain Good Environmental Status, regarding
underwater noise amongst other forms of pollution. A task group (the technical sub-group on
underwater noise) has formulated indicators for underwater noise pollution, resulting in an advice
to monitor low frequency sound in particular frequency bands (sound pressure level in the thirdoctave bands centred at 63 Hz and 125 Hz). Model generated noise maps were identified as a
monitoring tool to complement measurements.
Recently, the SONIC (Suppression Of underwater Noise Induced by Cavitation) project was awarded
within the European Seventh Framework Program to develop tools to investigate and mitigate the
effects of underwater sound generated by shipping activities. In this paper, we will present the
SONIC approach to generate shipping noise maps. The noise map generation tool uses Automatic
Identification System (AIS) as well as biological distribution data to generate maps representative
of the sound exposure that marine mammals and fish would experience. This tool uses a fast
acoustic model developed specifically for this purpose that was compared to normal modes and
parabolic equation models. Results of the models compared with reference models are presented
in this paper.
Practical spreading laws: the snakes and ladders of shallow water acoustics
Michael A. Ainslie, Peter H. Dahl, Christ A. F. De Jong, Robert M. Laws
Presenter: Michael A. Ainslie
Geometrical spreading laws are widely used in underwater acoustics because they provide - if
chosen carefully - an accuracy that is sufficient for many applications (source characterisation,
impact assessment, sound mapping, regulation) for negligible computation time. The simplest and
most widely used form is that corresponding to spherical spreading, with propagation loss, PL(R),
equal to 20*logR, which can provide a suitable approximation for deep water. In shallow water,
propagation is influenced by multiple reflections from the seabed and sea surface, and a
modification is then needed. The resulting effect depends on the kind of source (e.g., continuous or
transient) and its directivity (e.g., monopole or dipole). The result often simplifies to the form PL(R)
160
UA2014
= A + B*logR, where the values of A and B depend on the conditions. If the source level is known,
the received SPL can then be calculated for a continuous source using sound pressure level, SPL(R)
= SL – PL(R) = SL – (A + B*logR), but the expected behaviour depends on source directivity, and the
values of A and B need to be adjusted accordingly. For a transient source, there are no simple
expressions for sound pressure level, but the sound exposure level can be related in a similar way
to the energy source level. Guidance is provided for the choice of A and B in shallow water for
different activities, including seismic surveys, shipping, explosions, pile driving and use of active
sonar. Guidance is also offered on where not to use these simple rules, such as for the calculation
of SPL for short transient sources and the pitfalls associated with applying far-field concepts such
as source level to distributed sources such as pile drivers.
Signal Grouping by Correlation of Cepstra
Mark K Prior, David Brown
Presenter: Mark Prior
Data recorded on deep-sound-channel hydrophones, connected by cables to shore stations, are
described. The data are gathered for purposes of nuclear test-ban monitoring and are routinely
processed to determine noise spectra and the distribution of sound across arrival azimuth. These
parameters can be used to help describe the underwater sound field at the hydrophone stations
but a fuller depiction of the ocean soundscape requires some level of identification of the sources
responsible for the various components of the sound field.
A method is described in which groups of similar discrete signals are identified via correlations
between their cepstra. Cepstral correlation coefficients are used to form clusters of signals with
clustering thresholds designed to identify the largest non-trivial groups. The method is applied to
data gathered at a hydrophone station operated by the Comprehensive nuclear-Test-ban Treaty
Organisation (CTBTO) at Diego Garcia in the Indian Ocean. Large clusters are formed and these are
identified as containing signals from baleen whales. Residual inter-cluster correlations are
identified as resulting from the presence of overlapping signals of different types.
The views expressed are those of the authors and do not necessarily reflect the view of the CTBTO
Preparatory Commission.
Feasibility of reef health monitoring using passive acoustics
A.J. Hunter, L. Fillinger, M.C. Clarijs
In December 2013, TNO made underwater measurements in Aruba to assess the feasibility of reef
health monitoring using passive acoustics; this work was conducted in collaboration with Aruba
Ports Authority, Aruba Marine Park, and Aruba Reef Care Foundation. Ambient noise recordings
were made at various locations around the reef over a period of several days and marine biologists
performed a survey to assess the species present and to identify healthy and unhealthy regions of
reef. This paper presents results from the ambient noise analysis. The underwater soundscape was
found to be dominated by snapping shrimp noise. The sounds from other species known to be
161
UA2014
present were not observed, presumably due to their low levels and due to measurements made at
a distance from the reef. We present an analysis of the shrimp noise, for which we observed a
diurnal trend, and conclude with some comments on the feasibility of reef health monitoring.
162
UA2014
Session 21b
Sound Maps Of The Dutch North Sea For Natural And Anthropogenic Sound
Sources
H.Ozkan Sertlek, Geert Aarts, Sophie Brasseur, Hans Slabbekoorn, Carel Ten Cate,
Michael A. Ainslie
Regulations for protecting and preserving marine environment (for example in the USA and EU)
often require investigating the potential effect of sound on marine life. The origin of noise can be
natural as well as anthropogenic. To assess the potential importance of various types of sounds,
we constructed sound maps for the Dutch North Sea for both natural (wind, rain) and
anthropogenic (shipping, explosions, seismic surveys). Different sources are important to different
species, partly because of different frequency ranges and partly because of their distribution in time
(e.g., continuous or intermittent; changing suddenly or gradually). Our maps take into account
different averaging times, different swimming depths and frequency-weighting according to
different hearing sensitivities. The underwater acoustic propagation is modeled mathematically by
combining Weston’s average intensity method and adiabatic normal mode theory, which can
provide a fast and accurate results without calculation of normal mode eigenvalues and tracing
rays. These maps, combined with knowledge of animals’ distribution and their physiological and
behavioural reactions to sound, can be a useful indicator for understanding the impact of sound on
marine life in the Dutch part of North Sea.
Analysis of soundscapes in the East coast waters of the UK
L. S. Wang, S. P. Robinson, P.M. Harris, D. Partridge, J. F. Borsani, K. Brookes.
Presenter: L. S. Wang
This paper describes the analysis undertaken of several large sets of ocean acoustic data collected
with commercial autonomous recorders along the East coast of UK. The work is part of a project
that aims to improve the mathematical and statistical methods used to analyse data gathered
using a range of environmental sensor networks. Each data set has been collected over periods of
several months as part of programmes to assess the acoustical environments, the recorders having
bandwidths of up to 48 kHz, and sampled with medium to high duty cycles. The recorders were
placed near to the East coast of the UK: some are remote from heavy shipping areas, but some data
were acquired close to busy shipping lanes. AIS ship traffic information is available for this data,
allowing the effect of this source to be modelled in detail.
The data provide samples of the soundscapes at the sites over extended time periods. Preliminary
analysis of the data has been carried out, and significant artefacts have been identified on some
data sets (for example, caused by flow noise, mooring noise). Further analysis has been undertaken
to determine the frequency composition and temporal variation in the acoustic field, and to model
163
UA2014
the influence of specific sources such as ship traffic. A discussion is provided of some of the issues
for such sensor networks with regard to data assimilation, sensor calibration and temporal and
spatial sampling.
164
UA2014
Session 22: Synthetic Aperture Sonar: State-of-the-art
Organizer: Roy-Edgar Hansen and Daniel Brown
Location: Lecture Room C, Friday 27th June 2014, 8:30 - 9:30
Change detection in topographic structures using interferometric synthetic
aperture sonar
Roy E Hansen, Torstein O Saeboe, Ole J Lorentzen, Oivind Midtgaard
Presenter: Roy E Hansen
Interferometric synthetic aperture sonar (SAS) is a sensor technology well suited for imaging and
mapping of the seabed, providing high resolution and large area coverage at the same time. When
repeated passes are made over the same area, the data can be used to detect changes at the
seabed between the passes. Automated change detection may be useful in many areas within
marine research, offshore oil and gas, and military applications.
A critical stage in change detection processing is co-registration of the images, that is, accurately
positioning the image grid from the second pass on top of the image grid from the first pass such
that image based techniques can be used to detect changes. Traditional navigational accuracy on
towbodys and autonomous underwater vehicles (AUV) is not high enough to supply co-registered
images without using data driven techniques (the sonar images themselves).
Topographic structures may be a challenge in SAS imaging, potentially producing layover,
shadowing, multiple reflections, defocusing and projection errors. This causes also difficulties in the
co-registration process and the change detection process for repeat pass data. An interesting
possibility is to use all the data products from interferometric SAS in the process of co-registration
and applying image based change detection.
In this paper, we consider an area outside Trondheim, Norway, containing coral reefs. Data were
collected by a HUGIN AUV carrying an interferometric SAS in repeated passes December 2012 and
December 2013. The coral reefs are topographic structures with relatively large vertical variations,
and high acoustic backscattering strengt. We test different strategies for co-registration and show
example images and bathymetries from single pass and repeat pass processing.
Comparison of Fusion Approaches for the Displace Phase Centre Antenna
Method
Johannes Groen, Holger Schmaljohann, Wolfgang Jans, Ursula Holscher-Hobing
Presenter: Johannes Groen
Although autonomous underwater vehicles (AUVs) with synthetic aperture sonar (SAS) have been
employed by navies successfully for several years now, further improving its robustness is
nonetheless an important research theme. For synthetic aperture imagery the theoretical azimuthresolution is enhanced compared to real aperture imagery, but only if the sonar signals are
processed coherently. The prerequisites for that are ping-to-ping coherence and adequate motion
estimation. Typical problems that jeopardise these requirements are multipath effects, noise, or
165
UA2014
uncompensated sonar motion. This paper focuses on the motion estimation, which is performed by
fusing data from the inertial motion unit and the sonar itself. The latter estimates ping-to-ping
displacements based on the Displaced Phase Centre Antenna (DPCA) method that uses sonar data
from overlapping phase centres. In cases where DPCA provides inadequate accuracy to enable full
resolution SAS imaging, incorporating additional techniques can be beneficial. The time delay
estimation inside the DPCA method is typically performed many times per ping. The vast amount of
time delay measurements generated provides access to the motion estimation accuracy, e.g.
sudden jumps from ping-to-ping may be easily filtered out. Knowing the estimation accuracy at all
times also enables to fuse information when several SAS arrays are in use. In this paper data is
analysed from the SeaOtter AUV fitted with the Vision1200 dual-sided interferometric synthetic
aperture sonar (SAS), a long range autonomous mine hunting system. Four SAS arrays are included
in the analysis and different fusion techniques are compared. In the analysis the relevance of
estimated motion between port and starboard arrays is regarded in particular.
Alternative SAS processing for gas seep detection
Ann E. A. Blomberg, Andreas Austeng, Roy Edgar Hansen
Presenter: Ann E. A. Blomberg
An unknown amount of the greenhouse gas methane is continuously seeping into the oceans,
impacting marine life and potentially reaching the atmosphere. The introduction of subsea CO2
storage further increases the need for accurate monitoring methods to ensure that potential leaks
are detected. Gas seeps in the water column appear as characteristic ``flares'' in single- and
multibeam sonar images due to the high contrast in acoustic impedance between water and gas.
We investigate the potential of using synthetic aperture sonar (SAS) for seep detection. SAS has the
advantage of offering seafloor images with significantly higher azimuth resolution than
conventional sonar methods, by coherently combining images from multiple along-track pings.
However, a plume of rising bubbles is poorly imaged by a coherent SAS system because it violates
the assumption of time-stationarity. We propose an alternative SAS processing scheme aimed at
detecting and accurately localizing gas seeps at the seafloor. In standard SAS imaging, the nonstationarity of rising bubbles causes a local loss in image intensity. An alternative approach is to
form a SAS image through incoherent combination of individual images. This is sub-optimal in terms
of image resolution, but assures that the image intensity is retained in the presence of a seep. We
combine these two approaches in order to maximize seep detection ability, taking advantage of the
fact that a seep has different characteristics in the two SAS images. The difference in mean pixel
intensity reveals the presence of a seep. We collected data from two seep locations in the North
Sea where shallow gas is escaping through a cracked cement well casing. The seep was imaged
using the HISAS 1030 sonar carried by the Hugin AUV. We show how the proposed processing
scheme can be used to detect and accurately localize even a modest seep.
166
UA2014
Session 23: Tank Experiments
Organizer: Jean-Pierre Sessarego and Dominique Fattaccioli
Source level measurement in deep water conditions: Are free field condition
met whatever the source frequency?
A. Guillou, J-P Sessarego, R. Guillermin, D. Fattaccioli
Since the last years, there has been a growing interest in the measurement of the acoustic level of
underwater sources. This is related to the protection of environment and more specifically to the
protection of marine mammals. One of the main contributions to ocean noise is due to ship traffic.
In order to measure the radiated noise level from ships, measurements are usually made in deep
water conditions (> 300m), and at short distance of the hydrophones, the working frequency being
in the range [50 Hz - 50 kHz].
If lower frequencies are produced in the water column (development of oil industry and pile driving),
these free field conditions are violated. The interactions between surface and bottom affect the
measurements and even at short distance from the source, it is difficult to consider that free field
conditions are satisfied. The question which arises is: what is the error made in the estimate of the
sound level. To answer this question an experiment was performed in a large water tank at LMA.
In that experiment a scale factor of 1/10.000 was applied. The bottom of the tank was made of a
thick layer of sand simulating an infinite bottom at sea. An omnidirectional source emitting tone
bursts in the range (50 kHz-500 kHz) was used and its level was measured first in the large part of
the tank (free field conditions), and then in a water layer of 6 to 8cm (shallow water conditions).
The source was placed at different depth in the waveguide. The transmission loss as a function of
distance source-receiver was measured and this measurement was compared to a calculation
including all modes. In this computation the attenuation in the sediment was taken into account in
order to be as close as possible to the experimental conditions.
Calibration of ultrasound transducer heads using short preprocessed
ultrasonic pulses
Panagiotis Papadakis, George Piperakis, Spyros Kouzoupis
Presenter: Panagiotis Papadakis
In a previous work the authors have presented a method for the generation of short ultrasonic
pulses in medium size tanks. In this work an application of this method for the calibration of
ultrasound transducer heads will be compared to typical calibration methods. The Laboratory of
Underwater Acoustic Measurements at IACM-FORTH, Crete (Greece), owns a medium size water
tank (3m x 1.5m x 1.3m) which is mainly used for reflection and calibration experiments. The
transducers mainly used in these experiments are piezoelectric elements with central frequencies
around 500 kHz and bandwidths of 200 kHz. The beamwidths of the narrow beams depend on the
167
UA2014
transducer head sizes. The manufacturer of the transducers does not provide calibration charts and
documents for these instruments. Two different transducer heads, one with small diameter and one
with large diameter, are calibrated using a conventional method and the one proposed here
through the use of pre-processed ultrasonic pulses. Calibration was performed by the two methods
at different distances using a calibrated reference hydrophone. Results from the two methods are
compared and conclusions are drawn regarding the applicability of the new proposed method at
calibration experiments.
De-Coherence Effects in Underwater Acoustics: Scaled Experiments.
G. Real, J.-P. Sessarego, X. Cristol, D. Fattaccioli
We reproduce, using scaled experiments in a water tank, the effects of scattering phenomena
responsible for the degradations of sonar system performances in oceanic environment (typically,
the small sound speed fluctuations associated with linear internal waves). We reproduce a wide
panel of scattering effects, spanning from “simple” phase aberrations up to radical changes in the
sound field structure (appearance of caustics).
An experimental protocol was developed. It consists in transmitting a high-frequency wave train
(ultrasonic pressure field around 1MHz) through wax lenses with randomly rough faces, that induce
distortions comparable to those that would be observed at sea at around 1kHz in the case of a
lower frequency acoustic signal travelling through a linear internal wave field. Using a 3-D printer,
we were able to manufacture lenses with a randomly rough face. Several lenses were realized, each
characterized by the amplitude and vertical and horizontal correlation lengths of the random face
roughness.
The dependence of the various parameters involved in the experiment (related to the object,
distance of propagation, frequency, …) were studied using simulation programs allowing to
measure the average number of eigen rays and the phase difference between the extreme micro
paths. Those two quantities are useful to compare our results to what was obtained in the
literature, in particular to Flatte’s dimensionless analysis (see companion paper in session
“Modeling Sonar Performance in Uncertain Environments”).
The propagation through the lenses was then studied in a water tank using virtual arrays
(automatic displacements of a hydrophone). We represent the results using the acoustic envelop in
order to observe wave front distortions or appearance of caustics. Measurements of the coherence
function and, hence, of the radius of coherence, are carried out. Finally, we observe degradation of
the performances of a localization algorithm.
Acoustic Echo Reduction and Insertion Loss of Tiles
Shan Victor Pereira, D D Ebenezer, S K Bhattacharyya
Presenter: Shan Victor Pereira
A method is presented to determine the Echo Reduction (ER) and Insertion Loss (IL) of multilayer
planar panels with very large (infinite) lateral dimensions by doing measurements on panels with
168
UA2014
small lateral dimensions. The method is of interest because panels or tiles are often used to cover
areas with dimensions that are large with respect to acoustic wavelengths but made in large
numbers in small sizes. In the method, a designed signal is fed to a power amplifier that drives an
electroacoustic projector. The resulting acoustic waves are incident on a multilayer planar panel.
The time-dependent incident, reflected, and transmitted pressures are measured using
hydrophones that are close to the panel and on either side of it. As a result of the design, the support
of the incident pressure is compact in time and the response of the hydrophone in front of the panel
to the incident wave ends before the wave reflected by the panel reaches it. Further, the response
of the hydrophone to the reflected wave ends before waves diffracted by the edges of the panel or
other parts of the tank reach it. Similarly, the response of the hydrophone behind the panel to the
transmitted wave ends before other disturbances reach it. The signal is designed by first measuring
the transfer function of the power amplifier, electroacoustic projector, and the hydrophone. Timedomain measurements and the Fourier transform are used for this. Then, the input signal that will
result in a broadband short-duration output from the hydrophone is determined and used. The ER
and IL of a metal plate measured using this approach are compared with theoretical values for an
infinite plate. It is shown that accurate results are obtained even when the normalized frequency,
ka, is ? 1 where k is the acoustic wavenumber in water and a is the smallest lateral dimension of
the panel.
A high intensity pulsed laser as a wide band acoustic source for underwater
acoustic applications
J.-P. Sessarego, R. Guillermin, A. Jarnac, A. Houard, Y. Brelet, J. Carbonnel, Y.B. Andre, A.
Mysyrowicz, D. Fattaccioli
In this paper we describe an experiment which was designed to generate an acoustic source using
a high energy pulsed laser. The final goal of this work was to study the possibility of using such a
laser as a wide band acoustic source for underwater applications. The first objective of this study
was to investigate the physical characteristics of the acoustic pulse. This was achieved by
performing acoustical measurements in a large water tank equipped with hydrophones which can
be displaced along X, Y and Z directions by a positioning system controlled by computer (step
increment: 0.1 mm). For this experiment a Ti: Saphire Chirped-Pulse Amplification laser which can
deliver 300 mJ and 50 fs pulses at 800 nm was used. The initial laser beam was deflected in order
to penetrate at normal incidence in water. The laser pulses were focused at different depths in the
water tank (from 20 cm to 50 cm). The frequency range of the acoustic pulse was measured using
several hydrophones covering all together a very wide frequency band [50 kHz-20MHz]. The
directivity of the acoustic pulse in the plane containing the laser propagation axis and for different
conditions of focalization of the laser beam was measured.
In a second step we have investigated in a small water tank what were the parameters affecting
the efficiency of the electro-optic to acoustic conversion. In order to get better light propagation
conditions in water we have changed the laser wavelength (400 nm).This was done by using of a
169
UA2014
KDP cristal (for second harmonic generation). We also successively studied the influence of optical
pulse length and total laser pulse energy on the level of the received acoustical signals. In this
experiment three hydrophones were used, covering all together the frequency band [0-20 MHz].
Finally, we give some conclusions on the applicability of this technique to generating sound signals
in the ocean.
170
UA2014
Session 24: Three-dimensional sound propagation models
Session 24a
Location: Lecture Room D, Tuesday 24th June 2014, 8:30 -10:30
Three-dimensional ray modelling of high-frequency under-ice shallow-water
sound propagation
Sven Ivansson, Ilkka Karasalo, Erland Sangfelt
Presenter: Sven Ivansson
An existing 3-D ray model, REV3D, is extended to allow modelling of high-frequency sound
propagation in the presence of an ice cover. The extension includes computation of propagation
loss and propagation time series, by incoherent as well as coherent summation of ray contributions.
Ice thickness data are provided at the node points of a rectangular grid, with bilinear interpolation
in between, and a rough ice sheet with keels can readily be modelled. So far, the upper ice interface
is flat without ridges. An option is included to construct realizations of poorly known ice thickness
grid data stochastically. Elliptically shaped ice floes without keels are first laid out randomly. Ice
keels, shaped as ellipsoidal bowls with random thickness increments, are subsequently laid out
between the floes. When a ray from a sound source in the water reaches the water-ice interface, it
can be reflected back into the water or refracted into the ice as a compressional wave or as a shear
wave. In order not to get a dramatic increase of rays to follow, a random choice can be made guided
by the expected energy distribution among the different possibilities. Similar random choices are
made when a ray reaches the upper ice interface or the ice-water interface from above. When the
number of rays is increased, the total propagated energy is shown to converge to its correct value,
but random contributions may remain in the coherently computed time series at a particular
receiver. With a flat ice-water interface without keels, rays from the water are typically totally
reflected. With ice keels, ice reflections may cause ray steepening, and rays are more easily
transmitted into the ice, particularly as shear waves. As a result, additional energy losses are
suffered. Computational results are compared to experimental 10 and 20 kHz propagation loss data
from the Gulf of Bothnia, for a propagation range of 5 km.
Numerical applications of a higher order square-root Helmholtz operator
splitting method on modeling three-dimensional sound propagation
Ying-Tsong Lin
Presenter: Ying-Tsong Lin
A higher order numerical algorithm has recently been proposed to split the square-root Helmholtz
operator employed by the parabolic-equation (PE) method for modeling sound propagation. This
operator splitting method includes multidimensional cross terms to yield a more accurate
approximation, and, most importantly, it still permits efficient three-dimensional (3-D) PE
UA2014
numerical solvers, such as the Split-Step Fourier method and the Alternative Direction Implicit (ADI)
Pade method. This talk will first review the numerical applications of this splitting method on
forward models, and the importance of the cross terms in reducing approximation errors will be
addressed. Numerical benchmark of a 3-D wedge problem will be shown to demonstrate the model
performance, along with other computational examples of underwater sound propagation in
internal wave fields, submarine canyons and seamounts. Development of 3-D tangent linear models
and sound field sensitivity kernels will then be introduced with examples considering complex
environmental conditions. This square-root Helmholtz operator splitting method is readily
applicable to modeling sound scattering from rough sea surface, and a numerical implementation
employing the ADI Pade method is introduced. Lastly, suggestions for setting up 3-D benchmark
problems to test accuracy and efficiency of numerical models will be provided. [Work supported by
the Office of Naval Research, USA]
Numerically exact 3D propagation
Ahmad T. Abawi
Solving propagation in a three-dimensional environment is one of the most challenging problems
in computational physics. The problem stems from the large size of the computational domain,
which needs to be discretized at a fraction of the wavelength for the system of governing equations
to yield a numerically accurate solution. For cases where the environment possesses some kind of
symmetry be it rotational or translational, the propagation problem can be simplified considerably.
It can be shown that in problems where the environment is translationally invariant, the 3D wave
equation can be Fourier transformed along the direction of translational symmetry to reduce it to
a 2D equation for each spectral component. The 3D solution can be obtained by solving the 2D
wave equation for each spectral component and performing the inverse Fourier transform. In this
paper we use the above technique to compute propagation in an ideal and a penetrable wedge.
For the ideal wedge, the pressure-release boundary condition is applied to both boundaries and for
the penetrable wedge, the pressure-release boundary condition is applied to the horizontal surface
and the continuity of pressure and normal velocity is imposed on the sloped interface. To obtain a
numerically exact solution, we use the virtual source technique to solve the 2D problem for each
spectral component. As an example of a full 3D problem, we use the boundary element technique
to compute propagation around a conical seamount that extends to the ocean surface.
An explicit analytical solution for the problem of adiabatic sound propagation
along an underwater canyon with penetrable bottom
Pavel Petrov
Presenter: Pavel Petrov
The problem of 3D sound propagation in a shallow-water waveguide featuring a canyon-type
bottom inhomogeneity is considered. In the adiabatic case when the canyon depth is sufficiently
small this problem may be solved using the mode parabolic equations. For the acoustic track
172
UA2014
directed along the canyon the analytical expressions for the mode amplitudes are obtained from
these equations in terms of eigenfunctions determined by the shape of the canyon. This solution
allows us to answer the question whether the sound will undergo usual cylindrical spreading of
become partially trapped inside the canyon for a given shape of the latter. It may be also useful for
the benchmarking of the 3D sound propagation models.
Data driven three-dimensional modeling of shallow water waveguide during
broadband acoustic propagation in the presence of internal waves
During the past two decades attention has been devoted to the propagation of broadband acoustic
signals in shallow water environments. During this time, advances have been made in assessing the
effect of internal waves (IW) on the spatial and temporal acoustic intensity fluctuations. However,
to understand the spatiotemporal effects that a time-varying environment asserts on the
propagation of broadband signals, one needs to consider realistic three-dimensional (3D)
environmental input to acoustic models. In this paper, (1) we extract the IW parameters from
experimental data obtained during shallow water 2006 experiment where 3D effects are taken into
account; (2) we utilize the reconstructed IW field as the 3D environmental input to predictive
acoustic propagation models; (3) we use a combined 3D parabolic equation (PE) model with
Horizontal Rays and Vertical Modes models to check the effects of the curvature of the IW fronts
on the 3D acoustic propagation. The efficiency of the combined modeling exercise is discussed in
the context of the potential acoustic data-model comparison.
Three-dimensional Split-Step Pade Modelling (Peregrine)
Peregrine is the OASIS recoding of the RAM Split-Step Pade Parabolic Equation developed in the
90’s by Michael Collins (Collins, JASA 1993). Peregrine accesses the environment via efficient
memory access of 4D binary database files and is well suited for large-scale Nx2D propagation and
3D propagation environments. In this paper we present the 3D version of the model, which
alternately applies the propagation algorithm in range and azimuth in cylindrical coordinates. The
model has been applied to basin scale propagation at 8 Hz, including Perth-Bermuda and the longrange observation of an underwater volcano. Concerns about the phase stability of the algorithm
are addressed in the comparison of the Peregrine solution with the benchmark solution for the ASA
absorbing wedge problem.
173
UA2014
Session 24b
Location: Lecture Room D, Tuesday 24th June 2014, 10:45 - 12:05
Mode parabolic equations with mode interaction for the 3D modeling of sound
propagation
M.Yu. Trofimov, S.B. Kositskiy, A.D. Zakharenko
Presenter: M.Yu. Trofimov
The mode parabolic equations with the mode interaction are derived via the method of multiplescale expansion. These equations take into account all 3D features of shallow water acoustic
environment. The method is validated in both 2D and 3D benchmark problems. In a 3D benchmark
problem our solution is compared to that of Harrison (Harrison, 1992, JASA 93(4)) and they are
shown to agree well. A 2D penetrable wedge ASA benchmark is also used to demonstrate the
accuracy of mode interaction modeling in our method.
A coastal wedge propagation model including shear in an absorptive bottom
Piotr Borejko
Presenter: Piotr Borejko
A representative model of the propagation of underwater sound on a coastal wedge is that of an
acoustic wave in a wedge-shaped layer of fluid over a penetrable bottom (penetrable wedge).
Acoustic propagation in penetrable wedge model has been analyzed by various methods since the
early 1980s, but there have been a few observations of the predicted three dimensional (3-D)
propagation effects. It was not until 2007 that the 3-D sound was quantitatively measured in a pair
of acoustic transmission tests on the Florida shelf; and then modeled by applying highly accurate
3-D hybrid and adiabatic-mode based approaches.
In this paper, the method of generalized ray is applied to acoustic propagation calculations in a 3
[deg] wedge of water over an exposed low-loss limestone bottom, in which, like on the Florida shelf,
the shear wave speed is slightly less than the speed of sound in the water. The predicted 3-D crossslope propagation effect is that some acoustic signals coming in along paths of multiple bottom
interactions (in-shore refracted paths, propagating up the slope and back to the receiver) may
substantially be stronger than those coming in along paths of only a few bottom interactions (direct
paths, traveling near the straight source-to-receiver path that is parallel to the wedge apex). This
theoretical prediction is consistent with observations of two distinct signal arrivals in the abovementioned tests: one signal coming in along a direct path; and the other, sometimes substantially
stronger than the direct arrival, coming in later along an in-shore refracted path.
174
UA2014
Incorporating cross-multiplied terms in a three-dimensional parabolic
equation model
Frederic Sturm
Presenter: Frédéric Sturm
Wide-angle parabolic equation (PE) based models are efficient and accurate tools for solving sound
wave propagation problems in three-dimensional oceanic waveguides. For practical (or historical)
reasons, most of them neglect cross-multiplied operator terms that appear naturally in the wideangle square root operator approximation of the Helmholtz equation. It has been shown recently
both numerically and theoretically, that the use of a series of higher-order cross terms allows a
reduction of the phase errors inherent to any PE computations and can thus handle greater
propagation angles. The cross terms were efficiently incorporated in a split-step Pade 3D PE
algorithm. The objective of the present paper is to report some numerical results obtained
incorporating a leading-order cross-term correction in an existing 3-D PE model, written in
cylindrical co-ordinates, based on Pade approximations in both depth and azimuth, and a splitting
operator technique. Various aspects of the numerical techniques used to handle the additional
leading-order cross term are discussed. The improved accuracy of the now-fully wide-angle 3D PE
model is assessed on several benchmark numerical solutions and also tested on experimental data
obtained in a small scale tank.
Benchmarking a Three-Dimensional Gaussian Beam Tracing Model
Michael B. Porter
Presenter: Michael B. Porter
Beam tracing methods have become widely used in ocean acoustics applications. Like parabolic
equation models, they can treat range-dependent and fully 3D scenarios. Beam tracing methods
are particularly attractive at higher frequencies, or for broadband sources where other models
become computationally impractical. Full time-series calculations are easily done in the beam
tracing framework by simply summing the echoes. Thus, they are widely used for modeling acoustic
modems, passive acoustic monitoring of marine mammals, and active sonar. This presentation will
describe BELLHOP3D, which is the fully 3D extension of the BELLHOP beam tracing model. We will
focus particularly on the benchmarking of the 3D results using a variety of test cases.
175
UA2014
Session 25: Towards Automatic Target Recognition: Detection, Classification and
Modeling (of Underwater Targets)
Session 25a
Location: Lecture Room D, Thursday 26th June 2014, 8:30 - 10:30
Iterative Target Recognition for Port Protection System
Mohamad Nakcha, Einas Alhaji, Mohamd Alcahaita
Presenter: Mohamad Nakcha, Einas Alhaji
Coastal areas’ protection is an important aspect in providing security for sea side establishments of
high economic importance and/or military bases. The rise in underwater intrusion technologies,
promotes terrorists to use unmanned vehicles in their attacks. A monitoring system approach
discussed in this paper, considering the site investigation, using combining equipment types such
as sonar, hydrophones, communication tools, multithreading software programs and so on, to get
the maximum available data, which will be analyzed then sent to an appropriate processing part
to detect any intruders’ threat. Design simulation was developed and tested. Hardware prototype
was built with zero software versions, the trial result is shown. After trial, the current hardwaresoftware version will be modified to produce a first version of our proposal system, which will
correct some expected misleading results because of iterative method.
The Main Peculiarities of Automatic Target Recognition
Andrei Mashoshin
Presenter: Andrei Mashoshin
ATR is one of the most challenging problems of applied underwater acoustics.
Underwater targets are classified using the target signature containing target features (TFs), i.e.
scalar or vector parameters of the target signal or echoes, which possesses information about the
target class and can be measured at the sonar receiver output.
The purpose of the paper is to discuss the main peculiarities of the ATR problem and to propose a
general approach to the synthesis of optimal ATR algorithms which can be applied to both passive
and active sonars.
The main peculiarities of the ATR problem are the following:
o ATR is based on 3 types of TFs caused by the peculiarities of:
o sound generation (or sound reflection);
o underwater sound propagation;
o target behavior;
o since the ATR is fulfilled as a rule at small signal-to-noise ratios (SNR) the TFs do not possess
much useful information and that is why the only way to reach high effectiveness of ATR is correct
usage of all measured TFs;
UA2014
o the information concentrated practically in each TFs parameter depends on the current hydro
acoustic conditions and SNR, that is why ATR algorithms must be adaptive;
o practically all TFs are mutually dependent, which dictates the need to base the ATR algorithms
on mutual probability distribution function (PDF) of TFs;
o different TFs are measured during different time intervals, that is why ATR algorithms must be
dynamic.
The experience shows that the lack of knowledge about the above-mentioned peculiarities does not
allow the ATR problem to be solved with the required quality. For example, the algorithms which
do not consider current hydro acoustic conditions and also those in which the classification
decisions are delivered with the use of separate TFs with subsequent weight summing of these
decisions are ineffective.
The paper contains a procedure for substantiation of the synthesis of an optimal ATR algorithm. It
is shown, that an optimal ATR algorithm must be Bayesian for the case of equal hypothesis.
Taking in account the fact that the main procedure of the Bayesian algorithm is calculation of the
mutual conditional PDF of TFs, we show how to calculate this PDF in the case of mutually dependent
TFs.
A simple but important example is given to illustrate the use of an optimal ATR algorithm for
classification of submarines and surface ships in the passive sonar.
Independent views in MIMO sonar systems
Yan Pailhas, Yvan Petillot
The main advantages of MIMO (Multiple Input Multiple Output) sonar systems come from the
assumption of independent observations between each transmitter/receiver pairs.The
independence of the observations ensures a unbiased set of measurements and then provides true
statistics on the target. In this paper we study the correlation between views in MIMO sonar
systems. A traditional tool used to study the dependency between two random variables is the
Pearson product-moment correlation coefficient. However this measure suffers numerous defaults:
it only estimates linear correlation, it is not a proper distance and in particular a null measure of
the Pearson coefficient does not insure the independence of the tested random variables. For these
reasons we will use the distance correlation introduced by Szekely. From the distance correlation
we will derive the inter-views distance correlation matrix which assess the correlation of the full
MIMO system (i.e. the dependencies between each views). This independence measure matrix gives
a guideline to how to build truly uncorrelated MIMO sonar systems and then maximise the
performances of such system.
177
UA2014
Automatic classification for mid-frequency anti-submarine warfare sonars –
recognizing pipelines
Karl Thomas Hjelmervik
Presenter: Karl Thomas Hjelmervik
The Norwegian Trench is practically littered with wrecks, pipelines, and underwater terrain
features, all capable of generating false contacts in anti-submarine warfare (ASW) sonars. For
modern sonar systems, the sonar operator may easily discard such false alarms simply by
overlaying the detections on a chart containing the positions of pipelines and wrecks. However,
autonomous systems lack human perception and intelligence, and therefore require failsafe
algorithms to filter out false contacts.
Here we study the theoretical behavior of echoes from pipelines and also derive an algorithm for
classifying a given detection as originating from a pipeline or not. The method is applied on track
level data and exploits the kinematical properties of tracks to estimate a probability that a given
track originates from a given pipeline. A modification to the algorithm that allows for accurate
classification in unchartered waters is also suggested.
The proposed method is tested on a data set called the Clutter Experiment 2002 (CEX02). This
experiment was carried out in the NAT3 programme, a collaboration between Thales Underwater
Systems, TNO, FFI, and the Dutch, French, and Norwegian navies, with the intent to experiment
with and evaluate different clutter-reducing techniques. The area of the Norwegian Trench where
the test was carried out is particularly difficult with regards to false alarms, due to strong upslopes
and the presence of both pipelines and seamounts.
Acoustic obstacle detection for safe AUV surfacing
Imen Karoui, Isabelle Quidu, Michel Legris
Presenter: Isabelle Quidu
Automatic sea surface detection is necessary for securing autonomous underwater vehicle (AUV)
surfacing. However, few studies addressed this issue despite several accidents reported in the last
decade.
Here we propose a pattern analysis based scheme for automatic detection of sea surface objects
by processing forward looking sonar images. The considered sea surface obstacles are man-made
: buoys, drifting containers, boats, vessels (motorboats or sailboats). According to the object type
and state (moving or fixed), the surface acoustic signature in sonar images can be:
- Strong intensity beams due to stationary self noise from ships. This signature indicates the ship
bearing and is generally the strongest among all surface object signatures and the easiest to detect;
- A high contrasted intensity feature in case of noise-free objects like buoys, sailboats, not moving
ships such as fishing vessels;
- Some high intensity lines due to the wake behind a moving vehicle.
A hierarchical detection scheme is proposed in order to manage these various target signatures.
The first step consists in detecting stationary ship noise. In case of detection, the strong-intensity
178
UA2014
strip corresponding to the ship bearing is removed to avoid ship noise disturbance during other
target detection processes. The next step consists in detecting the other types of obstacles. It is
based on an adaptive CFAR (Constant False Alarm Rate) thresholding. The final step consists in
analyzing the area around every detected position in order to state that this latter is a reliable
obstacle and not a wake signature.
Promising results are obtained using real data collected at sea with various objects and scenarios.
Tracking underwater objects using large MIMO sonar systems
Yan Pailhas, Emmanuel Delande, Jeremie Houssineau, Yvan Petillot, Daniel Clark
MIMO sonar systems can offer great capabilities for area surveillance especially in very shallow
water with heavy cluttered environment. We present here a MIMO simulator which can compute
synthetic raw data for any transmitter/receiver pair in multipath and cluttered environment.
Synthetic moving targets such as boats or AUVs can also be introduced into the environment. For
the harbour surveillance problem we are interested in tracking all moving objects in a particular
area. So far the tracking filter of choice for multistatic systems has been the MHT (Multiple
Hypothesis Tracker). The reason behind this choice is its capability to propagate track identities at
each iteration. The MHT is an extension of a mono object tracker to a multi object problem and
therefore suffers from a number of drawbacks: the number of targets should be known and the
birth or death of new tracks are based on heuristics. A fine ad hoc parameter tuning is then required
and there is a lack of adaptivity in this process. To overcome those restrictions we will be using the
HISP (Hypothesised multi-object filter for Independent Stochastic Population) filter recently
developed by [1]. The HISP filter relies on a generalisation of the concept of point process that
integrates a representation of distinguishability. As a consequence, this filter deals directly with the
multi-object estimation problem, while maintaining track identities through time without using
heuristics. While filters track the objects after processing in the digital domain, we show as well in
this paper that we can adapt acoustical time reversal techniques to track an underwater target
directly with the MIMO system. We will show that the proposed modified DORT technique matches
the prediction / data update steps of a tracking filter.
[1] Houssineau, Del Moral, Clark. "General multi-object filtering and association measure." IEEECAMSAP 2013.
179
UA2014
Session 25b
An information-based Johnson’s criteria model for UUV system performance
prediction
Richard Brothers
Presenter: Richard Brothers
The performance of traditional, ship-mounted, mine hunting sonars have been typically modeled
using detection theory. However, the move toward unmanned, offboard systems for minehunting
in recent years has afforded the opportunity to employ high resolution, imaging sonars to
characterise targets. The sonar imagery generated by these systems presents information not
readily dealt with using detection theory alone.
This paper presents an information-based Johnson’s criteria model, adapted from the electrooptical domain for modeling the performance of unmanned system imaging sonars. The model
calculates the information content of intensity (contrast) and shape contributed by both contact
highlights and shadows, using their combined value to estimate the probability of target
recognition.
Performance prediction using the adapted model is verified against experimental data from a
number of unmanned system sonars and target characterization by both human and automated
recognition processing.
Application of the model to the performance of dual frequency SAS is investigated to demonstrate
how the information content available to target characterization processing changes with
operational frequency. This understanding is then used to help initial design of a target
characterization process that exploits the combined information of both low and high frequency
SAS.
Quantifying the complexity in sonar images for MCM performance estimation
Marc Geilhufe, Oivind Midtgaard
Presenter: Marc Geilhufe
Seafloor characteristics like roughness and clutter density affect the complexity in sonar images,
significantly influencing the achievable performance for detection and classification of bottom
targets, e.g. mines. The introduction of autonomous underwater vehicles (AUV) into mine
countermeasures (MCM) operations has thus created a need to measure such properties in imagery
from high-resolution side-looking sonar (real or synthetic aperture) using a consistent metric. In this
paper, we present our approach for quantifying the complexity in sonar images with a continuous,
multi-scale measure based on local variances of wavelet coefficients obtained by the twodimensional maximal overlap discrete wavelet transform (MODWT). Moreover, we determine the
presence of anisotropy, i.e. image textures with a clear directionality like sand ripples, through a
combination of ratios from wavelet variances and rotated integral images. The proposed method
180
UA2014
is applied to synthetic aperture sonar (SAS) images, where promising results are achieved both in
terms of validity and computational time. We see image complexity as an essential parameter for
MCM performance estimation. Further it can also be used to improve the performance of automatic
target recognition (ATR) by adapting the processing to the given environment.
Towards Automatic Target Recognition in Low-Frequency Sub-Sediment Sonar
Imagery
A.J. Hunter, R. Van Vossen, A.L.D. Beckers
Detection of naval mines, unexploded ordnance, and improvised explosive devices is challenging in
the underwater environment, particularly when object burial occurs. A capability to detect buried
targets has been demonstrated previously using TNO’s MUD low frequency sediment-penetrating
sonar and other similar sonars. However, the high clutter rates encountered in practice have the
potential to impose severe operational limitations in absence of a robust capability to distinguish
targets from clutter. To this end, we are taking the initial steps towards development of an
automatic target recognition algorithm for detecting targets and suppressing clutter in lowfrequency sub-sediment sonar imagery. The initial implementation presented in this paper uses a
previously developed wavelet shrinkage algorithm to suppress the background reverberation,
followed by automatic thresholding and segmentation to isolate individual seafloor objects for
subsequent extraction of their acoustic signatures. Using the MUD-2011 data set, we show
preliminary results from the detector and explore possible features of the acoustic response that
could be used for classification.
A GPU Sonar Simulator for Automatic Target Recognition
Jo Inge Buskenes, Jon Petter Asen, Herman Midelfart, Oyvind Midtgard
Presenter: Jo Inge Buskenes
Template matching is a common technique used when classifying objects in synthetic aperture
sonar (SAS) images. The principle is to isolate an image segment containing an object of interest,
correlate it with a set of template images, and assign it to the class of the template yielding the
highest correlation coefficient. The challenge is to come up with a suitable set of template images
considering that no seabed or object is truly alike.
We target this challenge with a sonar simulator that first take as input a seabed model derived
from the real sonar image. Then it places an object model on the seabed, renders the scene, and
adds the resulting image to the template set. For any object position, alignment, type and material,
the procedure is repeated, and a correlation coefficient computed. The faster we are able to
perform these simulations, the better we can expect the classification result to be. Therefore the
simulator is written in OpenGL and OpenCL and run on graphics processing units (GPUs).
181
UA2014
The result is a fast performing, mobile and portable on-the-fly template generator which can adapt
its behavior to the nature of the current scene. We believe this can prove a powerful tool for mobile
sonar imaging platforms such as autonomous underwater vehicles (AUVs).
Efficient Superellipse Fitting based Contour Extraction for mine-like shape
Recognition
Daniel Kohntopp, Benjamin Lehmann, Dieter Kraus
Presenter: Daniel Köhntopp
For the classification of mine-like objects in sonar images, the contour of its shadow and the
features derived from it play an important role. Therefore, the extracted contour needs to be as
good as possible, i.e. undisturbed by noise and artifacts. In this paper superellipse fitting is
presented as a way to enhance the extracted contour after segmentation. A new linearization of
the fitting task is introduced to speed up the computation time. Moreover, linear tapering and
circular bending are considered in the context of fitting more complex contours to the shadow of
mine-like objects. In the experimental investigations several different methods of fitting are
compared with regard to the resulting fitting error and computation speed. Results suggest that
the linearization indeed speeds up the computation by a factor of 2 up to 4 without any of the fitting
accuracy.
Identifying Contents of Low Profile Targets in a Cluttered Environment
Yan Pailhas, Keith Brown, Chris Capus, Nicolas Valeyrie
In the mine countermeasures context, image-based automatic target recognition is confronted by
two major limitations. Firstly, in cluttered or heavily cluttered environments, recognition algorithms
are plagued by a dramatic increase in PFa (probability of false alarm) making output of meaningful
results impractical at best. The second main limitation comes from previously unseen threats such
as IEDs (Improvised Explosive Devices) or simply from unknown types of mines.
In this paper we present classification results from trials in Portland Harbour, UK in October 2012.
The aim of the trials was to identify the contents of a set of test targets from wideband acoustic
responses alone. The test set comprised nine identical gas cylinders (65 cm height, 30 cm diameter):
three cylinders were filled with water, three with sand and three with gravel. Sonar data was
acquired for these targets on two different and reflective seabed types. The estimated SNR for these
low profile targets in these environments is negative making them almost undetectable in
traditional sidescan sonar imagery. Here, we demonstrate the capability of the BioSonar to identify
the cylinders and distinguish between the 3 types of contents using in-situ learning from spiral
reacquisition patterns and show that the wideband classification is robust to certain environmental
variations.
182
UA2014
Session 26: Underwater Acoustic Measurement Facilities and Standards
Session 26a
The Evolution of Acoustic Transducer Calibration Underwater Sound Reference
Division (USRD)
Anthony Paolero, Victor Evora
Presenter: Anthony Paolero
The U.S. Navy maintains acoustic standards for measurement to certify performance parameters
of systems and materials to ensure consistent and accurate measurements for research,
development, inspection, acceptance and maintenance of ships, aircraft, and other systems. The U.
S. National Institute of Standards and Technology (NIST) perform the governmental function of
providing standards in these areas, except in the area underwater sound. The USRD fills this gap by
serving as the U.S. recognized NIST-deferred activity for underwater acoustic metrology, with
traceability to NIST. Historically established in Orlando, Florida in the early 1940s as the Underwater
Sound Reference Laboratory to perform underwater acoustic measurements in support of the U.S.
Navy, the USRD today provides acoustic transducer calibration services and evaluation of Sonar
transducers to national and international navies, universities, and private industry. The USRD also
participates in Consultative Committee for Acoustics, Ultrasound and Vibrations (CCAUV) Key
Comparisons for international Round-Robin calibration of hydrophones. To support these critical
roles, the USRD maintains several world-class facilities, including the Acoustic Pressure Tank Facility
(APTF), Open Tank Facility (OTF), Low Frequency (LOFAC), and Leesburg Facility (LEFAC).
Additionally, it maintains the U.S. only Transducer Standard Loan Program which develops,
manufactures, and provides underwater standards to the national and international acoustic
communities. The presentation will cover a historical overview of the laboratory, capabilities of
measurement facilities, and the latest in calibration measurement methodologies for primary and
secondary (loan) standards
Provision Of Standards At Simulated Ocean Conditions
G A Beamiss, S P Robinson, Gary Hayman
Presenter: G A Beamiss
The Acoustic Pressure Vessel (APV) has been established at NPL since 2000 and is used for
characterizing underwater acoustic devices, determining the acoustic properties of viscoelastic
materials and conducting integrity testing of artifacts while at simulated ocean conditions. The APV
enables acoustic measurements to be undertaken at simulated ocean conditions: over a
temperature range of 2 ?C to 35 ?C, and at hydrostatic pressures of up to 6.9 MPa.
183
UA2014
This paper describes the APV facility and its operating capability, along with how underwater
acoustic standards are provided. This includes how primary standard calibrations are realized, and
how these are disseminated via comparison calibrations. The determination of the coefficients for
transducer sensitivity variation as a function of temperature and pressure are an important factor
in the dissemination process, both for measurements made in the APV and at the NPL open water
facility (where temperature varies with the seasons). The methods adopted are described along
with the influence of a number of factors which contribute to the uncertainty.
Recent advances in methods for the calibration of linear hydrophone arrays at
low frequency
Steven E. Crocker
Comparison calibrations of hydrophone data channels in linear arrays are challenged by a variety
of factors. First, calibration of an array may require the simultaneous measurement of complex
sensitivity (e.g., magnitude and phase) for data channels numbering in the hundreds and
distributed over an extended aperture length. As a consequence of high channel counts and long
electrical transmission paths, the acoustic data are frequently digitized within the array and
transmitted over a telemetry system that cannot easily be synchronized with calibrated reference
standard hydrophone data collected by commercial data acquisition systems. In addition,
calibrations at low frequency are hampered by an inability to use pulsed waveform techniques to
approximate free field propagation due to the arrival of boundary reflections before sufficient
reflection-free data can be acquired. Finally, the minimum calibration frequency may be dictated
by the operating bandwidth of the acoustic projector used to transmit the calibration waveforms.
The Naval Undersea Warfare Center has developed methods to measure complex sensitivity in the
data channels of hydrophone line arrays with digital telemetry systems. These methods have
improved the quality of low frequency calibration data, while maintaining efficiency commensurate
with a production test environment. A requirement for calibration at arbitrarily low frequency is
that the equivalent electronic noise levels in the data channels are six to ten (or more) decibels
below sea state zero; a requirement that is frequently satisfied in arrays designed for scientific,
geophysical and naval applications.
The method of complex free-field calibration of a pressure gradient receiver
Alexander Isaev, Anton Matveev
Presenter: Mr. Anton Matveev
Absolute free-field calibration of a pressure gradient receiver (PGR) is considered. The calibration is
carried out using reciprocity method procedure during radiation of continuous linear frequency
modulation signals in a water tank with reflecting boundaries. To obtain free-field frequency
dependences the technique of complex moving weighted averaging (CMWA) is used.
The phase frequency response of PGR is obtained by including to the calibration procedure
measurements phase angle of transfer impedance of the projector-receiver pair. To increase
184
UA2014
accuracy of phase frequency response measurements the position of the PGR acoustic center is
defined by its displacement relative to PGR geometrical center.
Proposed method was used to calibrate at a frequency range 500 Hz– 12,5 kHz combined PGR which
has three receive channels for vector quantity of acoustic field (pressure gradient) and one scalar
channel (sound pressure).
The successful results of absolute complex free-field calibration of combined receiver allow to use
phase closeness of instantaneous values of sound pressure and oscillation velocity in the same point
of sound field as a plane wave criterion instead of traditional criterion based on 1/r law.
Analysis of results, advantages and new possibilities of proposed way for PGR absolute free-field
calibration are discussed.
185
UA2014
Session 26b
Underwater techniques to characterize the near scattered acoustic vector field
Robert J. Barton Iii, Georges A. Dossot, Kevin B. Smith
Presenter: Robert Barton
In this study, we investigate the properties of the scattered acoustic vector fields generated by
simple geometric objects, including the infinite rigid plate, disk, and sphere. Analytical solutions are
derived from acoustic target strength scattering models in the near?field region. Of particular
interest is the understanding of the characteristics of energy flow of the scattered acoustic vector
field in the near? to far?field transition region. We utilize the time and space separable
instantaneous active and reactive acoustic intensities to investigate the relative phase properties
of the scattered field. Numerical results are presented for the near region scattered acoustic vector
field of simple objects in both two and three dimensions. Previous in-air measurements are
summarized, and an approach to taking water-borne measurements is offered.
Calibration of hydrophones in the frequency range 1 kHz to 200 kHz using
optical method
Shiquan Wang, Yi Chen, Yongjun Huang, Yuebing Wang
Presenter: Shiquan Wang
A technique for absolute calibration of hydrophones using the optical method is described. In this
method, the acoustic particle velocity is obtained at the air-water interface without a reflective
pellicle using a commercial laser doppler vibrometer. The measurement principle is introduced and
a calibration facility is set up. A B&K8105 hydrophone is calibrated over the frequency range from
1 kHz to 200 kHz using the calibration system. The result is compared with that of three-transducer
reciprocity method and good agreement is shown. The measurement scheme eliminates the
acousto-optic effect and pellicle mounting. The calibration is convenient to carry out.
The design of acoustic absorbers for test tank linings
To provide good free-field region or reduce reverberation in test tank, it is essential to use acoustic
absorbers as linings. Generally, these absorbers are made from polymer materials, which are
expensive for most users.
In this paper, a new type of absorbers is presented, and its performance is analyzed and measured.
Made of porous materials and in shapes of wedge, these absorbers are suitable for applications as
test tank linings in the frequency range from 10 kHz to 200 kHz.
186
UA2014
Calibration methods of the interferometric fiber-optic hydrophone
Yi Chen, Jun Zhang, Min Zhang
Presenter: Yi Chen
The calibration methods of the phase-shifted sensitivity of the interferometric fiber-optic
hydrophones are described and investigated experimentally. At first we introduced the
measurement theories and their calibration facilities of three methods briefly: Bessel Function Ratio
(BFR) method, Fringe Counting (FC) method, and Phased Generator Carrier (PGC) method. Then by
the use of comparison calibration method, the calibration experiments of the phase-shifted
sensitivity of a fiber-optic hydrophone based on the Michelson interferometer are carried on in a
vibrating column vessel and free field respectively. By analyzing the experimental results, it is
demonstrated that the phase-shifted sensitivity of the interferometric fiber-optic hydrophone can
be calibrated correctly, by considering the different practical frequency range of these methods.
The calibration and characterisation of autonomous underwater recorders
Gary Hayman, Stephen P. Robinson, Paul A. Lepper
Presenter: Stephen P. Robinson
Methodologies for the calibration and characterisation of autonomous recorders used for in-situ
measurement of underwater noise are presented. The increasing use of autonomous recorders is
motivated by the need to monitor underwater noise, such as in response to the requirements of the
EU Marine Strategy Framework Directive, and the aim is to provide traceability for underwater
noise monitoring to underpin the protection of the marine environment from anthropogenic noise.
The performance of these systems is a crucial factor governing the quality of the measured data. In
this paper a discussion is presented of measurement methodologies for key acoustic performance
characteristics of the recorders, including the self-noise of the hydrophone and system and the
absolute sensitivity as a function of frequency (including hydrophones, amplifiers and digitisation
system). Consideration is given to effects due to the proximity of the recorder body to the measuring
hydrophone: where the hydrophone is attached close to the recorder body, scattering from the body
can have a significant influence on the frequency and directional response of the overall system.
Examples of the results obtained are given and a discussion is presented of the implications of
system performance on the quality of the measured data.
Long term underwater third octave sound levels at a busy UK port
Joanne K Garrett, Matthew J Witt, Lars Johanning
Presenter: Joanne K Garrett
The sound levels of the third octave bands with centre frequency 63 Hz and 125 Hz will be used as
indicators for the EU Marine Strategy Framework Directive (MSFD) descriptor 11 to monitor low
frequency continuous sound in the marine environment. To explore this, long term underwater
sound data from a busy UK port were investigated. Two autonomous multichannel acoustic
recorders (AMAR Generation 2; Jasco Applied Sciences) recording broadband sound in the effective
187
UA2014
frequency range 10 Hz to 32 or 48 kHz, for half an hour in every hour, have been deployed
alternately at the Falmouth Bay Test site for marine renewable energy devices (FaBTest) off the
south coast of Cornwall, UK from March 2012 to August 2013. Data collected during periods of
wave energy device testing were removed for this analysis. The area supports considerable
commercial shipping and recreational boating along with diverse marine fauna, including
bottlenose dolphins, harbour porpoises and fish. Custom MATLAB scripts were used to derive third
octave levels (TOLs). The mean minute TOLs were found to vary by season with a mean TOL for the
63 Hz band increasing from 95.61 dB re 1 µPa in July 2012 to a mean of 106.04 dB re 1 µPa for
December 2012. The yearly mean TOLs of 100.85 ± 7.94 dB re 1 µPa and 103.48± 6.11 dB re 1 µPa
for the 63 and 125 Hz bands respectively (number of half hour sound files = 6992) were found to
exceed the suggested target of 100 dB re 1 µPa for the period March 2012-March 2013. This
provides information on the current sound levels from which a trend can be monitored at this site.
The empirical data presented here offers an exploration of the proposed MSFD indicator bands in
order to inform future use.
188
UA2014
Session 27: Underwater Acoustic Studies in Asian Seas
Session 27a
Observation of ambient noise induced by the internal solitary wave in the
center of Kuroshio northeast of Taiwan
Yiing-Jang Yang, Wen-Der Liang, Jeff Chih-Hao Wu, Hsian-Chih Chan, Ruey-Chang Wei,
Chi-Fang Chen
Presenter: Yiing-Jang Yang
A subsurface acoustic Doppler current profiler (ADCP) mooring was deployed in summer of 2012 to
monitor the ambient noise and current velocity on the I-Lan Ridge, northeast of Taiwan. The water
depth at the mooring was 275 m. The mooring was close to the northeastward Kuroshio
mainstream. Several internal wave packets were recorded during spring tide periods. When the
wave passed the mooring, the ADCP instrument dived into deep water and temperatures increased.
A southwestward current accelerated with this downwelling as well as the temperature rise, and
then followed by an opposite pattern in the upper ocean; therefore, a mode-1 depression internal
solitary wave packet can be easily identified. The maximum horizontal current perturbation and
vertical amplitude were around 150 cm/s and 40 m, respectively. The mode-1 depression waves
could generate the surface wave breaking in the convergence zone and produce ocean noises. The
ocean ambient noises would be enhanced in the convergence zone and reduced in the divergence
zone due to rough and smooth sea surface conditions, respectively. This study observed the sound
source from surface wave breaking affected the high sonic bands between 100 Hz and 10K Hz. The
maximum noise was around 10 dB increment at 1.5K Hz and coincided with the maximum
convergence (downwelling), which agrees well with the theory. Strong horizontal currents induced
by internal solitary waves would produce self-noise in the infra and low sonic bands below 300 Hz.
Meanwhile, internal solitary waves can change the thermal structure and modify the aspects of the
underwater acoustic signal. Temperature profiles will be required to establish an underwater
ambient noise model for further study.
Some characteristics of bottom scattering provided by single-mode
reverberation
J. Zeng, W.Y. Zhao, D.Y. Peng, H.F. Li , Y.Ge, T.F. Gao , E.C. Shang
Presenter: J. Zeng
The characteristics of scattering due to interface roughness are usually described by the
backscattering matrix (BSM). The BSM based on the Bass perturbation theory has significant
differences from that based on the popular empirical scattering law (Lambert’s law), especially at
low grazing angles. In an experiment with a point source, it is very difficult to extract the
189
UA2014
quantitative characteristics of the BSM at low grazing angles from the experimental data because
of the difficulties in acquiring low-grazing-angle scattering data and separating the scattering data
between different modes (grazing angles). In contrast, the use of single-mode excitations as sources
in shallow-water waveguides enables acquisition of good quality low-grazing-angle scattering
data. In this paper, the characteristics of the BSM were obtained from different single-mode
reverberation experiments in shallow-water. The experiments were carried out at different sites
during different seasons: one was carried out in the south sea of China with water depth about 70
m during the summer of 2012, and the other two were in the Yellow Sea of China with water depths
about 36 m during the summer and winter of 2011, respectively. Model-data comparisons were
made and the results showed that at low grazing angles (20~50), the BSMs based on the Bass
perturbation theory were in good agreement with the experimental data, but the BSMs based on
Lambert’s law were not.
Measured Channel Impulse Responses for a Mobile Source in the Northeastern
Sea off Taiwan
Linus Y.S Chiu, Andrea Chang, Chifang Chen
Presenter: Linus Y.S Chiu
Data communication is of interest in numerous naval and civilian applications. Examples include
communication among autonomous underwater vehicles (AUVs) for collaborative operations,
harbor security systems, tactical surveillance applications, oceanographic data retrieval from
underwater sensors over geographically large areas, offshore oil and gas explorations, etc. While
our understanding of underwater acoustic communications has improved over the last 2 decades,
communications involving mobile assets e.g., Gliders and AUVs_remains quite challenging. First,
the harsh multipath, with delay spreads up to hundreds of symbols for high data rates, and
temporal variations of the underwater acoustic channels, with Doppler spreads up to several tens
of hertz, are major issues in underwater mobile communication. Secondly, underwater acoustic
systems typically operate at low signal-to-noise ratio (SNR) environment and thus require some
form of spatial diversity e.g., array to enhance SNR and mitigate channel fading effects. A point-topoint communication between a mobile source and a single fixed receiver where spatial diversity is
gained from a virtual horizontal array obtained by relative motion between them, is referred to as
synthetic aperture communications (SACs).
In September 2011, the MArine Cable Hosted Observatory (MACHO), administrated by the Central
Weather Bureau, was launched in the northeastern sea off Taiwan for surveillance of submarine
earthquakes, tsunami, and the ocean environment. The MACHO system is the first submarine
observatory for various science research projects in Taiwan and was used to study underwater
acoustic channel in 2012. In May 2012, an acoustic experiment was conducted by using a mobile
source with a bottom mounted hydrophone on the MACHO system. This paper presents preliminary
results of measured underwater acoustic channel impulse response for mid-frequency band of 46kHz in the northeastern sea off Taiwan. The measured channel characteristics are also predicted
by an acoustic numerical model, and the modeling result has good agreement with experiment
190
UA2014
data. Besides, this paper also advances the understanding of the truth of performance of the SACs
operated in a highly range-dependent and challenging channel, a track of canyon, which is with low
SNR at certain locations as the mobile source was moving along the canyon. The data transmitted
by the moving source along a track of canyon, listened by the fixed bottom mounted hydrophone
in the system of marine hosted cable, is analyzed and compared to the more typical channel: the
flat bottom channel. Typical q-functions obtained by passive phase conjugation processing are used
to present and compare the SAC performance. The results shows that the across-canyon track
presents low SNR at certain locations as the mobile source was moving across the canyon, which
will break the communication between mobile sources and receiving stations. However,
communication performance can be improved by spatial diversity, obtained from a virtual
horizontal array generated by the moving transmitter and the fixed receiver, referred to as a
synthetic aperture.
Observe Seismic Activities and Ambient Noise of Underwater Acoustic Data
from MACHO Hydrophone
Yin-Ying Fang, Shih-En Chou, Chifang Chen, Chien-Kang Huang
Presenter: Yin-Ying Fang
Since Taiwan is located in the Circum-Pacific Seismic Zone, there are thousands of earthquakes
which had been detected within one year. In addition, tsunamis, which were caused by
earthquakes, have been suffered parts of areas in Asia-Pacific recently years. For the reason that
Central Weather Bureau (R.O.C.) has built the first submarine cabled observatory to investigate
earthquake, tsunami and ocean environment for monitoring the natural disasters at the
northeastern of Taiwan’s offshore in 2011 - Marine Cable Hosted Observatory (MACHO) system.
The National Science Council of the Republic of China has also sponsored a project “Marine
Observatory in the Northeastern Taiwan (MONET)” to efficiently analyze the large quantity of longtime monitoring raw data from hydrophone sensors in the MACHO system. The detector in this
paper includes time-varying ambient noise level estimation via Leq (equivalent continuous sound
level, averaged over 30 seconds) and estimation theory, and an energy detector with the estimated
ambient noise level as threshold. Two years of data are analyzed with the detector, and the results
are showed seismic activities of local environment, the seasonal variation of the local ambient noise
level in MACHO system and the efficiency of analysis of large amount of data. In addition to this, a
user friendly UI (user interface) is written to enhance its utilization as an useful intermediary for
Central Weather Bureau detecting earthquakes by the underwater acoustics. (This work is
sponsored by National Science Council (R.O.C.), under Project entitled “Physical Oceanography and
Acoustic Applications at the Marine Observatory in the Northeastern Taiwan (MONET) coordination
project and subproject 1: Study of Underwater Acoustic Signature Extraction and its Data Base
Automation" Project No. NSC 101-2221-E-002 -028 -MY2).
191
UA2014
An Overview of Ocean Ambient Noise around Taiwan: Measurement and
Analysis
Jeff Chih-Hao Wu, Chi-Fang Chen, Ruey-Chang We
Presenter: Jeff Chih-Hao Wu
Ocean ambient noise, the important parameter in sonar application, includes diverse noise sources
like waves, ships, and marine life, etc. On the other hand, ocean environment (such as bathymetry,
sediment, and sound speed profiles) which plays an important role in ocean ambient noise is
complex and changeable especially in Taiwan. For example, the water depth is less than 100 meter
in the Taiwan Strait, but over 3,000 meter in regions offshore of eastern Taiwan. Moreover, the
warm and saline Kuroshio passes through eastern Taiwan, and the intrusion of its branch occurs in
the northern and southern Taiwan. Since 2005, many underwater recording systems are deployed
in different seasons and locations around Taiwan. Above-mentioned systems include Several
Hydrophone Recording Unit (SHRU), SM2M marine recorder, DSG-Ocean acoustic datalogger, and
Passive Aquatic Listener (PAL), etc. The numerous ambient noise data contain irregular transient
signals like strumming noise, marine mammal, and sonar. In this study, automatic event detection
technique is applied for removing transient signals to extract the “continuous” ambient noise. The
characteristics of the ocean ambient noise can be quantified using statistical methods to know the
daily, monthly, seasonal difference in each location. Furthermore, Features of ocean ambient noise
in different locations are compared and analyzed to understand the spatial changes of ocean
ambient noise. The result of this study provides the idea how the ocean ambient noise changes
around Taiwan for scientific or military application.
192
UA2014
Session 27b
Large-amplitude internal solitary wave of the second mode in Luzon Strait:
observations and numeric modeling of its propagation
Andrey Serebryany, Andrey Belogortsev, Cho-Teng Liu
Observations of a solitary large-amplitude internal wave of the 2-mode which was met in Luzon
Strait during survey of R/V “Ocean Researcher 1” in May, 2006 are discussed. The wave was
observed at western side of Heng Chun Ridge and registered by echo sounder, ADCP, and other
devices. The wave had 50-m height and propagated to north-west with fast speed close to 3 m/s.
Passage of the internal wave accompanied by lateral rip band on the sea surface which resulted in
enhancement of underwater noise level up to 5-8 dB at frequency of 1 kHz. The observational data
were used for numeric modeling (in the frame of KdV equation with variable coefficients) of process
of the internal wave propagation through South China Sea from Luzon Straight up to continental
shelf. Scenario of internal wave generation in Luzon Strait has similarity with processes taking part
at Mascarene Ridge in the Indian Ocean where the 2-mode internal waves were observed too.
Reverberation Modeling in Range-dependent Waveguide
J.R.Wu T.F.Gao E.C.Shang
Presenter: J.R.Wu
A shallow water range independent reverberation model based on Perturbation theory has been
proposed recently. In this paper, the range independent reverberation model was extended to
range dependent waveguide. The bottom composite roughness has been considered. Small scale
bottom rough surface provides dominating energy for reverberation. While large scale roughness
has the effect of forward and back propagation. Its backscattering energy can be neglected
compared to small scale roughness backscattering. Small scale roughness backscattering theory
used in range independent reverberation model has been used in range dependent waveguide. And
the Green function of range dependent waveguide was calculated using modal spectrum of PE field.
Numerical analysis and experiment data show that the reverberation model in range dependent
waveguide works well.
193
UA2014
Session 29: Underwater Communication and Networking
Session 29a
High rate uplink acoustic communication from AUVs to surface platform
Xiao Zhang, Adam Zielinski, Jingwei Yin
Autonomous underwater vehicles (AUVs) are widely used in ocean operations. They often require a
robust and high data rate communication link in order to transmit the large amount of survey data
back to the surface platform. Those requirements face the problem of inter symbol interference (ISI)
caused by the multipath reflections. Methods used to alleviate the effects of the ISI include linear
or decision feedback equalization. However, these techniques require a high signal to noise ratio
(SNR) and a large number of equalization filter taps needed to cover the length of the acoustic
channel’s impulse response. Both of those conditions are restrictive. A new receiver structure that
employs a vertical array with narrow steerable beam that suppresses multipath has been proposed
in the paper. The received SNR is improved by the array processing, resulting in enhanced
communication reliability. The acoustic narrow radiation beam pattern of the array adaptively
tracks the direct path channel in order to suppress the multipath signals that are the main causes
of ISI. This makes the follow-up channel equalization easier and more effective due to shorter
impulse response of the channel. The simulation results show that the proposed uplink acoustic
communication technique has a higher data rate and reliability compared with the common
receiver structure under the same channel conditions. At the same time the channel equalization
computation complexity is reduced by at least half. The results attest to the robustness and
practicality of the proposed uplink acoustic communication scheme.
Clock Synchronization in Underwater Acoustic Networks During Payload Data
Exchange
K.G. Kebkal, O.G. Kebkal
Modern underwater acoustic modems, besides receiving and transmitting data, can measure signal
propagation time and, evaluate the distance between the transmitter and the receiver. Such
capabilities can be exploited for synchronization of underwater acoustic sensor networks, where
sensing and actuation must be coordinated across multiple nodes.
However, such problem as the accuracy of propagation time measurement, especially in highly
reverberant environments, as well as the evaluation of clock skew on interacting modems, have
been investigated rather poorly. One of the objectives of this paper consists in experimental
evaluation of stability/precision of propagation time measurements, particularly between
UA2014
underwater acoustic modems communicating in highly reverberant environments, as well as in
demonstration of the modems capabilities to precisely measure their mutual clock skews.
Another objective is to demonstrate a “comfortable” way for clock synchronization of underwater
acoustic networks nodes directly during payload data exchange. This opportunity is provided with
underwater acoustic modems built upon the sweep-spread carrier technology (S2C technology). A
distinctive future of this modems consists in their capability to exchange specific messages, so called
Synchronous Instant Messages, which enable precise timing of signal transmission into propagation
medium, as well as precise estimation of clocks skew between interacting modems right during
their payload data exchange. Owing to special cross-layer links between medium access control
and physical layers of the modem, the accuracy of signal transmission into water is about 1
microsecond. Owing to the ability of the S2C modems to resolve (or isolate) individual path of a
multipath signals, the physical experiments demonstrated high accuracy of the propagation time
measurements. Even in channels with severe multipaths (transmission distances 1-2 km, water
depth 20-25 m, near to La Spezia, Italy) the measurement accuracy of signals propagation time,
particularly for signals having duration 2 ms and comprising linear frequency spread (linearly
wobbulating carrier) in the range between 18 and 34 kHz, was very high remaining in the range
between 1 and 3 microseconds (rms value).
Such accuracy of signal propagation time measurement allowed to perform an accurate estimation
of mutual clock skews between interacting modems and so enable accurate clock synchronization
of underwater acoustic network nodes directly during payload data exchange (particularly, using
synchronous instant messages as containers of the data).
Experiments were carried out in laboratory conditions and in conditions of underwater acoustic
channels with several modem pairs and several mutual positions (underwater acoustic network
with five nodes).
Multi-user communication by adaptive time reversal in deep ocean
Takuya Shimura, Yukihiro Kida, Mitsuyasu Deguchi, Takami Mori, Yoshitaka
Watanabe, Hiroshi Ochi
Presenter: Takuya Shimura
Recently, demand for multiuser underwater acoustic communication has increased, e.g. for
multiple AUVs operation. Passive time reversal is an promising approach to realize multiuser
communication, because signals from differently positioned sources can be separated, while
intersymbol interference (ISI) is removed, by its spacio-temporal focusing. Additionally, to enhance
cancelling crosstalk, adaptive time reversal has been proposed by Kim et.al. In this paper, the
effectiveness of adaptive time reversal for multiuser communication in deep ocean is discussed with
experimental data and simulation using normal mode method including source movements,
comparing with multiuser-multichannel DFE. As results, the performance of adaptive time reversal
is better than that of multiuser-multichannel DFE, especially in case that sources are positioned
closer.
195
UA2014
Stealth underwater acoustic communications based upon steganography
techniques
Jm Passerieux
Presenter: Jm Passerieux
The proposed paper addresses the application of steganography to underwater acoustic
communications. More precisely, we propose to slightly modify previously recorded short pieces of
ambient noise (biological noise, underwater works, ship noise, rain, etc.), in order to hide some
information inside them before re-transmitting them. Unlike other stealth transmission techniques
(e.g. spread spectrum techniques with long spreading sequences) the transmitted signals will likely
be detected by an adversary. However, it is expected that they will be confused with harmless
ambient noise, thus not identified as communication signals.
With respects to similar techniques, such as the one already proposed in [1] (transmission of
sequences of simulated Dolphin clicks, with information encoded by the delay between clicks), the
main interest of our approach is that the transmitted signals will appear perfectly natural (they do
look like simulated signals). Moreover, they can also be perfectly compatible with the
environmental conditions of the day and location of the transmission.
In practice, standard audio steganography techniques are not applicable here (they are not robust
to the underwater acoustic channel and/or they can be detected by simple signal analysis
techniques). Therefore, the paper will focus on an original technique which consists in preparing,
with a given ad hoc recipe, an auxiliary signal from the initial signal. Then, before retransmission,
this auxiliary signal is added to the original one with low amplitude and a phase which carries the
information. At reception, the auxiliary signal is re-generated from the received signal. Then, the
phase and hidden information are retrieved by matched filtering.
The possibility to reliably transmit data at a low bit rate –- a few bits/sec at a few kilometers -- in a
realistic environment will be demonstrated by a few simulations, using recorded biological (clicks
of whales) or rain noise, and replay [2] of typical time-varying underwater acoustic channels.
The design of wide band transducers for underwater acoustic communication
To implement effective underwater communication, it is essential to design wide band transducers
with compact structures. tube transducers have been used as hydrophones for underwater acoustic
measurements, it is found that by combing different sizes of PTZ elements, this type of transducers
can be applied in wide frequency range with moderate transmitting response and sensitivity.
In this paper, the working principle of transducers are introduced, and transducers in different
ranges are manufactured. Measurement results are given which verify that this type of transducers
can be designed to cover the frequency range from 15 kHz to 100 kHz.
196
UA2014
Measurement and Modelling of Fading in Ultrasonic Underwater Channels
A. Sanchez, E. Robles, F. J. Rodrigo, F. Ruiz-Vega, U. Fernandez-Plazaola, J. F. Paris
Presenter: J. F. Paris
A crucial aspect for underwater acoustic communication (UAC) is the statistical characterization of
the communication channel. High-frequency UAC for short-range applications is very challenging;
the promise of small-size transducers has the drawback of fast time-varying channels. Shallow
waters UAC channels experience significant fading even when the transmitter and the receiver are
not intentionally moving relative to each other. A number of researchers have measured and
modelled fading UAC channels in the audio band; however, fading of ultrasonic UAC channels has
been barely investigated.
First, this paper reports measurements for ultrasonic UAC channels in Mediterranean shallow
waters performed by the company SAES and the University of Malaga. Secondly, statistical fit to
the measured data is performed. The measurements have been conducted when the transmitter
and the receiver are spaced 50, 100 and 200 m, approximately. Both transducers (B&K 8105 and
RESON TC4032) have been placed at depths 3, 6 and 9 m from anchored boats. Faded channel
sounding signals have been recorded at frequencies 32, 64 and 128 kHz. After preprocessing the
measured data, statistical fit to the recently proposed κ-μ shadowed fading model is performed;
this model includes, among others, the Rayleigh, Ricean and Nakagami-m fading.
Our conclusions are that fading in ultrasonic UAC is very significant and that the κ-μ shadowed
fading model is quite appropriate for its statistical characterization.
197
UA2014
Session 29b
Joint Pilot based Channel Estimation with Selected-Mapping to Reduce PAPR in
Underwater Acoustic MIMO-OFDM System without Side Information
Siyu Xing, Gang Qiao, Wei Wang, Songzuo Liu, Zhimeng Zhu
Presenter: Siyu Xing
In this paper, a peak-to-average-power ratio reduction scheme that requires no side information in
underwater acoustic MIMO-OFDM system is proposed. The key idea of the scheme is that different
phase sequences is represented by different comb pilot sequences, and it uses channel estimation
scheme to distinguish the number of phase sequences while completing the channel estimation at
the receiver. Therefore, the proposed scheme does not need to reserve bits for transmitting side
information, so that the data rate can be increased. Simulation results show that the PAPR
reduction performance was not reduced when comparing with the traditional selected mapping
and the BER performance is approximately the same as those selected mapping scheme with
perfect SI. Experiment carried in the tank also demonstrates the proposed scheme can differentiate
phase sequences, and significantly enhance the quality of the underwater acoustic MIMO-OFDM
communication system.
Ultrasonic Diversity OFDM Transceiver architecture with Impulsive Noise
Cancelling for shallow sea communication
Tran Minh Hai, Yasuto Matsuda, Taisaku Suzuki, Tomohisa Wada
Presenter: Tran Minh Hai
In order to support shallow sea under water communication to explore marine natural resources
using remote robotic control or to enable rapid information exchange between divers and so on, a
robust Digital Communication method under the multiple delayed refraction wave circumstance is
necessary. We propose OFDM Ultrasonic communication system with Diversity receiver. It utilizes
20-28 (KHz) ultrasonic channel and Subcarrier Spacing of 46.875 (Hz), 161-subcarriers OFDM
modulation. Living creatures in shallow sea generate Impulsive Noise so called Shrimp Noise. Then
Our OFDM diversity receiver has Time and Frequency Domain Impulsive noise Canceller with
Maximum ration combiner. The paper shows the proposed Diversity OFDM Transceivers
architecture and Experimental results taken at a fishing port in Okinawa Japan, which has shown
QPSK communication more than 50m distance shallow sea. In addition, an Inter-Carrier
Interference Canceller is incorporated, and experiments with moving receivers at 0.6 (m/s) and 0.9
(m/s) are conducted as well.
198
UA2014
PIC-DDFE-IDMA Detection For Uplink Shallow Water Acoustic Channels
S.N. Qader, C.C. Tsimenidis, M. Johnston, B.S. Sharif
Presenter: C.C. Tsimenidis
In this paper, a new receiver structure is proposed for the uplink of interleaved division multiple
access (IDMA) based transmission in shallow water acoustic channels. In a conventional IDMA rake
detector, jointly removing inter symbol interference (ISI) and multiple access interference (MAI) is
achieved by exchanging extrinsic log-likelihood ratio (LLR) chips between an elementary signal
estimator (ESE) and a posteriori probability decoders (DEC) in a turbo-like manner. This principle
can be extended to centralized decision feedback equalizer (CDFE) and DEC in CDFE-IDMA systems
to offer improved performance. In contrast, the proposed IDMA receiver can be considered as a
combination of a Parallel Interference Canceller (PIC) for eliminating MAI with a decentralized
decision feedback equalizer (DDFE) to remove the ISI effects. The PIC utilizes the same ESE principles
in removing MAI effects from received symbols, which is more efficient than cross-over filters in a
CDFE-IDMA system. Moreover, the ISI subtraction in DFE is more effective compared to ESE.
Moreover, by analyzing the computational complexity of the proposed PIC-DDFE-IDMA and
comparing it with both rake IDMA and CDFE-IDMA on the basis of the required operations, it can
be demonstrated that the PIC-DDFE-IDMA provides much lower complexity than rake IDMA
detection, however, it is more complex than CDFE-IDMA. Utilizing experimental channels obtained
by sea-trials conducted by Newcastle University in the North Sea, simulation results demonstrate
the superiority of PIC-DDFE-IDMA over the other two detection methods.
199
UA2014
Session 30: Underwater Unexploded Ordnance (UXO) Detection and Remediation
Organizer: Mike Richardson and Wolfgang Jans
Location: Lecture Room B, Monday 23rd June 2014, 10:45 – 12:45
First results for buried object detection from the “sounding ammunition
(SOAM)” 2013 experiment
David Rose, Wolfgang Jans
Presenter: David Rose
The German attempt to move away from fossil fuels and abandon nuclear power has put the focus
on renewable energy, and especially on offshore wind farms in Germany. On the other hand
remains from World War II like dumped ammunition and other unexploded ordnance (UXO) in the
North Sea and the Baltic Sea become in this context a more and more urgent problem.
Currently marine magnetometers are often used to detect UXO at sea, although their range is very
limited. Furthermore, some objects are made of alloys, making them undetectable by
magnetometers. Hence, acoustic and electromagnetic methods can help with UXO detections.
One part of the Sounding Ammunition Project (SOAM) is to compare different acoustic sonar
systems. Therefore, several sea trials on a test site with submerged objects were planned. The test
site contains several metal test objects. These objects are situated in different depths between flush
buried and 1 m under the seafloor consisting of sand and muddy sand.
The focus in the first sea trial 2013 was on the seafloor (side scan sonar, bathymetry) and on cross
sections with two different sub-bottom profilers. A magnetic sensor was used to correlate acoustic
and magnetic signals. This sea experiment will be described and first results will be shown.
Additional, some prospects on measurements with different multibeam sub bottom profilers and
magnetic sensors will be made.
Target detection with low-cost imaging sonars
Christian De Moustier, Joe Calantoni
Presenter: Christian De Moustier
The probability of detecting targets of interest on the seafloor with imaging sonars depends on the
spatial resolution and contrast of the image. This paper presents a "proof-of-concept"
demonstration that "low-cost" imaging sonars augmented with some signal processing can be used
effectively to detect targets of interest in a scene under surveillance by a sonar at a fixed location
or on a moving platform (e.g. AUV or ASV). In general, these sonars output beamformed time series
of acoustic backscatter for each ping, with little or no corrections applied. Our approach consists
in estimating, then removing, the deterministic spatial gain introduced by the transmit and receive
beam patterns. In the process, we normalize the measured backscatter value for the size of the
instantaneous area ensonified at each range increment within each beam, and compensate for
transmission losses incurred during the round-trip path between the sonar and the seafloor.
200
UA2014
Low frequency feature extraction for target discrimination on a Biosonar
dataset
Tara Leblanc, John Fawcett, Yan Pailhas, Duncan Williams
Presenter: Tara Leblanc
Low frequency broadband sonar is an emerging technology in the field of underwater unexploded
ordnance (UXO) detection and identification. These lower frequency acoustic signals can penetrate
an object and allow one to extract target features based on both elastic effects and internal
structure that will allow for more accurate target classification. One such sonar, Hydrason’s (UK)
Biosonar is a broadband side-scan sonar that operates between 30 kHz and 130 kHz and uses
dolphin inspired pulses. This paper examines data from trials conducted by the UK with Biosonar in
the fall of 2012. Particular focus is given to the classification of a series of cylindrical targets with
identical structures yet differing internal fillers. For these targets, numerous features are considered
in order to differentiate between the targets i.e. their fillers. These features include time-domain
features, spectral-based features, and aural-based features. Simple classification schemes are
applied and the suitability of the features is discussed in the context of an ATR (automated target
recognition) scheme.
State of the Art in Commercial Offshore Unexploded Ordnance Detection
Kay Winkelmann
Presenter: Kay Winkelmann
Millions of tons of explosive remnants of war, including both conventional and chemical stockpile
munitions, have been dumped in the North Sea, the Baltic, the Irish Sea, the Barent Sea, the
Mediterranean and other maritime environments worldwide. Also, military exercises, testing and
warfare at sea have left behind large quantities of unexploded ordnance (UXO) at sea. In the North
Sea and the Baltic alone, approximately 700,000 mines were laid in the First and Second World
Wars, most of which were never recovered.
With the increasing utilization of the maritime environment for energy production (offshore oil and
gas, offshore wind energy, offshore tidal power), international trade (harbor construction and
extension), and the production of maritime food, the clean-up of UXO and dumped conventional
and chemical munitions becomes more and more important. Compared to the year 2010, the
market for offshore UXO detection and removal has multiplied.
Standards for offshore UXO detection and removal as they have been adopted in land applications
remain yet to be developed. Because of a lack of experience on the side of service providers – many
of which are new to this market – and due to a lack of awareness for the dimension and risks of the
offshore UXO problem on the side of the clients, many surveys carried out fail to detect the UXO
sought.
Careful consideration of the characteristics of state-of-the-art detection systems and objects of
interest and environmental parameters allows to design UXO surveys such that the objectives are
met. Systems used in offshore UXO detection include scalar and vectorial magnetometers, metal
201
UA2014
detection systems, sidescan sonar systems and high resolution multibeam echosounder systems.
Examples of safety considerations for offshore construction activities with respect to UXO, large
marine UXO left undetected in surveys and good practice examples are discussed.
Offshore Unexploded Ordnance Recovery and Disposal
Jan Koelbel
Presenter: Jan Koelbel
Large quantities of excess munitions, including both conventional and chemical munitions, have
been dumped in maritime environments worldwide. Also, military exercises, testing and warfare at
sea have left behind large quantities of unexploded ordnance (UXO) at sea.
Prior to offshore construction activities including geotechnical investigations, grapnel runs,
trenching, pipe and cable laying, construction or ramming of foundations, risks from UXO must be
excluded. During the UXO survey, typically many objects of interest are detected. Typically (except
for dumping areas), most of the objects detected turn out to be innocuous during visual inspection.
As operating costs are high for offshore UXO recovery and disposal, efficient technologies need to
be applied for the task.
Because of the water depths, environmental conditions (currents, visibility, and temperature) and
hazards of UXO inspection and recovery, the application of technical, remotely operated systems
must be preferred over divers.
Issues that need to be addressed in offshore UXO recovery and disposal included
• Suitable high-precision underwater positioning for relocating of targets detected in geophysical
surveys
• Sensor systems to assist the re-location of targets
• Maximization of operational time (24-hour-operations) for maximum efficiency
• Application of remotely operated vehicles with powerful manipulators and visual sensors
(workclass ROVs) instead of humans (divers) for work on potentially lethal objects in order to reduce
exposition and maximize productivity
• Application of tools for the recovery of bulk or heavy UXO or scrap (electromagnets, underwater
excavators)
• Enhancement of underwater visibility to allow for the clear visual identification of targets
(imaging sonars)
• Remotely operated placement of explosive devices and environmental protection
considerations for open detonation of UXO at sea in case recovery and disposal on land are not
possible
202
UA2014
Using A 3D Subbottom Profiler For AUV-Based Pipeline Detection And
Localization
Aneta Nikolovska
Presenter: Max Abildgaard
A Deep Diving AUV (DD-AUV) with sensors and processing capacity to detect and to inspect surface
laid and buried pipes/cable is presented. The detection and localization capability is achieved by
fusing data from different sensors: optical cameras; magnetic gradiometers; side scan sonar; multi
beam sonar and synthetic aperture sub-bottom profiler. The detection was performed on the basis
of the Constant False Alarm Rate (CFAR) method.
This paper describes the current status of work on the vehicle; the sensors and the pipe/cable
detection method.
Initial tests are performed with demonstrator testbed which verified the consistency of the
hardware and software concepts.
203
UA2014
Session 31: Unmanned Vehicles (AUV, USV and Gliders) for Underwater Acoustic
Surveillance and Monitoring
Session 31a
Monitoring the underwater acoustic pressure field using two spatiallyseparated hydrophones with application to forward-aft sensors onboard an
undersea glider
Brian G. Ferguson, Kam W. Lo
For an acoustic sensor onboard an undersea glider, the effect of hydrodynamic flow noise (or
broadband pressure fluctuations induced by unsteady flow or turbulence) is small and platform
noise is observed only briefly when either the buoyancy change pump or trim adjustment
mechanism is actuated. A single hydrophone onboard an undersea glider can be used to detect
and range surface watercraft for passive surveillance and collision avoidance purposes. The
instantaneous range of a source can be estimated by measuring the multipath time delay between
signals arriving at the sensor via the direct and indirect (boundary-reflected) propagation paths.
Bearing estimation of the source would require the addition of another sensor. To enhance the
bearing resolution, the two sensors would need to be mounted on the forward and aft extremities
of the glider so as to maximize the intersensor separation distance. In the present paper, the
outputs of two fixed hydrophones (positioned 1 m above the sea floor and 14 m apart) are crosscorrelated to enable measurement of the time difference of arrival (TDOA) of a signal wavefront at
the two sensors. The TDOA measurement is then converted to a source bearing, which is measured
with respect to the sensor pair axis. Using real data, the variation with time of the instantaneous
bearing estimates is shown for the transits of various sources: rigid-hulled inflatable boat (RHIB),
maritime services work boat, autonomous underwater vehicle (AUV), diver propulsion vehicle
(underwater scooter), open-circuit scuba diver, and a helicopter. Also, the source bearings of a
series of underwater acoustic communication transmissions detected during an experiment are
presented. The bearing errors for each of the sources depend on the signal-to-noise ratio, signal
bandwidth, integration time, intersensor spacing, and source bearing. For a typical undersea glider
fitted with two hydrophones (one forward, the other aft), the bearing errors would be an order of
magnitude larger due to the smaller intersensor separation distance.
204
UA2014
Passive acoustics embedded on gliders – Weather observation through
ambient noise.
Pierre Cauchy, Pierre Testor, Laurent Mortier, Marie-Noelle Bouin
Presenter: Pierre Cauchy
Underwater gliders can provide high resolution (~4h / 4kms) water temperature and salinity
profiles. Being able to associate them with a surface weather conditions estimation would allow to
better study sea-air interactions. Since in-situ observations of the marine meteorological
parameters are difficult, the development of a glider embedded weather sensor has been studied,
based on the WOTAN approach (Weather Observation Through Ambient Noise), as described by
Ma & Nystuen (2004).
In the 1kHz-30kHz frequency range, the background underwater noise is dominated by wind
generated noise, at any depth. Focusing on the sound pressure level at 5kHz, 8kHz, 10kHz and
20kHz, we are able to provide an estimation of the sea surface wind speed. Thus, deploying a glider
with an embedded hydrophone gives an access to the surface weather conditions around it's
position.
We have deployed gliders in the Mediterranean sea, with passive acoustic monitoring devices
(Acousonde) onboard, during the 2013 Moose, HyMeX/MERMeX and MED-REP13 experiments. 4
months of data have been recorded and post recovery processed. Wind speed estimations have
been confronted to weather buoys observations and operational atmospheric models predictions,
to validate the method and estimate the results accuracy. Wind estimates have been obtained with
a ~2m.s-1 error. A specific emphasis has been placed on the robustness of the processing through
multi frequencies analysis and depth induced attenuation correction, as well as on the acoustic
sampling protocol on which a downscaling study has been performed in order to meet the low
energy consumption glider standards, for a future real time embedded processing.
The glider generated noise and its vertical movement are not perturbing the estimation. Moreover,
the surface behavior of the Slocum gliders (vane effect induced by its tail), allow an estimation of
the wind direction.
Small vessel detection through the use of an underwater glider
Tesei, A., Been, R., Troiano, L., Dymond, R., Maguer, A.
Presenter: Tesei A.
Monitoring the marine traffic of small- and mid-sized boats is of major interest for many
applications, ranging from the surveillance of boarders against illegal immigration or illegal goods
traffic, to the protection of marine parks or assets. In this work the approach selected is conducting
passive acoustic monitoring of marine traffic by hosting a small volumetric hydrophone array of
eight elements on an underwater glider.
Passive underwater acoustic monitoring technologies applied to mobile autonomous underwater
platforms may allow:
205
UA2014
o detection of boats not equipped with an AIS antenna and with low radar signature (such as
inflatable boats);
o minimum environmental impact;
o covertness ;
o wide area coverage;
o near real-time, continuous (d7/h24) monitoring;
o long endurance;
o relatively easy deployment and recovery;
o availability of several functionalities, from detection to classification of multiple acoustic noise
sources at the same time.
In particular, the main objective is to detect and estimate the direction of arrival of boats by
processing acoustic data directly on board the glider. Appropriate algorithms are developed to
achieve the automatic detection of surface vehicles and their direction of arrival; the processing
chain is applied to the acoustic data on the glider during its dive and the results are sent to the
glider’s control station each time it comes to the sea surface.
This paper will describe the mobile acoustic measurement system, will present the processing chain
developed, and will show the preliminary results obtained during at-sea tests. This work is partially
funded by the European Union in the context of the FP7 PERSEUS Project.
Automated Detection of Fishing Vessels using Smart Hydrophones on an
Underwater Coastal Glider
Ray Mahr, Mark Wood
Presenter: Ray Mahr, Mark Wood
AUVs operating in coastal waters where fishing vessels are also operating are at risk of becoming
entangled in nets or long lines close to the surface or other fishing equipment in the water.
Autonomous gliders operating in these coastal waters are probably more susceptible to the dangers
of these fishing vessels than powered AUVs, but the risks exist in all cases.
A method is proposed to minimize this risk of potential interaction with the fishing gear by using
two or more smart hydrophones on the coastal glider and to provide glider maneuvers that prevent
interaction with the gear deployed from the fishing vessel. In order to command a proper glider
maneuver, a fishing vessel must be detected, and classified as to type..
This paper describes both aspects of this fishing vessel avoidance issue – developing the necessary
detection techniques for identifying these vessels, and the necessary maneuvers that the glider
must take to avoid any interaction with the fishing vessel or its gear.
Ocean Sonics of Great Village, Nova Scotia, Canada has developed a Smart Hydrophone with the
ability to detect the radiated-noise signatures of different types of fishing vessels, and has the
memory to store known signal characteristics of these vessels. With these stored acoustic
signatures and characteristics, estimates can be made of the type of fishing vessel detected as well
as the approximate range to the vessel and fishing equipment deployed. Exocetus Development of
Anchorage, AK has developed a glider named the Coastal Glider which has a complete set of
206
UA2014
behavioral maneuvers, and with this fishing vessel information, the control system can decide what
maneuvers would be best for avoiding interaction with the fishing vessel or its gear.
It is likely that fishing vessels around the world have their unique characteristics, and the proposed
detection algorithms used to automatically detect different types of fishing vessels will be customtuned for the areas where the coastal gliders will be deployed. These algorithms will be discussed
as well as ongoing work to improve these algorithms using data provided by operators of the
Exocetus Coastal Glider.
207
UA2014
Session 31b
NeXOS objectives in multi-platform underwater passive acoustics
Eric Delory, Daniel Toma, Joaquin Del Rio, Pablo Ruiz, Luigi Corradino, Patrice Brault,
Frederic Fiquet
Presenter: Eric Delory
The objective of the NeXOS project is to develop cost-effective, innovative and compact integrated
multifunctional sensor systems in ocean optics, ocean passive acoustics and for an Ecosystem
Approach to Fisheries (EAF), which can be deployed from mobile and fixed ocean observing
platforms. Resulting downstream services will contribute to the GEOSS, the Marine Strategy
Framework Directive (MSFD) and the Common Fisheries Policy of the European Union. The
development of innovative hydrophones will focus on the pre and post-processing of acoustic
information and improved transducer integration, reducing size and overall procurement and
operations cost while increasing functionality. An important part of the effort will focus on the need
for greater dynamic range and the integration on autonomous platforms, such as gliders and
profilers. Embedded processing will be reconfigurable, allowing the monitoring of MSFD Good
Environmental Status descriptors 1 (Biodiversity) and 11 (Underwater Noise) as minimal
requirements. The first phase of the project consists in interacting with scientific communities and
the industry in order to narrow down initial requirements and possibly extend the planned
functionalities to new applications. The presentation will provide an overview of the project and an
update on current progress, with a focus on unmanned vehicles and mobile platforms more
generally. NeXOS is co-funded by the European Commission 7th Framework Programme, the Ocean
of Tomorrow 2013.
The fusion of digital terrain models measured from multiple acoustic sensors –
Application to the DAURADE autonomous underwater vehicle
Ridha Fezzani, Benoit Zerr, Michel Legris, Ali Mansour, Yann Dupas
Presenter: Ridha Fezzani
Building an accurate digital terrain model (DTM) of the seabed is a key issue for various military
and civilian hydrographers applications.
In the past decades, the emergence of autonomous underwater vehicles (AUV) offers new
methodologies to collect the bathymetric data used in the estimation of the DTM. In our study, we
use the DAURADE AUV platform which is capable of acquiring bathymetry with two acoustic
sensors: A multibeam echo sounder (MBES) and an interferometric sidescan sonar (ISSS). The two
sensors (MBES and ISSS) are synchronized to operate concurrently. In fact, the final DTM can be
improved by performing a fusion of the data; the two systems acquire the bathymetry with different
resolutions, geometries and error models; these parameters are introduced in the fusion process
to improve the estimation of the DTM and to increase its accuracy.
208
UA2014
The aim of this paper is to describe the fusion method and discuss our simulated results. First, the
modeling of two acoustic sensors (MBES and ISSS) will be briefly described. The input data sets are
simulated by applying the sensor models on simplified seabed models. The use of seabed models
provides ground truth and, therefore, allows for quantifying the accuracy of the fusion process.
We applied the fusion process on actual data from the two bathymetric sensors of DAURADE (Reson
7125 MBES and Klein 5000 Inteferometric); the obtained results will be presented and discussed.
Key words: AUV, multibeam, interferometric sidescan, fusion, bathymetry, seafloor.
Real time improvement of the seabed mapping with AUV-borne sensors using
statistical analysis
Naveed Islam, Ahmed Nait-Chabane, Benoit Zerr, Yann Dupas
Presenter: Naveed Islam
Sonar data is commonly affected by noise due to the processing of scatter signals and interference
of acoustic waves scattered from the seabed. To overcome this problem and limit the noise in sonar
images, the sonar operator can change the sonar settings (e.g. range, pulse length, modulation,
inter-track distance, etc.) to acquire the best possible acoustic data. On board autonomous
underwater vehicles (AUV), due to the low bandwidth of the communication with the robot, the
real time definition of the best settings by an operator is nearly unfeasible. For these reasons, we
have developed a real-time analysis method for automatically assessing the quality of the data.
The results of this process are then sent to the AUV planning module which can change the sonar
settings (e.g. inter-track distance).
The classical approach is based on the correction of the artifacts related to the wave propagation
in water column and the characteristics of the sonar system. This approach requires strong a priori
knowledge of the system and the conditions of acquisition of the sonar data.
The main objective of this paper is to propose a statistical measure of quality of the sonar data
acquired using AUVs. This statistical measure would be representing a quality map for the input
sonar data. As no prior measurement of similarity or dissimilarity of sonar images is given, the
decision to whether accept the quality of data as noisy/non-noisy will be based upon statistical
hypothesis testing.
To accomplish the quality mapping, different noise models are analyzed over the sonar data using
linear/non-linear filters. To determine the best fit for noise distribution, the goodness-of-fit (GoF)
test for each noise distribution is carried out at different significance levels in different regions of
the sonar range image. An adaptive threshold based on the confidence level of the noise distribution
is used to identify and separate noisy or non-noisy regions of the sonar data. The experimental
results on sonar data images acquired using DAURADE AUV are presented and discussed.
209
UA2014
Real-time acoustic monitoring of baleen whales from autonomous platforms
Mark Baumgartner, David Fratantoni
Presenter: Mark Baumgartner
In the past decade, much progress has been made in real-time passive acoustic monitoring of
marine mammal occurrence and distribution from autonomous platforms (e.g., gliders, floats,
buoys), but few systems are capable of detecting the calls of multiple species simultaneously. We
have combined the low-frequency detection and classification system (LFDCS; Baumgartner and
Mussoline, 2011, JASA 129:2889-2902) with the digital acoustic monitoring (DMON) instrument to
detect, classify, and report in near real time the calls of several baleen whale species, including fin,
humpback, sei, bowhead, and North Atlantic right whales. The DMON/LFDCS has been integrated
into the Slocum glider and APEX profiling float, and we have integration projects currently
underway for the Liquid Robotics wave glider and a moored buoy. In a recent evaluation study, we
deployed two DMON/LFDCS-equipped Slocum gliders in the central Gulf of Maine for 3 weeks
during late November and early December 2012. The gliders reported over 25,000 acoustic
detections attributed to fin, humpback, sei, and North Atlantic right whales. Real-time detections
were evaluated after recovery of the gliders by (1) comparing the acoustic detections to
continuously archived audio recorded by the DMON/LFDCS, and (2) comparing species-specific
detection locations with nearby sightings collected from both an aircraft and ship. The overall false
detection rate for individual calls was 14%, and for right, humpback, and fin whales, false
predictions of occurrence during 15-minute reporting periods were 5% or less. Agreement between
acoustic detections and visual sightings from aerial and shipboard surveys was excellent (9 of 10
visual detections were accompanied by real-time acoustic detections of the same species by a
nearby glider). We envision that this autonomous acoustic monitoring system will be a useful tool
for both marine mammal research and mitigation applications.
SONOBOT - an autonomous unmanned surface vehicle for hydrographic
surveys, hydroacoustic communication and positioning in tasks of underwater
acoustic surveillance and monitoring
K.G. Kebkal, I. Glushko, T.Tietz, R.Bannasch, O.G.Kebkal, M.Komar, S.G.Yakovlev
Unmanned surface vehicles (USVs) evolved over decades, becoming more effective and affordable,
and with the modern benefits of wireless data transmission technologies and precise global
positioning systems, the USVs are becoming more and more attractive for a wide range of
commercial, scientific and military operations. For modern unmanned vehicles, advanced radio and
satellite communication technologies enable live feedback from onboard sensors, remote control
or complete autonomy. Further USV research and development efforts focus on greater
communication ranges, more autonomous functionality, easier and more reliable launch and
recovery, greater sensor performance, better power management and longer operation times. The
USVs are presented in various hull and craft types, such as semi-submersible craft, conventional
210
UA2014
planing hull craft, semiplaning hull craft, hydrofoils, catamarans and many others. With sizes
ranging from small vehicles to large unmanned boats, more and more prototypes and commercial
USV solutions are becoming available on the market. Although there are still significantly fewer
USVs than their unmanned underwater vehicle counterparts (UUV), demand for USVs grows for
many applications. USVs for naval operations are often based upon traditional surface vessels, as
controls, navigation and telemetry systems can transform almost any conventional craft into a USV,
remotely operated by crew ashore or on other vessels. Although unmanned, these systems depend
on telemetry and are more similar to underwater remotely operated vehicles than autonomous
underwater vehicles (AUVs). A greater variety of hull forms was developed for academic and
commercial survey purposes. Survey applications trend toward small, low-cost USV platforms that
can be used for a variety of missions. Smaller USVs are particularly attractive for hydrographic
surveys aimed at measuring the depth and bottom configuration of water bodies, as an
autonomous or remotely controlled surface vehicle can be equipped with on-board echo-sounder,
side-scan sonar and other sensors to collect valuable data in automated mode. As geological
mapping of the seafloor in nearshore or shallow water areas presents technological challenges due
to the dynamics of the environment, high volume of data collected, and the limitations of operating
in very shallow water, a remotely operated or autonomous seafloor mapping USV is a great solution
for the problems of shallow water settings.
The SONOBOT, an autonomous USV, was designed and built by EvoLogics GmbH, to operate in
harbors, inland and coastal waters and to address the industry need for a light and affordable
platform for hydrographic surveys, hydroacoustic communication and positioning in tasks of
underwater acoustic surveillance and monitoring. Design of the SONOBOT vehicle is discussed in
the paper, followed by an overview of the platform and practical results of the SONOBOT survey
missions.
The Persistent Maritime Monitoring System (PMMS)
Willcox S., Leroy F., Wyatt P.
Presenter: Leroy F., Willcox S.
This paper presents a system concept for persistent acoustic surveillance and monitoring of large
ocean areas by a self-deploying system of surface, submerged, and bottom-mounted sensor nodes.
We begin with a brief review of maritime security and fisheries management missions that provide
technical, operational, and economic context for the system. We then describe the system concept
in detail, including the USV and potential bottom-mounted and submerged nodes, and the seafloorto-shore communications architecture. Next, we discuss four key enabling technologies that we
have developed or are currently developing, both internally and with partners: the persistent USV
platform (“SV-3”), the seafloor to shore communications payload (“Gateway”), a family of acoustic
surveillance and monitoring payloads (“Sentinel”), and a multi-vehicle centralized autonomy
(“MVCA”) system for coordinating and managing the control of networks of vehicles at the fleet
scale. For each of these technologies, we provide technical details of the existing capabilities, review
performance results from sea trials and operational deployments, and discuss our technology
211
UA2014
roadmaps for further development. Finally, we present results from towing-capacity sea trials with
the USV. Taken together, these mature and developing technologies and operational capabilities
validate our concept for persistent and economical acoustic surveillance and monitoring of large
areas of the ocean with a self-deploying, unmanned system of systems.
212
UA2014
Session 31c
Virtual Ocean Testbed For Autonomous Undersea Sensing Networks
Henrik Schmidt, Erin Fishell
Presenter: Henrik Schmidt
The MOOS-IvP Nested Autonomy architecture provides an open-source, behavior-based artificial
intelligence framework for robust operation of undersea sensing networks with their inherently
severe communication constraints, allowing for mission objectives to be achieved without the need
for constant operator supervision and intervention. However, the reliance on a high level of
autonomy requires extensive testing for performance demonstration and risk reduction ahead of
extended operational deployments. To lower the cost and time required, a comprehensive
simulation environment, the Virtual Ocean Testbed, has been developed, where the MOOS-IvP
payload autonomy system on an arbitrary number of nodes is operated unchanged with the
physical platform and the surrounding environment being replaced by physics-based simulators,
including state-of-the-art ocean circulation models, high-fidelity acoustic propagation and
scattering models, and advanced models for platform and sensor array dynamics. This paper
describes the architecture of this Virtual Ocean Testbed and provides examples of its use for the
planning of an experiment demonstrating the autonomous detection, classification and tracking of
seabed objects using bistatic active acoustics. [Work supported by the Office of Naval Research
and the NATO Undersea Research Centre].
Networking underwater, surface and air vehicles: tools and experimentation
Ricardo Martins, Joao Borges De Sousa
Presenter: Ricardo Martins
In this paper we describe the unmanned vehicles, software and communication architecture and
devices used by the Underwater Systems and Technology Laboratory (LSTS) in at-sea operations.
Novel achievements like data mulling using unmanned aerial vehicles and adaptable operator
interfaces for networks of multiple vehicles and sensors are also presented.
The LSTS fleet includes two remotely operated submarines, eight autonomous underwater vehicles,
one autonomous surface vehicle, and twelve autonomous air vehicles.
The software toolchain consists on three main entities: DUNE onboard software, Neptus command
and control software and a common IMC message-based communication protocol.
213
UA2014
CMRE's use of AUVs and USVs for the demonstration of network concepts for
multistatic active ASW
Kevin D. Lepage, Ryan Goldhahn
CMRE has been pursuing the development of an ASW Autonomous Security Network demonstrator
based on AUVs for detection and USVs performing communications gateway services. Already the
demonstrator has been to sea 7 times in the last six years, with two additional deployments planned
for 2014. In this talk the characteristics of the demonstrator are described and some of the
challenges and research directions for underwater ASW networks are identified. CMRE's experience
to date shows that the concept of unmanned UUVs deployed in a multistatic network, performing
real-time DLCT and making autonomous decisions, is practicable. Challenges to the further
development of the concept fall roughly into four catagories: 1) Reducing the detection threshold
while keeping the number of false alarms roughly constant, 2) improving inter-vehicle and vehicle
to gateway communications, 3) developing control or behaviour strategies to maximize ASW
network performance, and 4) overcoming limitations to the endurance and deployability of AUVs
and USVs. Results obtained with the current network are briefly reviewed and the work on-going
to address the challenges is described.
214
UA2014
Session 32: Vector Sensors: Development and Applications
Session 32a
An experimental study on DEMON spectrum direction estimation of multitarget with a low frequency vector hydrophone
Hualin Lan, Dajun Sun, Jidan Mei, Tingting Teng
Presenter: Hualin Lan
Due to the periodic motion of the ship propellers, its radiated noise will carry rich discrete
components including shaft and blade frequency and their multiplications information. And these
discrete components may be modulated by the continuous spectrum involved in the radiated noise
and produces the DEMON spectrum. The DEMON spectrum is actually a kind of modulated
spectrum of the ship radiated noise, from which the information about the shaft and blade can be
extracted. What’s more, the information can be used to distinguish the targets of interest. Owing
to the DOA estimation ability of a single vector sensor, such discrete component can be used to
distinguish the targets distributed in different directions. Most of the existing DEMON methods
make use of the high frequency (above 1 kHz) information, from which the blade information is
obtained. In this paper, the study focuses on the low frequency DEMON spectrum extraction for
multi-target and their DOA estimation with a single vector sensor. The sea trial data acquired by a
submerged buoy was processed using the above method. Results indicated that DEMON spectrum
is rich in low frequency, from which the shaft component and DOA could be determined. Compared
with the method based on cross-spectrum DOA estimation using a vector sensor, the presented
method could achieve better ability for multi-target discriminate and DOA estimation.
Signal Processing for circular Vector-Sensor Array mounted around a
cylindrical baffle
D.S. Yang, Z.R. Zhu
Presenter: D.S. Yang
A method of the Acoustic Vector-Sensor Array Processing for a uniform circular acoustic VectorSensor Array (UCAVSA) mounted around a cylindrical baffle is presented. Using the elastic thin shell
theory, the analytic expressions for the scattered pressure and particle velocity are derived. It is
found that the pressure and the particle velocity fields near the surface of the cylindrical baffle are
characterized by complex interference structure. Then the total pressure field and the total particle
velocity field near the surface of the cylindrical baffle are analyzed theoretically by applying the
method of spatial Fourier transform. The so-called modal vector-sensor array signal processing
algorithm, which is based on the decomposed wavefield representations, for the UCAVSA mounted
around the cylindrical baffle is proposed. Simulation and experimental results show that the
UA2014
UCAVSA mounted around the cylindrical baffle has distinct advantages over the same manifold of
traditional uniform circular pressure-sensor array (UCPSA). It is pointed out that the acoustic
Vector-Sensor (AVS) could be used under the condition of the cylindrical baffle and that the UCAVSA
mounted around the cylindrical baffle could also combine the anti-noise performance of the AVS
with spatial resolution performance of array system by means of modal vector-sensor array signal
processing algorithms.
Broadband DOA estimation in phase modal space for circular acoustic VectorSensor Array
Z.R. Zhu, D.S. Yang
Presenter: Z.R. Zhu
An approach to Broadband direction of arrival (DOA) estimation for a uniform circular acoustic
Vector-Sensor Array is proposed. The pressure field and the particle velocity field are decomposed
into an orthogonal set of phase modals by applying the method of wavefield decomposition. The
method to construct pretreatment matrix, which transform the received signals from element space
to phase modal space, is proposed for broadband farfield signals. Simulation and experimental
results show that the UCAVSA has distinct advantages over the same manifold of traditional
uniform circular pressure-sensor array (UCPSA) in anti-noise performance, resolution and accuracy.
The method is based on the principle of coherency between pressure and particle velocity, which
can suppress interference in isotropic noise field. The algorithm could combine the anti-noise
performance of the Vector-Sensor with spatial resolution performance of array system to solve the
problem of high-resolution DOA estimation of remote targets for UCAVSA.
Patch near-field acoustical holography based on vector hydrophone array
Hu Bo, Yang Desen, Sun Yu
Presenter: Hu Bo
Near-field acoustical holography is a powerful tool for identifying noise sources from partially
known sound pressure field. However, the current theory of NAH is not applicable to tracking large
scale noise sources. Patch near-field acoustical holography (PNAH) is related to the partially
measured pressure on the hologram surface in terms of sampling and bandlimiting matrices, which
cost more in computation. PNAH procedure based on measuring of vector hydrophone array is
described, including the mathematical formulation. The measurement array can been smaller than
the source, thus the practicability and efficiency of this technology is greatly enhanced. Then an
experiment has been carried out with vector hydrophone array. The experimental results have
illustrated the high performance of PNAH and the advantages of a vector hydrophone array in an
underwater near-field measurement.
216
UA2014
A Robust Noise Sources Localization and Identification method based on Vector
Sensor Array
Jie Shi, Desen Yang, Shengguo Shi
Presenter: Jie Shi
Minimum variance distortionless response (MVDR) focused beamformer based on vector sensor
array is widely used in the area of the noise source high-resolution localization and identification.
However, when an arbitrary unknown signal steering vector mismatch occurs or the training sample
size is small, the performance of MVDR focused beamformer will be severely degraded. In this
paper, we develop a new approach, which is based on the worst-case concept, to improve the
robustness of the original method. It is shown that the proposed algorithm improves its robustness
by imposing the array response constraint on the uncertainty set of the steering vector, and can be
reformulated in a convex form as the so called second order cone program (SOCP), and then solved
efficiently using the well established optimization tool, Sedumi. Theory analysis and computer
simulations show better performance of our robust beamformer as compared with the existing
methods: it can achieve a greater dynamic range,sharper focused peak,and lower back-ground
noise level.The results in this paper demonstrate our proposed method can be applied in the
underwater noise source high-resolution localization and identification.
Research Vector Hydrophone Measurement System
Hong Lianjin, Fang Erzheng
Presenter: Hong Lianjin
The pressure and particle velocity could be measured vector hydrophone at the same location and
the same time in underwater sound field, then, the message of sound field could be described
briefly. The vector hydrophone of resonant-column type was widely used in underwater acoustic
projects in virtue of its strong-points, such as the high sensitivity, the smooth response in the
frequency band of interest, the fine symmetry in direction performance, and the high resolution.
In the thesis, a kind of 3-D cylindrical vector hydrophone was designed for the project’s
requirements. The vector hydrophone was combined with pressure channels and vector channels.
The theory of acoustic reception under the case of rigid cylinder and elastic cylinder was researched
in this thesis. The hanging model of column receptor was founded and researched via
experimentation. The pressure channels and vector channels of the designed vector hydrophone
was researched via simulation by using ANSYS. The 3-D vector hydrophone designed based on the
result of simulation with the size Φ66?82mm, and the essentially flat response between 102000Hz.The vector hydrophone has high sensitivities, be convenience in hanging and smaller
volume. Its performance was measured both in standing wave pipe and anechoic lab. The result of
experimentation indicated: The pressure sensitivity of vector channels was -180dB (measure
frequency: 1000Hz, 0dB re 1V/μPa) with direction performance of cosine, and the pressure
sensitivity of pressure channels was -192dB (0dB re 1V/μPa).
217
UA2014
On this basis, a three-element vector hydrophone line array measurement system has been built,
the experiment showed that: the system can effectively carry through the radiated noise
measurement of .the underwater structure.
218
UA2014
Session 32b
Measurement of the vector field and its application on Geo-acoustic inversion
Fenghua Li, Liangming Zhu, Renhe Zhang
Presenter: Fenghua Li
Vector sensor, which can measure sound pressure and particle velocities simultaneously, has
attracted much attention in recent years. In spite of the measurement of the direction of arrival
and the increase of signal noise ratio by the vector sensor, theoretical analysis shows that the
vertical particle velocity has different normal mode distribution in comparison with the pressure,
which can be used to monitor the ocean environment. This paper reviewed several vector sensor
experiments performed in shallow water. The collected data show that the transmission loss and
the waveform of the vertical particle velocity diverge from that of the horizontal particle velocities
which is similar to the sound pressure. With the data analysis, a vector sensor array based geoacoustic inversion method has also been developed. The results indicate that the uncertainty of the
inversion can be decreased by a vector sensor array in comparison with that by a hydrophone array
only. [work supported by the National Natural Science Foundation of China No. 11125420].
characterization
Qunyan Ren, Jean-Pierre Hermand
Presenter: Qunyan Ren
A sequential Bayesian filtering technique is introduced to estimate sediment geoacoustic
parameters through processing of multi-frequency vertical impedance (VI) data due to a moving
ship as observed on a drifting receiver. The VI is defined as the ratio of pressure and vertical particle
velocity, and highly correlated with environmental properties. The particle filtering (PF) technique
is used here owing to its robustness in dealing with highly nonlinear problems as compared to other
sequential filtering techniques such as the extended Kalman filter and unscented Kalman filter. The
PF also does not require a restricted state space model and local linearization. A set of particles is
used to estimate the states (here geometric and environmental parameters) and their uncertainties
as described by posterior probability density (PPD). Each particle has a weight representing its own
probability and the whole particle set is evolved with sequentially received acoustic data. The
evolution makes the technique capable of successively updating the estimates when processing
continuous data set, here, ship noise. Preliminary synthetic tests are performed based on
environmental and acoustic data collected at the mouth of the north channel of the Amazon
(Hermand & Vinzon, 2012). The PF smoother is shown to provide accurate PPDs for the estimated
(here static) parameters of density and thickness of a fluid mud layer and water depth, whose
maximum a posteriori solutions are statistically close to the respective true values. The promising
results obtained with PF and the intrinsic characteristics of sequential Bayesian filtering to
219
UA2014
continuously track parameters of interest and quantify uncertainty suggest it can be applied to
other complex situations involving varying measurement geometries and environmental
conditions. [Work supported by Fonds de la Recherche Scientifique (FNRS)]
220
UA2014
Session 33: Bioacoustics
Bioacoustic Absorption Spectroscopy of physoclists
Orest Diachok
Presenter: Orest Diachok
This paper describes the results of a recent multi-disciplinary Bioacoustic Absorption Spectroscopy
experiment, which was focused on the effects of North Pacific hake, Merluccius productus, a
physoclist, on bio-absorptivity. Physoclists are a class of fish that can control the amount of gas in
their swim bladders. The volumes of adapted, i.e. fully filled swim bladders of physcolists are
independent of depth. This experiment was conducted near the shelf break off Oregon in August
2012 at a mesoscale biological hot spot, which was identified through echo sounder surveys and
trawls conducted by the Northwest Fisheries Science Center (NWFSC). The experiment included
coincident measurements of transmission loss (TL) vs. frequency (f) at 0.3 < f < 5 kHz, fish layer
depths, fish length distributions, and continuous temperature profiles. TL measurements were
conducted several times between a moving, ship-deployed, broadband source and a 24 element
vertical array at ranges between 0.1 and 10 km. Length distributions of hake, the dominant species
in this region, were measured by NWFSC. Measured resonance frequencies, which were attributed
to hake decreased with time during the day, and approached values, which were consistent with
theoretical calculations of fully filled swim bladders at night. These results are in accord with
previously reported laboratory measurements, which indicate that the swim bladders of physoclists
are generally adapted at night, and that the volumes of their swim bladders generally increase with
time and approach adapted volumes during the day.
Bluefin tuna behavioural response to anthropogenic noise in floating sea cages
Vicent Puig, Victor Espinosa, Ester Soliveres, Pedro Poveda, Fernando De La Gandara,
Jaime Ramis, Patricia Ordonez, Isabel Perez-Arjona, Jose L. Cort
Presenter: Victor Espinosa, Vicent Puig
The behaviour of bluefin tuna in feeding cages has been continuously monitored by means of
vertical echosounders and video cameras after purse seining. Long term monitoring reveals daily
reaction patterns to aquaculture plant boat navigation and feeding operations. Behavioural
changes when exposed to high levels of anthropogenic noise, like marine wind turbine recordings,
are reported together with a highly adaptive character to external inputs.
221
UA2014
Acoustical biomass estimation results in Mediterranean aquaculture sea cages
Ester Soliveres, Vicent Puig, Patricia Ordónez, Isabel Pérez-Arjona, Miguel Ardid, Jaime
Ramis , Pedro Poveda, Vicent D. Estruch, Dolores Lópezc, del Mar Agrasod, Rocio
Roblesd, Víctor Espinosa
Presenter: Victor Espinosa
Fish size and total biomass monitoring in Mediterranean off-shore cages have been investigated in
production conditions. Good results for total biomass estimation are reported for medium values
of sea-bream densities. Consistent sea bass large range size monitoring is possible from dorsal
target strength measurements, in apparent contradiction with previous works with other
commercial species. A novel approach from full-waveform echo-sounding permits to monitor small
size changes from direct acoustical measurement of fish biometrics.
Sound propagation over an elastic bottom –particle motions caused by seismic
interface waves
Jens M. Hovem
Presenter: Jens M. Hovem
Particle motion sensitivity may be important for fish responding to low frequency anthropogenic
such as sounds generated by piling and explosions. This article discusses particle motions of seismic
interface waves generated by low frequency sources close to solid rigid bottoms. The interface
waves are transversal waves with slow propagation speed and characterized with large particle
movements, particularity in the vertical direction. The waves decay exponentially with distance
from the bottom. The interface waves may be important to include in the discussion when studying
the impact of low frequency anthropogenic noise at generated by relative low frequencies, for
instance by piling and explosive charges, airguns and subsea construction works.
Acoustics marine survey of distribution and intensity of species in Arabian Sea
during monsoon variation (by mid-water trawl survey)
Karthikeyan, Abhay
Presenter: Karthikeyan Muniraj
The study investigated the stability of using the acoustic and catch data gathered during mid- water
trawl surveys conducted in SEAS- South East Arabian Sea and NEAS- North East Arabian Sea (Indian
Ocean) during the respective monsoon period. The main goal is to improve abundance and
distribution estimation of species of the OMZ in particular for mesopelagic fish species. First we
describe the methodology used for acoustic data collection and editing, fish species identification
and problems encountered. Next we compare the acoustic estimates (backscattering
coefficient,S_a) for different fish species with the catch data fall mid-water trawl tows. We also
present acoustic estimation along the survey tracks. The integrated analysis of acoustic and catch
data was successful in (i) detecting and differentiating among several groups of organisms across
222
UA2014
the SEAS and NEAS (Indian Ocean) including zooplankton, pelagic and myctophids (ii) Providing
proxy estimates of density and distribution of different groups of fish.
223
UA2014
Session 34: Underwater Communication
Session 34a
Data Smoothing algorithms for Phased Array Doppler Log
Songzuo Liu, Gang Qiao,Zongxin Sun, Wei Sun, Lu Ma
Presenter: Songzuo Liu
Doppler log is one of the navigation equipments on ship. With the development of marine
technology, more and more countries and institutions began to study it, especially the one for deep
sea environment. Velocity compensation was once a bottleneck to improve the accuracy. The
application of phased-array technique on sonar solved this problem in theory, which made the
performance of Doppler log greatly improved. The precision of speed which is measured by Doppler
log is high, but in some situations there are some outliers of the measured doppler value. In order
to obtain the high accuracy speed result, some smooth processing algorithms should be applied to
the speed estimated. The commonly smooth method is interchange method, average value filtering
and so on, but these methods are only suitable for that wild value spots are few. Three methods for
smooth filtering that moved window and weighted mean method, cylinder sieve method, kalman
filtering are studied in this paper. The relation of outliers’ elimination and tracing speed are
considered. The situation of the lake test is introduced, and the performances and application
conditions of the algorithms are analyzed based on the experimentation condition. Finally, Kalman
filter is realized on the real-time processor and work well.
Comparison of Modulation Techniques for Parametric Underwater
Communications
Karsten Wiedmann, Tobias Weber
Presenter: Karsten Wiedmann
In parametric underwater communications, nonlinear effects occurring during the wave
propagation in the underwater acoustic channel are particular used for data transmission. One
point of interest in parametric underwater communications is that different modulation techniques
can be applied to generate identical received signals, which is inherently due to the underlying
nonlinear channel.
This paper shows that different modulation techniques which create identical received signals
achieve different signal conversion efficiencies. Consequently, different transmit powers are
required to excite the nonlinear signal generation to the same amount.
One outcome is that for quadrature-amplitude-modulation the transmission of two modulated
signals using correlated data sequences outperforms the conventional approach, which is the
transmission of one modulated signal and an additional carrier.
224
UA2014
The results are verified by measurements, which are conducted using a model parametric
communication system in air.
Inquiring Flooding Algorithm For Underwater Acoustic Sensor SelfOrganization Network
Xiao Dong, Wei Liping, Chen Yan, Chen Geng, Ma Li
Presenter: Xiao Dong
Usually, UASN (Underwater Acoustic Sensor Network) is constructed by randomly deployed sensor
nodes. In order to construct network of certain function with these nodes, self-organization
algorithm is needed. There are many self-organization algorithms for WSN (Wireless Sensor
Network) onshore. But, in underwater acoustic communication, the phenomena, such as: severe
attenuation, high background noise, limited bandwidth, large time delay, complicated multi-path,
etc., make UASN different from WSN. Most self-organized algorithms for WSN are hard to be
applicable to UASN. For example, in OPNET simulation of flooding self-organization algorithm in
UASN, it takes an unreasonable long time to establish an effectual network, which means each
node could be accessed by others. Even worse, the network can not be established without
reference to the broadcast times of each node. An improved self-organization algorithm is
proposed, which appends an inquiry process to flooding self-organization algorithm. It is proved by
OPNET simulation that under the same conditions an effectual network could be successfully
established in a shorter time than simple flooding and probabilistic flooding algorithms, and the
energy consumed in self-organization period is decreased.
DSP implementation of Turbo Equalization based Underwater Acoustic Modem
Bo Peng, Hefeng Dong
Presenter: Bo Peng
In this paper, the real time implementation of frequency domain turbo equalizer for single carrier
underwater acoustic communication is investigated. The whole receiver system is implemented on
a floating point DSP TMS320C6727B board. Through careful adjustment, the real time frequency
domain turbo equalizer with burst symbol rate of 2 ks/s is realized with relative large margin. The
performance of the system is demonstrated by a tank experiment at NTNU. The system works at
relatively low SNR. With respect to the phase rotation due to CFO estimation error, a joint turbo
phase tracking and equalizer is proposed. Its performance is also demonstrated through the same
experiment data with off-line analysis. It improves the system performance around 2 dB and can
be easily implemented in real time with respect to current DSP processing burden.
225
UA2014
Orthogonal Multicarrier Underwater Acoustic Communication Experiments in
River, Lake and Shallow Sea
Feng Zhou, Gang Qiao, Yanling Yin, Lu Ma, Songzuo Liu, Zongxin Sun
Presenter: Feng Zhou
The purpose of this research is to test the algorithm’s performance of the OFDM system in a multisystem (spread spectrum & OFDM) prototype for underwater acoustic communication(UAC)and
assess whether the prototype meets the required criterion. The experiments were conducted in
three water types, including river, lake and shallow sea and were conducted in 2011.
The prototype used in the experiment had two communication systems. One was underwater
acoustic spread-spectrum communication (UASSC). The other was underwater acoustic OFDM. The
robust UASSC was used to transmit orders and status, so in the prototype the OFDM parameters
were modified by orders conveyed by UASSC. This meant that the OFDM parameters, including
points of FFT, cyclic prefix(CP), cyclic postfix, length of LFM(sync signal), pilot patterns and interval,
frame of signal and coding, could easily be adjusted as required according to the channel.
The details of the three tests will be outlined in this paper, including test location, test time, water
characteristics, and weather at the time of the test. The results of each experiment will be provided
and the results analyzed including consideration of the conditions at the time of the tests.
The results and analyses show that with enough SNR and without an obvious multipath the
performance of OFDM was mainly influenced by channel variations and the doppler spread of
signals. In the complicated UA channel, robust in more water areas, alterable parameters and
effective equalization are required.
Study on the impact of multi-path channel to the horizontal uniform linear
array beamforming
Jidan Mei, Dajun Sun, Yunfei Lv, Junjie Shi, Tingting Teng, Hualin Lan
Presenter: Jidan Mei
The horizontal uniform linear array beamforming technology can be used for measuring the
direction of arrival (DOA) of target of interest. In the free field space, it has good performance.
However, in the ocean due to the interference’s impact, the underwater acoustic channel has multipath characterization, which may have some influence on the DOA estimation. When the target is
far from the array, the multi-path azimuths are all approaching to the true target azimuth. Here
the multi-path impact can be ignored. But when the target is in a short range to the array, the
multi-path impact has distinct influence on DOA estimation and can’t be omitted. In this paper the
multi-path impact was analyzed theoretically based on the ray propagation model. And the
theoretic results were validated through numerical simulation and lake experiment, in which the
pulse signal form was used for convenient analysis. Theoretical analysis indicated that the multipath impact is related to the target range, DOA, the vertical distance from the array reference point
to the equiphase plane. In detail, the closer target range, the more deviation from the broadside of
the array and the longer the vertical distance from the array reference point to the equiphase plane,
226
UA2014
the worse influence subjected to DOA estimation. The simulation and experiment results all
validated the theoretical analysis.
227
UA2014
Session 34b
Development/Introduction of the bio-logging system to realize high data
recovery rate using acoustic communication
Yoshinori Miyamoto,Kazushi Miyashita, Nobuaki Arai, Takashi Kitagawa, Keiichi
Uchida, Hiromichi Mitamura, Toyoki Sasakura
Presenter: Yoshinori Miyamoto
Project to build a new bio-logging system was started in order to elucidate the dynamics of
populations and community of the apex predator fish in the open ocean. One of the goals is to
enhance the data collection of individual information by utilizing acoustic communication
techniques. And ultimately, to build a new bio-logging system using acoustic communication and
to complete the evaluation system will be applicable for the open sea. By using this new system, if
individuals equipped with the micro data logger are close to each other, the data recorded in the
loggers can be shared by mutual acoustic communication (Inter-individual communication logger).
Hence, it is possible to recover data from other data-logger, if it is difficult to recover multiple
loggers. In addition, acoustic receiving system designed for various platforms (eg. low-orbit small
orbiting satellites (Iridium) and mobile phones), will make this system a multi-platform correspond
data receiving and collecting system.
By applying different kind of gears (eg. sink-float type that utilizes mooring type, fishing boats,
fishing gear, and Argo floats, large biological-mounted) in this multi-platform correspond receiving
system, data collection in the open sea will be possible even without recovering the logger itself.
For this, a 20-50% decrease in the comparable conventional cost can be expected. Here, we
introduce the basic technology and protocols of this newly developed system and report the
verification of the method of transmitting the information, and the communication distance.
Full-duplex, relative clock based and collision free protocol for underwater
acoustic networks
Gang Qiao, Jiarong Zhang
Presenter: Gang Qiao, Jiarong Zhang
This paper presents some experimental results of underwater acoustic communication networks
both in lake trail and sea trail. Three special designed MAC protocols were introduced and tested:
full-duplex based protocol (FD-MAC), relative clock based protocol (RC-MAC) and collision free
protocol (CF-MAC). FD-MAC protocol was implemented on a full-duplex modem, vector hydrophone
was used to improve the receiving gain, and dual-mode communication and acoustic baffle
technology were applied to reduce the local emission interference. Experiment results show that
the exposed/hidden terminal problems can be well resolved within the cover range of 200 m. RCMAC protocol was an improved TDMA protocol suited for centralized networks in the underwater
environment, neither the global clock synchronization nor the periodically broadcasting of
228
UA2014
synchronize signal is needed. Net schedule was arrange by the main node according to the
propagation delays and the sub nodes were working on their local clock by mapping the net
schedule to their local time line separately. Working slot for each sub nodes can be customized
according to the information flow to improve the network throughput and the main node and sub
node can be reversal to prolong the network lifetime. Experiment results show that this protocol
was easy to be realized and work well. CF-MAC protocol was designed for the regional underwater
observation networks. This protocol avoids channel contentions by using the top-down channel
assignment. In this scheme, temporary channel access is assigned by the gateway/AUV using a
require-data-send signal (RDS). RDS is used to wake up the specific node and cannot be heard by
the other nodes. Two types of acknowledgement signal were designed to shorten the poll duration
and save energy. Practicability of this protocol was validated by the experiments. All modems that
we used in these trails are designed and implemented by our lab.
229
UA2014
Session 35: Acoustic Modelling
Session 35a
Comparison of RAM-based Calculations of Acoustic Propagation with Measured
Data in the East China Sea
Clare Nadig, David Bradley
Presenter: Clare Nadig
Towed CTD measurements and recorded acoustic transmissions taken during the Transverse
Acoustic Variability Experiment (TAVEX) in the East China Sea in 2008 are used to examine the
individual and combined effects of density-compensated thermohaline variations (spice) and
diffuse and propagating internal waves. The acoustic propagation conditions were long range (2034 km) and shallow water. Sound speed fields are constructed using two components: first,
measured profiles of real travelling internal wave packets; second, randomly generated variations
based on the observed characteristics of the diffuse waves and spice. Propagation through these
fields is modeled using parabolic equation simulations and compared with acoustic data recorded
during the experiment at 300 and 500 Hz.
Single global empirical equation for prediction of the peak pressure level of
airgun arrays signals in different marine environments
Marta Galindo-Romero, Alexander Gavrilov, Alec J Duncan
Presenter: Marta Galindo-Romero
Peak pressure and sound exposure levels of impulsive signals produced in seismic surveys using
airgun arrays must be estimated prior to undertaking the survey. The sound exposure level
(proportional to signal energy) can be accurately predicted with existing underwater acoustic
models if the environmental parameters are known. However, these models do not correctly predict
spatial variations of the peak pressure. Empirical analysis of recordings from three seismic surveys
in different marine environments was conducted, including (1) a highly range-dependent
environment in deep water over the continental slope off South Western Australia, (2) a nearly
range-independent environment of shallow water, over a calcarenite seafloor, and (3) a range
dependent environment in medium water depth over a layered elastic seafloor. The analysis
showed strong correlation between the peak pressure level and the sound exposure level in all
cases, with similar coefficients of linear regression. This paper presents a single linear equation to
predict the peak pressure level from the sound exposure level in different marine environments. The
difference between the empirical prediction of the peak pressure and the measurements follows a
nearly normal distribution of about 0-dB mean and 1.6-dB standard deviation. When
measurements are not available, the peak pressure level can be approximately predicted by
applying the linear equation to the values of sound exposure level predicted with an acoustic model
230
UA2014
suitable for each environment. For the deep-water environment, the difference between the
predicted and measured peak pressure levels is also nearly normally distributed with the mean and
standard deviation of 0.09 dB and 2.79 dB respectively. The peak pressure levels predicted directly
by a parabolic equation model exceeded the measured values by more than 10 dB at distances
greater than 48 km. Therefore, this empirical equation offers a significant improvement in
estimating the peak pressure level of airgun signals.
Peculiarities of the broadband interference pattern in a shallow sea with a
substantially varying bottom relief
E.L. Borodina, Yu.V. Petukhov
Presenter: E.L. Borodina
The focus of this paper is on the space - frequency interference patterns created by broadband
sources in shallow water environments with substantial variation of the bottom bathymetry. The
experimental data, discussed herein, were collected in the region characterized by the sharp coastal
slope connecting the shallow and relatively deep water areas with the seamounts located in the
latter. It was found out that when the source is located in the uniform shallow area, the space frequency interference structure typical for the shallow sea is formed in the deeper non - uniform
area at the certain distances from the slope and the corresponding frequency interval. This
phenomenon can be explained by formation of the fairly narrow directivity diagram of low - order
modes in the shallow water area, at the shelf boundary. When the source is towed in the deep part
of the waveguide, the interference structure formed in the latter is entirely determined by signals
scattered by the coastal slope and, only to a minor extent, by the elevations of bottom relief. At
that, in the shallow area, the interference structure is formed by signals scattered by the elevations
only.
Acoustic remote sensing of internal Kelvin waves dynamics in a stratified lake
Boris Katsnelson, Andrey Lunkov, Ilia Ostrovsky
Internal Kelvin waves (and any internal seiches in general) have a great ecological significance in
stratified shallow water basins with an area of about a few thousands of square km (lakes, gulfs,
bays) since they induce mixing, resuspension and material transport at the basin periphery,
affecting chemical regime and ecosystem productivity. Acoustic methodology is suggested for the
parameterization of the basin-scale internal waves. The effects of internal Kelvin waves (IKWs) on
spatiotemporal variability of the mid-frequency (0.5 to 1.5 kHz) sound field in a deep (~40 m)
stratified lake , such as horizontal shift of the interference structure and frequency shift at a single
receiver, are studied in numerical experiments. It is shown that the IKWs cause significant variations
of the sound field which can be easily measured using linear array or single receiver, respectively.
Basic relations connecting interference pattern shifts and IKW parameters are provided. These
relations can be utilized for IKW reconstruction.
231
UA2014
Work was supported by RFBR and BSF
On the importance of uncertain sea bottom parameters for the prediction of
pile driving noise
Tristan Lippert, Kristof Heitmann, Marcel Ruhnau, Stephan Lippert, Otto Von Estorff
Presenter: Tristan Lippert
The topic of hydro sound radiation from industrial offshore pile driving has gained a lot of attention
lately. The massive increase in constructed and planned offshore wind farms, especially in the North
Sea, leads to a new dimension of hydro sound immission both in quality and quantity. For the impact
assessment of future farms and the optimization of insulation systems, reliable numerical
predictions of the resulting sound pressure levels (SPLs) are needed. In this contribution, a hybrid
approach of a finite element (FE) model coupled to a wavenumber integration (WI) approach is
used: The complex phenomena in the vicinity of the pile can be modelled very detailed using the FE
model, while the prediction of SPLs for large source receiver separations, can be performed
efficiently by the WI method. After a brief introduction to the model, a validation by the comparison
to measurements in an offshore wind farm is performed. Here, different model environments are
used to represent the sea bottom and their effect on the prediction accuracy is investigated.
Subsequently, the problem of parameter uncertainties in the context of underwater noise
predictions of offshore pile driving is discussed. Looking at the sea bottom, the degree of
uncertainty is relatively high, due to the size of the domain of interest. To address this issue, MonteCarlo simulations are performed, varying the bottom parameters and the resulting SPL, probability
distributions in the water column are evaluated.
Modeling pulse propagation in a wedge environment with range-dependent
geoacoustic parameters
Melanie E. Austin, N. Ross Chapman
Broadband pulses were transmitted across a slope to a range of 10 km in an experiment in the
Florida Straits in 1999. The pulse structures of the long-range received data consisted of secondary
arrivals whose arrival times were consistent with an effect of horizontal refraction from the sloping
seafloor. Three dimensional sound propagation modelling was conducted to characterize the outof-plane propagation paths for the experimental geometry. A full-waveform pulse simulation
confirmed that the secondary arrivals could be attributed to horizontally refracted sound paths, but
the arrival times of the horizontally-refracted sound paths were not well matched by the model
simulation. Range-dependent geo-acoustic models of the three dimensional coastal environment
were generated to investigate the influence of the seafloor properties on the characteristics of the
out-of-plane arrivals. The models were introduced into the 3D propagation model to simulate the
broadband signal structure at 10 k. A geoacoustic model with parameters changing in the upslope
direction improved the match of the modelled arrival structure with the measured data.
232
UA2014
Session 35b
Finite Difference Time Domain Method for Acoustic Waves in Attenuate and
Absorptive Medium for Layered Underwater Acoustic Environments
J. Lakziz, S.Othmane, S.Ouaskit
Presenter: S.Ouaskit
In this work, we use Finite Difference Time Domain (FDTD) method to investigate the underwater
acoustic wave propagation problem; in order to understand the acoustic wave propagation
properties and determining the effects of of the attenuation and absorption for the sound intensity
in the sea.
The results are presented for calculated Transmission Loss (TL) of a propagating acoustic waves in
a layered underwater acoustic environments by using Finite (FDTD) method. In a first step; The sea
environments are modeled as one layered medium before generalizing the resolution to two
layered (fluid-fluid and fluid-solid) medium. The sea surface and the sea bottom are considered to
be planar surfaces. The discritized acoustic wave equations are used in the algorithm of the FDTD
method and Perfectly Matched Layer absorbing boundary condition is applied to eliminate
numerical reflections from the ends of the grid .Two cases are discussed (Taking account of
absorption and attenuation).The good performance of this method is validated with the results of
KRAKEN program and by comparison with the analytical method.
Sound fluctuations in the presence of nonlinear internal waves moving along
acoustic track in shallow water
Boris Katsnelson, Valery Grigorev, Jixing Win
Theoretical analysis and numerical modeling is carried out for low-frequency (<1kHz) acoustic
signals propagating in shallow water waveguide in the presence of nonlinear internal wave
traveling approximately along acoustic track. Fluctuations of the sound amplitude at the receiver
in this case are provided by modes coupling (in modal theory) or ray scattering (in ray
approximation). Modeling is carried out within the framework of modal theory and PE
approximation. It is shown that in spectra of amplitude fluctuations there are characteristic
frequencies, proportional to speed of internal waves. Maximal amplitude of fluctuations in
spectrum takes place at the so called predominating frequency, determined by the scale of
interference beating of modes, having the turning point in area of thermocline (or by the cycle of
rays, touching thermocline). Depth dependence and frequency dependence of fluctuations is
studied, rime-frequency diagram is considered as well. Results are compared with experimental
data.
Work was supported by RFBR and BSF
233
UA2014
Statistical Characterization of Wideband Channel Impulse Response
Observations in Shallow Water
W.S. Hodgkiss, H.C. Song, D.E. Ensberg
Presenter: W.S. Hodgkiss
The Kauai Acomms MURI 2011 (KAM11) Experiment was conducted in shallow water (~100 m deep)
off the western side of Kauai, Hawaii, in June-July 2011. Wideband channel impulse response
transmissions (24 kHz bandwidth LFM chirps and MLS centered at 23 kHz) were carried out every
2 hours for an extended period of time. These were transmitted from a moored 8-element source
array with 7.5 m element spacing and carried out in a round-robin fashion for 60 s per source
element. A pair of 16-element receive arrays with 3.75 m element spacing were moored at ranges
of 3 and 7 km. The fixed source, fixed receiving array geometry enabled observing environmentallyinduced fluctuations in the channel impulse response. The experiment region exhibited substantial
daily oceanographic variability. The mixed layer depth changed from as little as 20 m to as much
as 60 m or more over the course of 24 hours. Similarly, the wind speed and sea surface conditions
exhibited a daily pattern. Environmental data collected included continuous water column
temperature structure measurements (thermistor strings near the source and receiving arrays), sea
surface directional wave field (waverider buoy), and local wind speed and direction. The paper will
include selected examples of the temporal variability of the wideband channel impulse response for
various source-receiver pairs under different sea surface and sound speed conditions. Individual
eigenray paths will be identified and their fluctuation characteristics quantified including temporal
correlation scales as well as the relatedness of fluctuations between pairs of paths. In addition, the
variation of these statistical characterizations will be shown as the sea surface and water column
environment evolve over time. These path and cross-path characteristics then can be used for
tapped delay line model channel simulation purposes.
Modeling Reverberation Time Series Based on Full Wave Reverberation model
J.R.Wu, L.Ma
Presenter: J.R.Wu
Traditionally, reverberation time series was simulated by cell-scatter model or point-scatter model.
A new method has been proposed in this paper to simulate reverberation time series. The
reverberation time series was simulated based on the full wave reverberation model we have
developed. The effect scattering area was modeled firstly, then the initial signal was convolved with
the effect scattering area to get the scattering kernel, finally, sound propagation factors were
included in the method. Numerical analysis shows: the new reverberation time series simulation
model is faster than the cell-scatter model and point-scatter model. At the same time, this new
method has the clear physical picture, and can be used for reverberation time series simulation in
sonar performance model.
234
UA2014
Acoustic Backscattering from Layered Interfaces using Finite Elements
Marcia Isakson, Anthony Bonomo, Nicholas Chotiros
Presenter: Marcia Isakson
Understanding acoustic backscattering from ocean sediments is vital to understanding target
detection on the seafloor, bathymetric mapping and sediment classification. Modeling
backscattering from layered sediments is particularly challenging because of the complicated
coupling among the different modes and the possibility of multiple scattering between the layers.
In this study, a finite element model is proposed to study the backscattering from a variety of
sediment layers including fluid, elastic and poro-elastic. Results will be compared with
approximations to the Helmholtz-Kirchhoff equation including perturbation theory and the small
slope and Kirchhoff approximations. Special emphasis will be placed on the range of validity of
approximations with respect to the surface RMS height to acoustic wavelength ratio. [Work
sponsored by ONR, Ocean Acoustics.]
235
UA2014
Session 36: Acoustic Inversions
Location: Lecture Room D, Friday 27th June 2014, 8:30 - 9:50
Geoacoustic inversion using pile driving pulse and surface ship noise of
opportunity based on single vector sensor
Dajun Sun, Junjie Shi, Yunfei Lv, Hualin Lan, Jidan Mei
Presenter: Dajun Sun
A geoacoustic inversion method using pile driving pulse and surface ship noise of opportunity based
on single vector sensor was presented here to estimate geoacoustic parameters. Firstly, the
propagation between a couple of underwater source and receiver in the case of a real waveguide
was numerically studied via a predefined cost function to analyze the geoacoustic parameters’
sensitivity to the cutoff frequency and interference range. Then, the cutoff frequency and
interference range were estimated on the real data using transformation like short-time Fourier
transform for the pile driving pulse and surface ship noise respectively. And finally the best-fit
solutions involving sediment sound speed and its thickness and basement sound speed etc were
found using the defined cost function about the cutoff frequency and interference range. The idea
was effectively validated during the experiment that took place in June of 2013 in South China Sea
nearly 100m depth. [Work supported by the National 863 Project (No. 2011AA090502) and Defense
Industrial Technology Development Program (B2420132004).]
Inferring ocean temperature variations from shipping noise
Ana Bela Santos, Paulo Felisberto, Sergio M. Jesus
Presenter: Ana Bela Santos
Acoustic passive methods to infer properties of the oceanic medium have been subject of
progressive emphasis in order to obtain low cost, environmental friendly, long time characterization
of the ocean. To this end a first step consists in the estimation of the frequency response of the
medium or its time domain counterpart, the impulse response. In this work we consider distant ship
noise as an opportunity source characterized by a few low frequency discrete tones. Therefore, the
frequency response of a shallow water acoustic channel is estimated at these discrete frequencies,
between two vertical line arrays (VLA's). The influence on the estimates of several factors of
uncertainty such as range, depth and tilt variations in the VLA's is investigated. For validation
purposes, a phase conjugation method is considered. In a preliminary approach towards passive
ocean acoustic tomography implementation the estimates obtained by the proposed method are
applied in a matched field framework to track sound speed profile variations. Simulations are
conducted based on experimental setup and environmental parameters gathered during the
MREA07 sea trial that took place in the Tyrrhenian Sea, near Elba Island in May 2007. The results
show that, although shadowed, the obtained frequency response estimates allows to obtain a
permanent focus and enables the tracking of sound speed profile variations in the water column.
236
UA2014
Marine mammal’s directivity in geoacoustic inversion scheme
Amelie Barazzutti, Cedric Gervaise
Presenter: Amelie Barazzutti
Gervaise & al. UAM 2011 and Barazzutti & al. 2013, RTS described the general structure of a scheme
to estimate the nature of superficial sediment in shallow waters using marine mammal’s whistles
and a single receiver. The multipath structure of calls resolved by spectrogram is used to estimate
the source characteristics and the superficial sea bottom features. A field application of this method
was presented in Barazzutti & al. UAC 2013 using controlled signals similar to marine mammal’s
vocalizations in a shallow water environment on a sandy bottom (experiment using 10ms duration
chirps with [1kHz-2kHz] bandwidth in the Gulf of Lion, ~ 100 meters water depth).
However, contrary to the source used during that experiment, marine mammals are directive
sources and the directivity loss underwent by the multipath must be taken into account in our
inversion process. Indeed, the directivity, function of frequency and emission angle (sound-source
azimuth), impacts each path differently according to its emission angle. Thus the bottom path, once
corrected from transmission loss, must be corrected from directivity loss before being used to
estimate the bottom feature. The emission angle can easily be geometrically related to the
accessible arrival angle and a specific unknown angle we called “attitude” (source orientation in
space during the emission). However, the directivity patterns of marine mammals are not well
documented yet, especially for vocalizations (e.g. directivity model assumption - W.L. Au 1993,
directivity pattern measurement - Au & al 2012, etc.) and contrary to other mammals the unknown
“attitude” parameter is not that easy to observe (e.g. Dantzker &al 1999). Our communication aims
at describing different methods to estimate the “attitude” angle and the directivity loss for marine
mammals. Their performances and limits are evaluated using simulated data.
Bayesian reconstruction of seafloor shape from side-scan sonar measurements
using a Markov Random Field
Philipp Woock, Alexey Pak
Presenter: Philipp Woock
To explore the seafloor, a side-scan sonar emits a directed acoustic signal and then records the
returning (reflected) signal intensity as a function of time. The inversion of that process is not
unique: multiple shapes may lead to identical measured responses.
In this work, we suggest a Bayesian approach to reconstructing the 3D shape of the seafloor from
multiple sonar measurements, inspired by the state-of-the-art methods of inverse raytracing that
originated in computer vision. The space near the bottom is modelled as a grid of voxels, whose
occupancies are represented by random binary variables. Any assignment of occupancies
corresponds to some seafloor shape. A global multi-component energy potential describes how well
the resulting surface agrees with the sonar data and with the a priori assumptions. Minimization
of energy is equivalent to finding the maximum a posteriori (MAP) assignment to this Markov
random field (MRF) and is done using the iterated belief propagation (BP) algorithm.
237
UA2014
The critical step in this method is to compute messages from “factors” representing the sonar
beams to voxels. Naively, its complexity scales exponentially with the number of voxels traversed
by a beam. Unlike inverse raytracing, where a pixel value constrains voxels only along a single view
ray, a sonar beam involves voxels within a relatively wide cone. Employing dynamic programming
techniques and space-filling curves, we were able to develop a practical approximate solution to
this problem.
The algorithm is not restricted to side-scan sonar reconstruction and could be applied to medical
ultrasound or ultra wide-band (UWB) radar imaging.
238
UA2014
Location: Lecture Room B, Friday 27th June 2014, 8:30 - 9:00
From the pole to the equator: Utilizing a screw dislocation in an acoustic
wavefront.
Brian T. Hefner
A screw dislocation in a wavefront is characterized by a phase dependence about the dislocation
axis that varies as $expleft(-i m phiright)$, where $m$ is an integer and $phi$ is the angle about
the axis. This talk discusses two sources which generate an acoustic field with a screw dislocation
but for very different applications. The first is the helicoidal wave transducer which generates a
beam with a screw dislocation along its axis [Hefner and Marston, J. Acoust. Soc. Am. 106, 3313
(1999)]. At the axis, the phase is indeterminate and as a result there is a corresponding null in the
pressure magnitude. The screw dislocation is found to exist in both the far- and near-fields of the
transducer. This null then clearly indicates the axis of the beam at all distances and has the potential
to be used as an aid in the alignment of objects in sonar experiments or other similar applications.
This beam is also shown to carry angular momentum. The second source utlizes a screw dislocation
but far from the null axis. It generates a wavefront in the x-y plane that has a phase which is
proportional to the azimuthal angle about the source ($m$ = 1). This transducer is combined with
an omnidirectional, reference source to produce a spiral wavefront beacon. The phase difference
between these sources contains information about a distant receiver’s azimuthal angle relative to
the beacon and can be used for underwater navigation [Hefner and Dzikowicz, J. Acoust. Soc. Am.
129, 3630 (2011)]. Navigation using this beacon has been demonstrated experimentally and
propagation models have been developed to assess the performance of the beacon for the general
case of propagation in a horizontally stratified waveguide [Hefner and Dzikowicz, J. Acoust. Soc.
Am. 131, 1978 (2012)]. This talk discusses both of these unique sources and their applications as
well as the underlying physics which connects them.
239
UA2014
Posters
Location: Poster Area, Tuesday 24th June 2014, 10:00 - 13:00
Multi-ary code shift keying direct sequence spread spectrum using in
underwater acoustic communication
Sun Zongxin, Yu Yang, Qiao Gang, Zhou Feng,
Presenter: Sun Zongxin
Owing to it’s abilities of anti-noise and anti-fading, Spread Spectrum communication is widely
applied in underwater acoustic field (UWA). But the data rate of Direct Sequence Spread Spectrum
(DSSS) is very low, which cannot meet the needs of long-range UWA communication at present.
This paper proposed a method, named of M-ary CSK-DS, which combines M-ary DSSS schemes with
code shift keying (CSK). Gold sequences, as the basic sequences, occupy a fine periodic auto
correlation function (PACF), periodic cross correlation function (PCCF) and large code set size. This
method utilizes sequence information, code phase information, polarity information. The influence
of PACF and PCCF is analyzed over AWGN and UWA channels to the proposed method. The
performance of M-ary CSK-DS is evaluated through simulation over AWGN and UWA channels. The
results show that M-ary CSK-DS is superior to the conventional methods. In the last part of the
paper, an experiment testified the correctness of simulation.
Direct-sequence spread spectrum underwater acoustic communications with
Turbo equalization in time-varying channels
Jianchun Huang, Shengming Guo, Li Ma, Zhongyuan Guo, Geng Chen
Presenter: Jianchun Huang
A receiver combines direct-sequence spread spectrum (DSSS) and turbo equalization with MAP
algorithm has a potential to provide good performance in under-water acoustic communications.
Channel estimation is an essential part in branch metric computation for the state transition in
MAP. In time-varying channels, multi-path changes rapidly. Conventional MAP algorithm does not
update the channel estimation during information detection process, hence the performance
degrades a lot. In this paper, data-depended update of channel estimation is developed, which is
derived from per-survivor processing (PSP). PSP establishes a trellis, traverses all the state and picks
survivor paths, which can be used for channel estimation. Because no training data is needed, it
can update channel estimation anytime during the information detection process, that is a great
advantage in time-varying channels. PSP establishes the same trellis as MAP does, the method can
be applied to the receiver in this paper. By comparing current branch metrics of all states, the
variation of channel is measured, then time to update channel estimation is decided. Simulation
result shows that the receiver using data-depended update of channel estimation is a good solution
for underwater acoustic communication system in time-varying channels.
240
UA2014
Underwater Acoustic Communication System Simulation Based On Gaussian
Beam Method
Bing Li , Yuan Li ,Hai Bo Zheng ,Zhi Bo Zhang, Qihu Li
Presenter: Bing Li
Due to dynamic multipath propagation structure and Doppler spreading, the underwater acoustic
channel make the performance of underwater communication systems seriously degraded .
Comparing with costly sea trials,underwater acoustic channel models provide a reliable and
cheaper tool for predicting the performance of communication systems.The accuracy and efficiency
of the channel model is essential for prediction. The ray theory based on Gaussian beam method
has merits of high efficiency, clarity in physical meaning, and being easy to be parallel processing .
In this paper, we proposed a novel simulator based on Gaussian beam method to estimate impulse
response function of underwater mobile communication systems.
Simulator operates periodically .At the input port of the simulator ,the evaluated communication
systems will transmit a frame of communication signal every fixed intervals.According to the
relative position relation between transducer and hydrophone in the acoustic field, simulator will
estimate the current impulse response function . At the output port, a frame of communication
signal will be acquired and processed with a mosaic method.When transmission is over,the BER will
be calculated. Parameters such as depth ,sound profile,SNR and SL of transducer can be set up in
the simulator.Simulations and trial are performed to validate this method. The results indicate that
it help to assess the performance of underwater communication systems different in modulation
method.
241
UA2014
Location: Poster Area, Tuesday 24th June 2014, 14:15 - 17:15
The Study of Time Delay Estimation Technology Based On The Cross-spectrum
Method
Zhibo Zhang, Changyu Sun, Yuan Li, Haibo Zheng, Bing Li, Xizhong Bao, Qihu Li
Presenter: Zhibo Zhang, Yuan Li, Haibo Zheng, Bing Li, Qihu Li
Abstract: In Many fields of modern radar and sonar, parameters of the target distance and azimuth
often need to be measured accurately, one of its key technologies is the time delay estimation, time
delay estimation directly affects the accuracy of acoustic positioning effect. Traditional delay
estimation consists mainly of the generalized correlation method , phase-spectrum analysis,
parametric model estimation, and adaptive time delay estimation. Cross spectrum method is a
common method of time delay estimation, the method in high noise environment, can obtain more
accurate estimation of delay; but in low SNR environment, the performance of this method in sharp
decline. Cross spectral method is first transformed into the frequency domain, in order to get higher
accuracy, often need to segment average, one estimation of the time delay often need thousands
of points or even thousands of data points, and the individual outliers will make a serious decline in
accuracy. In order to reduce individual outliers brought by the nonstationarity of signal and improve
the accuracy of estimation, this paper presents an improved cross spectrum time delay estimation
method, theoretical analysis, and gives the specific implementation steps. Computer simulation
results show that, the improved method in the low SNR environment, can improve the estimation
precision. Cross-spectrum time delay estimation method is improved, the optimization in the
conventional cross-spectral method, is less computation complexity, and has a strong practical.
The Study of Passive Ranging Technology Based on Three elements Vector
Array
Haibo Zheng, Yuan Li,Bing Li, Zhibo Zhang,Xizhong Bao,Qihu Li
Presenter: Haibo Zheng
Traditional passive ranging technology based on three elements array is well known. Changes in
the curvature of a spherical wave front leads to relative time delay of each primitive. By measuring
the relative time delay of each base element, the target range and azimuth are estimated. Delay
estimation accuracy, target distance, orientation, aperture of array, array installation accuracy and
other factors have an impact on the ranging accuracy. The most critical factor is the delay
estimation accuracy. Published research literature on ternary array of passive positioning, mainly
focused on the sound pressure information. It is well known that sound wave has both scalar
quantity and vector field, while traditional acoustic pressure sensor system merely makes use of its
acoustic pressure information. Vector hydrophone, also called combined sensor, is combined by
traditional and omni-directional pressure hydrophone and natural dipole independent on
frequeney, which can co-locating and simultaneously measures pressure(scalar field)and particle
velocity(vector field)of acoustic field. This paper presents the passive ranging technology based on
three elements vector array, representing Traditional passive ranging technology based on three
242
UA2014
elements array, more fully applied sound field information to improve the accuracy of delay
estimation, and ultimately improve the passive ranging accuracy.
Adaptive Despeckling Method For Sas Images In NSCT Domain
Sheng Yan, Minggang Liu, Chaohuan Hou
Presenter: Sheng Yan
The synthetic aperture sonar (SAS) is an attractive high-resolution underwater acoustic imaging
technique and widely used in seafloor imaging. Due to the coherent nature of scattering
phenomena of the SAS imaging, a type of multiplicative noise called speckle affects the further
processing and understanding to SAS images. To despeckle in SAS images, a nonsubsampled
contourlet transform (NSCT) based adaptive despeckling method is presented in this paper. The
NSCT is a flexible multiresolution, multidirection, and shift-invariant image decomposition
transform that can be used to separate the speckle from SAS images. In NSCT domain, each high
frequency subband is adaptively divided into three types of regions: the targets, the seafloor, and
the shallow. Different strategies are selected to despeckle in these three types of regions to reduce
speckle while preserve detail information of SAS images. The experimental results of SAS images
despeckling show that the proposed method has better performance in speckle reduction and
targets contour preservation than other two comparing methods.
243
UA2014
Location: Poster Area, Wednesday 25th June 2014, 10:00 - 13:00
Influence of ship radiated noise level directivity on the assessment of
underwater noise maps
Thomas Leissing, Christian Audoly, Celine Rousset
Presenter: Thomas Leissing
Recent directives outline the need to mitigate underwater noise footprint due to shipping, to
prevent negative consequences to marine life. The need is becoming even more acute because of
shipping traffic, which increases steadily. In that context, the goal of the EU project AQUO is to
provide policy makers with practical guidelines, in order to mitigate underwater noise footprint due
to shipping. The retained definition of underwater noise footprint involves three different
quantities: (i) the radiated noise level of a ship, which is a physical, measurable quantity, used to
characterize a ship as a source of underwater noise; (ii) the underwater noise map due to shipping,
which is a physical quantity used to determine underwater noise due to shipping level in a maritime
area and (iii) the underwater noise impact on marine life. In the objective of the assessment of the
underwater noise footprint, the ship underwater radiated noise patterns must be accurately
evaluated. Whether or not accounting for the directivity of underwater radiated noise is a matter
of concern, since some ship types present significant directivity in the horizontal plane, even at low
frequencies. We consider in this paper three ship types (fishing vessels, merchant ships, fishing
research vessels), that show very different directional behaviour. Simulations on various test cases,
mixing different ship types and operating conditions, allow bringing into focus situations where
horizontal directivity may play an important role in the assessment of the underwater noise maps
and on related noise footprint indicators.
The fluid noise analysis on underwater high speed small vehicles
Lanyue Zhang, Yunsheng Li, Sichun Li, Lijing Sun, Bo Wang
Presenter: Lanyue Zhang
Fluid noise is one of the three most important kinds of noise source. When the small underwater
vehicles move in high speed, fluid noise is made up of fluid-induced vibration noise and the noise
radiated by the perturbation of turbulence boundary layer. In order to research the rules of fluid
noise radiated by small high speed underwater vehicles, the theories of turbulence and finite
element method (FEM) were used. The main factors that have influence on the fluid noise were
researched through the theoretical calculation and stimulation based on a model of underwater
high speed small vehicle. The results showed that the model made of aluminum would radiate
higher noise than that made of steel and the thickness of the model is bigger, the fluid-noise is
lower. Furthermore, the noise radiated by the perturbation of turbulence boundary layer will
increase with the ascending of flow velocity, and the shape of the vehicles' head will also influence
on fluid noise.
244
UA2014
The comparison of two way sound propagation in Hashirimizu Port
Hanako Ogasawara, Kazuyoshi Mori, Toshiaki Nakamura
Presenter: Hanako Ogasawara
A Reciprocal sound propagation experiment has been conducted in Hashirimizu Port, Japan.
Hashirimizu Port has only 6 m of the depth in maximum and the distance between the transceivers
are about 120 m. The experimental area is very shallow water. As the received sound includes many
of reflected and refracted waves from seafloor and sea surface, there were two high amplitude
signals which comes directly from the transmitter and reflected from the bank along the
propagation path. Authors compare these two arrival signals and explain the environmental
changes such as water temperature, at the experimental area.
245
UA2014
Location: Poster Area, Wednesday 25th June 2014, 14:15 - 17:15
A combined GIS-2DFTT multi-parameter analysis of very high resolution
bathymetric data: a case study from the Venice lagoon
A. Kruss, F. Madricardo, J. Tegowski, F. Foglin, L. Janowski
Presenter: A. Kruss
Coastal and transitional environments undergo strong morphological changes due to natural and
anthropogenic pressure. In these environment the bathymetric surveys are extremely important in:
a) the monitoring of the long term environment evolution; b) the managing of the changes. The
recent technological development of the multibeam systems enables them to achieve very high
performances also in the very shallow waters of coastal and transitional environments.
The morphology of the extremely shallow Venice Lagoon, that surrounds the historical city of
Venice, is changing rapidly. This is due to the relative mean sea level rise and at the same time to
the strong human induced modifications that started in the 14th century and that are presently still
ongoing. In the year 2013 an extensive bathymetric survey was carried out in the Venice Lagoon.
During this survey all the channels of the lagoon were mapped with a Kongsberg EM 2040 DC
multibeam system with a grid resolution up to 5 cm. Due to the need to process this large dataset,
we developed a semi-automatic method based on a combined GIS-2DFTT multi-parameter analysis
and spectral parameterization of the bathymetric data. This analysis allowed us to identify and
parameterize the geometrical characteristics of the main morphological features of the channels,
like dunes, scours, crests and troughs and sedimentation areas and to extract the channel bottom
roughness.
Passive acoustic detections of odontocetes in the Ionian and Aegean Seas,
Greece.
Nikoletta Diogou, Holger Klinck, Julie Oswald, Evangelos Papathanassiou, Stratis
Georgakarakos, Jeffrey Nystuen
Presenter: Nikoletta Diogou
The Ionian and Aegean Seas, both part of the eastern Mediterranean Sea, are habitat for several
odontocete species including striped (Stenella coeruleoalba), short-beaked common (Delphinus
delphis), Risso's (Grampus griseus), and bottlenose (Tursiops truncatus) dolphins as well as Cuvier’s
beaked (Ziphius cavirostris), and sperm (Physeter macrocephalus) whales. Common dolphins and
sperm whales in the study area are listed as endangered in the IUCN Red List of Threatened Animals.
Furthermore, several standings of Cuvier’s beaked whales in Greece have been linked to naval sonar
exercises. Currently, very little is known about the seasonal abundance and distribution of cetacean
species in the eastern Mediterranean Sea. This information is crucial for the development of
effective protection measures for these animals.
In a first attempt to collect baseline data on the occurrence patterns of odontocetes in Greek
waters, two passive-acoustic recorders were deployed in 2008 in the Ionian Sea in the vicinity of
Ocean Station Pylos (36.8N, 21.6E) and in the northern Aegean Sea in the vicinity of Ocean Station
246
UA2014
Athos (40.0 N, 24.7E). These recorders operated for 19 and 10 months, respectively. Preliminary
results of the data analysis revealed that delphinid species were present at both locations yearround. As expected, sperm whales were predominately detected in the Ionian Sea. Current work
focuses on species identification of delphinid whistles using the Real-time Odontocete Call
Classification Algorithm (ROCCA) to allow an analysis of species-specific occurrence patterns.
Future work will include an examination of oceanographic data collected at each site
(http://poseidon.hcmr.gr) to identify environmental drivers of the occurrence of odontocetes in the
Ionian and Aegean Sea.
Work supported by the Hellenic Centre for Marine Research and the University of the Aegean,
Greece.
Sound Pressure Field Focused by Off-Axis Aplanatic Straubel Acoustic Mirror
Yuji Sato, Hanako Ogasawara, Koichi Mizutani, Toshiaki Nakamura
Presenter: Toshiaki Nakamura
Underwater acoustic lenses and acoustic mirrors made an effort on reducing sound attenuation,
aberration, and having stability for water temperature change. In the past study, we designed an
aplanatic Straubel (AS) mirror. The AS mirror is an aplanatic back-surface mirror. The AS mirror
could correct spherical and coma aberrations in result of experiment. However, the AS mirror has a
problem. The receiver array is located in front of the mirror, which results in the interruption of
incident sound waves, and then we designed an off-axis AS mirror and evaluated the convergence
properties using numerical calculations. The off-axis mirror is the effective area of the ordinary
mirror that is not hidden by the receiver array. The focal length and aperture of the original AS
mirror are both 400 mm. Therefore, the aperture of the off-axis AS mirror is 200 mm. In this report,
we made an off-axis AS mirror with silicone rubber and brass, and measured the convergence
properties at the frequency of 500 kHz for different incident angles in a water tank. We could not
measure at larger incidence than 10° due to the restriction of our experimental apparatus. At large
incident angles, -3 dB focal areas moved close to the mirrors. This deformation seems to be caused
by field curvature. The beam patterns along the spherical imaginary surface were measured at 0°,
5° and 10°. The peak value at 10° was about 2 dB lower than the value at 0°, but the beam widths
of the off-axis mirror were almost the same width at all incident angles. From these results it is
confirmed that spherical and coma aberrations are corrected. The side-lobe levels at oblique
incidence were below -20 dB and lower than the value at normal incidence.
247
UA2014
Location: Poster Area, Thursday 26th June 2014, 10:00 - 13:00
Preliminary investigation on the potential of using low power ultrasound to
induce low frequency vibrations on an immersed object.
Spyros Kouzoupis, Panagiotis Papadakis, George Piperakis
Presenter: Spyros Kouzoupis
In this paper, an investigation is undertaken as to whether low power ultrasonic projectors in water
can induce vibration on a soft object immersed in water. Experiments were conducted in a water
tank by using one transducer emitting an amplitude modulated (AM) pressure wave or two
converging transducers each one emitting a sinusoidal wave in the 500 KHz range, while their
frequencies differed by 500 to 2000 Hz. Short pulses with a rise time around 7 ns were also emitted.
The objective of these experiments was to check whether ultrasound can be used to induce
vibrations of fish swimbladder in the range of its resonant frequency (500-2000 Hz). In some
experiments focal transducers at 5 MHz were also used. In all cases the beams were aiming at a
light surface, made of plastic or nylon, where an accelerometer is mounted for picking up vibrations
in the frequency range mentioned above. Results for all different setups, using directional and focal
transducers are presented and discussed. Prospects for applying the method in order to cause a
sound response from fish or simply study their responsive behavior, are also considered.
248
UA2014
Parallel Activities-Meeting Point
Location: to be announced
Monday 23rd June 2014, 17:40-18:40
Institute of Acoustics, Underwater Acoustics Group, AGM meeting
Wednesday 25th June, 14:15-18:00
CTBTO’s Hydroacoustic Network
Expert Group Forum on Innovation pertinent to the medium- and long-term sustainment and
upgrading of the Hydroacoustic component of the International Monitoring System of the
Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO).
The purpose of this Expert Group Forum is to discuss innovative ways for the sustainment of
the Hydroacoustic component of the International Monitoring System (IMS) of CTBTO, and to
investigate the utilization of data and IMS assets by the scientific community for novel civil
applications and data quality verification. The expert discussion will be centred on two main
themes:
1. IMS network sustainability: modular design technology transfer from ocean
observatories
 Reliability of modular components in ocean applications
 ROV and modular design repairs
 Ocean observatories contributing to the IMS network
 Next generation observatories
2. Data processing and civil applications: can CTBTO’s hydroacoustic network be used
as an “ocean observatory” that provides data for novel civil applications?
 Novel civilian applications of IMS data
 Hydroacoustic event detection and evaluation via three different types of
sensors: Suspended IMS hydrophones, Ocean Bottom Seismometers (OBS) from
Ocean Observatories and IMS T-stations (Seismic stations)
 Modelling the propagation/reception of T-phase signals – potential of improving
T-station performance
Forum set-up: open discussion on the above themes with approximately 30 experts. One
introductory presentation will be given by the Forum organizers; no presentation is expected
by the participants (however, participants who wish to contribute material are welcome to
discuss it with the Forum organizers prior to the meeting). Follow up actions, feedback ideas
and recommendations will be incorporated into a report which will be distributed for
comments by all participants.
249
UA2014
This page intentionally left blank
250
UA2014
Conference Venue Map
251
UA2014
Exhibition
Teledyne RESON A/S
The Teledyne RESON name is the hall mark of class leading sonar equipment, transducers,
hydrophones and survey software that you can count on.
Headquartered in Denmark, Teledyne RESON has a global presence with offices and
representatives in the United Kingdom, Scotland, The Netherlands, Germany, Singapore, China and
the United States, Teledyne RESON provides worldwide support around the clock.
Teledyne RESON employs a team of roughly two hundred professionals dedicated to disciplines
including Installation Support, Research & Development, Program Management, Quality
Assurance, Manufacturing, Software Development, Security, and Administration.
Teledyne RESON is renowned for providing innovative
solutions to complex underwater surveying problems
in both commercial and military environments. Our
focus is to assure our clients receive the highest level
of operational support and system utility.
Beginning in 1976 as a supplier of ultra-sound sensors,
Teledyne RESON has become a world-leading
developer, manufacturer, and distributor of advanced
underwater acoustic sensors and sensor systems.
SeaBat 7125SV2
Teledyne RESON has more than 2000 SeaBat
Multibeam systems operational worldwide and
measured by the number of systems sold, Teledyne
RESON is the largest individual supplier of multibeam
systems today.
RESON manufactures over 20 different COTS SeaBat models, providing a solution to almost every
type of Multibeam sonar application. We are a true sonar systems Solutions provider, catering our
252
UA2014
products to the Hydrographic & Offshore Market, Civil Engineering & Dredging, Marine Research,
Military and Defense & Security.
The flagships in our product portfolio are the Seabat Multibeam Echosounders and our Data
Acquisition Software PDS2000.
SeaBat T20-P
Teledyne RESON is part of Teledyne Marine, a group of sixteen companies providing products and
services to the oceanographic community and owned by Teledyne Technologies Inc.
SeaBat 7160
www.teledyne-reson.com
253
UA2014
RTSYS
RTSYS is a R&D oriented company specialized in the design
and manufacturing of advanced underwater acoustics
systems (recorders, buoys, hydrophone networks) dedicated
to a wide range of shallow and deep water applications. Our
systems allow to measure different types of underwater
noise such as mammals noise, ships and boats noise, pile
driving and air guns or other anthropogenic noise.
What is it to know about the RTsys recording systems specifications:
Wide dynamic: 24bits .WAV files
Up to 4 synchronized inputs: beamforming, tracking..
Broadband: from 3Hz to over 1000KHz
Compatible with most existing passive and pre-amplified hydrophones sensors (Neptune
Sonars, RESON, B&K, HTI, Aguatech…)
Additional sensors: GPS, CTD, 3 axis accelerometer & gyrometer, multi-parameter probe…
Embedded processor for Real-Time processing
RTSYS offers versatile solutions that are useable from any computer via web embedded interface.
Moreover, RTSYS systems can be used as:
Autonomous dataloggers
Real-Time towed system (via Ethernet/VDSL/Internal network)
WiFi / Radio link connected systems
BASDA14: WiFi remote Sound & GPS buoy
BA-SDA14 WiFi buoy can record, stream and process sound data in real-time up to 1.5km distance
thanks to a WiFi antenna. It is a compact and flexible monitoring solution that allows multiple
applications such as mammals and fish monitoring as well as noise impact studies for piling,
offshore constructions.
254
UA2014
Four hydrophones, either passive or pre-amplified can be
simultaneously connected to the BA-SDA14 buoy with variable
cable length. GPS and other sensors such as CTD and
accelerometers & gyrometers can also be synchronized with
the acoustic data.
Easy to deploy, use and recover by one person, BA-SDA14
weighs only 34Kg in air and is equipped with rechargeable
batteries.
EASDA14: Versatile Brodaband Multichannel recorder
EA-SDA14 is a compact and versatile multi-hydrophone recorder that can be deployed either in
towed or autonomous mode and that offers excellent flexibility and user friendliness.
EASDA14 allows to record signals from up to 4 passive or pre-amplified broadband hydrophones
simultaneously at over 1000kHz in 24bits.
Acoustic data acquisition can be synchronized with many other sensors such as CTD, GPS, 3 axis
accelero & gyrometers, multi-parameter probe.
SYLence: Low-power Autonomous Sound recorder
SYLence is a very low-power and easy to use autonomous sound recorder. Only 55cm long and
with up to 2TB storage capacity, SYLence offers a compact and cost-effective solution for
anthropogenic and biological underwater noise recording.
SYLence allows to record signals from a single hydrophone at over 192kHz in 24bits.
www.rtsys.eu – +33 (0) 297 898 580 – info@rtsys.eu
255
UA2014
Ocean Sonics
11 Lornevale Road
Great Village, Nova Scotia
Canada B0M 1L0
Info@oceansonics.com
+1 902-655-3000
Ocean Sonics provides researchers, operators and acousticians with instruments that gather ocean
sound. Our goal is to produce the best digital acoustic instruments available.
The icListen Smart Hydrophone is the culmination of the team’s
decades of experience developing high quality hydrophones and
acoustic logging instruments.
The icListen is ideal for precise measurement of sounds such as
mammals, fish, crustaceans and other aquatic life, ships and boats,
wave and wind, seismic, pile driving and air guns or other
anthropogenic noise in the ocean.
Use the icListen as a digital hydrophone, acoustic data logger or
both.
The icListen Smart Hydrophone is a digital hydrophone that
Processes & stores acoustic data, and detects events
Transmits waveform or spectral data over cable or radio in real-time
Is low power & runs from an internal or external battery
Gathers waveform, spectral or event data in standard formats
Is calibrated with a wide dynamic range (24 bit, 120dB)
Has very low self-noise
Can be configured as a synchronized multi-channel system
256
UA2014
Ocean Sonics Products
icListen HF Ethernet Hydrophone 10Hz to 200kHz
IcListen AF Ethernet Hydrophone 1 Hz to 12kHz
IcListen LF Serial Hydrophone 1 Hz to 1600Hz
Radio Link WiMAX or GSM Radios
icTalk Smart Projector 10 to 200 kHz @ 140 dB re uPa@1m
Lucy PC software for instrument control & data visualization
Buoys, battery packs and cable solutions
Come and talk to us about your ocean listening and recording projects at the Ocean Sonics stand
at the 2nd Underwater Acoustic Conference in Greece.
Technical Paper Presentation by Ocean Sonics and Exocetus. “Automated Detection of Fishing
Vessels using Smart Hydrophones on an Underwater Coastal Glider.”
www.OceanSonics.com
257
UA2014
NORBIT
NORBIT GROUP was established in 1995 and is
renowned for providing innovative solutions in
many different markets. Our industrial
corporation is divided in four major divisions;
SubSea, Intelligent Traffic systems, Original Design Manufacturing and Electronic Manufacturing
Services, which consists of companies such as ASTI and Norbitech.
Our company has received the prestigious Gaselle distinction several times, awarded to the top 5
% Norwegian companies for long-term growth and sound financial status. NORBIT GROUP is
headquartered in Trondheim but also have offices in Poland, Australia and in the US. A network of
distributors is supplying our products to a global market.
NORBIT SUBSEA designs and develops wideband multibeam sonars for hydrographic application,
forward-looking applications as well as advanced subsea leakage detection. Our solutions are
based on the latest in analog and digital signal processing and our products provide wide coverage
monitoring combined with high sensitivity and accuracy.
NORBIT subsea provides solutions for applications such as:
Leakage detection:
NORBIT SUBSEA has developed a range of products. The NORBIT WBMS series are ultra compact
sonars optimized for long-range subsea leakage detection. On the example below the display
shows an example of NORBIT long-range automatic gas leakage detection, the red dots are the
build-in algorithms auto-detect functionality. 3D detection of a gas leakage rising to the surface is
shown on the top right illustration.
258
UA2014
Forward looking and obstacles detection: NORBIT’s Forward Looking Sonars (FLS) is designed to
detect obstacles in front of ships, such as seafloors- and in-water obstacles, tracking sea mammals,
monitoring pollution or underwater structures. The display below shows raw data from an ROV
survey, several targets are shown in the acoustics window and the video footage is shown as well.
Bathymetry and hydrography: Bathymetric sonar is used to measure elevation and depth of ocean
seafloor for hydrographic and inspection surveys and finding/identifying underwater objects. The
image below shows a dam survey with very complex bottom topography and close to 90m waterdepth (blue regions of the image).
www.norbit.no
259
UA2014
EvoLogics GmbH
EvoLogics GmbH
Ackerstrasse 76
13355 Berlin, Germany
Tel. +49 30 4679 862-0
Fax. +49 30 4679 862-01
www.evologics.de
sales@evologics.de
EvoLogics GmbH is a high-tech enterprise from Berlin, Germany. The company was founded
in 2000 by a group of leading international scientists and R&D experts aimed to develop
innovative key technologies for maritime and offshore industries through interdisciplinary
cooperation between engineering and life sciences.
EvoLogics GmbH designs and manufactures underwater information and communication
systems based on bionic concepts, combining cutting edge engineering with the best ideas
found in nature. The advanced product features have become enabling technologies for deep
water exploration and production.
EvoLogics’ products offer highly reliable, flexible and cost-effective solutions for multiple
underwater communication, positioning, navigation and monitoring applications. Our team
strives for innovation and invests its vast experience into developing, manufacturing and
supporting products that deliver an excellent performance and solve the most challenging
tasks.
260
UA2014
EvoLogics' developments are based on the patented S2C (Sweep Spread Carrier) technology the most reliable acoustic telemetry available for subsea and offshore applications that
provides an independent bidirectional data link along with acoustic positioning, broadcasting
and full networking capabilities. Our solutions can simultaneously facilitate optimal operation
and navigation of unmanned underwater vehicles - both remotely operated and autonomous.
Our systems enable retrieving information from various subsea sensors and other subsystems
and allow to control complex processes by combining communication with high-accuracy
positioning and multiple monitoring options.
EvoLogics caters to scientists, developers and commercial customers with a series of
underwater acoustic devices and software tools that offer an open framework for
development and testing, providing endless opportunities for new implementations.
EvoLogics S2C product lines have been carefully designed for operations in harsh underwater
environments and enhanced with special algorithms for signal processing and data
management. The company’s extensive experience with sensor integration allows it to provide
customers with turn-key solutions ranging from initial deployment up to recovering the
equipment.
Aware that unique application scenarios might require an individual approach and additional
customization, our experts are always ready to address special requests and find a solution,
tailored for the customer’s needs.
www.evologics.de
261
UA2014
MarSensing Lda.
MarSensing – Marine Sensing & Acoustic Technologies, Lda.
Centro Empresarial Pav. A5
Campus de Gambelas
8005-139 FARO
PORTUGAL
contact@marsensing.com
http://www.marsensing.com
MarSensing is a company consisting of scientists and engineers with activities in Underwater
Acoustics (UA). The company provides consultancy services in UA, develops instrumentation for
acoustic data acquisition, and participates in R&D projects. Our experience comes from over a
decade of research and technological development in UA.
SERVICES
In situ noise measurements & data analysis: wide range
of temporal and spatial configurations, real-time
monitoring.
Noise mapping & prediction: acoustic modelling for
noise level prediction over space and time.
Planning of noise monitoring programmes.
Custom-made technological solutions tailored to
specific problems.
PRODUCTS
The MarSensing digitalHyd line of products is designed to provide a user-friendly and compact
equipment solution in underwater acoustic signal acquisition activities. MarSensing has developed
a very compact autonomous acoustic recording device, the SR-1, and a telemetry based acoustic
recorder with real-time streaming of acquired data for remote visualisation, and internal data
processing capabilities, the TP-1, ideal for integration into existing systems, such as buoys, or for
real-time monitoring.
262
UA2014
TECHNOLOGY
MarSensing has many years of accumulated system development and design experience in
electronics, software and mechanics in the marine environment. This knowledge is applied in the
development of specialised devices with focus on underwater acoustics and integration of other
marine sensors into monitoring platforms:
Autonomous acoustic recording systems.
Multi-Channel Acoustic Systems.
Monitoring buoys.
Shore Connected Underwater Monitoring Stations.
REFERENCES
C. Soares, E. Cruz, et al., Environmental inversion with an autonomous hydrophone in a wave
energy device deployment site. In Proc. Underwater Acoustics 2014, Rhodes, Greece, June 2014.
A. Silva, A. Matos, C. Soares, et al., Measuring underwater noise with high endurance surface and
underwater autonomous vehicles, OCEANS 2013, San Diego, USA.
www.marsensing.com
263
UA2014
Wildlife Acoustics
Wildlife Acoustics, Inc. is launching its new, third
generation range of Song Meter bioacoustics
recorders. The Song Meter SM3M Deep Water,
SM3M Submersible and SM3 Aquatic are the
only recorders of their kind that double as noise
loggers.
The new SM3M and SM3 marine systems are the results of Wildlife Acoustics' grounds-up redesign
effort of its Song Meter technology platform. Incorporating lessons learned in the past ten years
making over 16,00 Song Meters and deploying them to over 60 countries, this new recorder line
represents new advances in bioacoustics technology.
This flexible, cost-effective and power-efficient platform allows the SM3M Deep Water, SM3M
Submersible and SM3 Aquatic to be quickly and easily configured to meet a vast array of
deployment missions.
Easy to Program: The SM3M and SM3 Aquatic, are some of the easiest-toprogram recorders on the market. Users can quickly program the recorders
aboard ship or in the field, with an easy-to-read, backlit LCD screen and
simple-to-use keypad.
Quick to Configure: All Song Meter recorders can instantly recognize which
type of microphone or hydrophone is attached and assure appropriate
settings, saving valuable set-up time.
Song Meter files can be created, viewed and edited with the free Song Meter
Configuration Utility. The utility allows users to conveniently program multiple
Song Meters using a single SD card. Simply save the program on a single SD
card and download it to other units before heading out on deployment.
Plan for Extended Deployments: The new Song Meter product range also
boasts of significantly increased deployment lengths – roughly 40% better
than its predecessors, the SM2+ and SM2M+. Using new power circuitry, the Song Meter SM3M
Deep water, for example, can record acoustic data continuously for up to 10 months, saving users
precious travel time and money.
Adapt to Multiple Roles: The SM3 Aquatic can not only record marine activity below the surface
of the water with its cabled 20 meter hydrophone, the recorder can simultaneously capture abovesurface avian and terrestrial acoustic or ultrasonic call data with an assortment of available Wildlife
Acoustics-supplied microphones.
264
UA2014
The SM3M and SM3 are designed to log SPL levels during
the recording process. The hydrophone sensitivity settings
are entered and the Song Meter can then automatically
convert recording levels to actual SPL levels. A running log
of maximum, minimum and mean values are stored for
easy post-analysis and reporting. Calibration data is
provided with all Song Meters to assure accuracy of the SPL
readings.
Quickly Analyze Data: For those needing to quickly
visualize sound data on a large scale Song Scope high
speed spectrum analysis software is a proven, mission
critical tool.
Using powerful and patented classification algorithms,
Song Scope enables users to scroll through large audio
files (up to 2GB) quickly and easily, saving valuable
analysis time. The analysis software allows researchers to
recognize visual patterns corresponding to specific vocalizations of interest. Song Scope's Sound
Scape feature enables researchers to visualize sound on temporal scales on the order of months
or years.
Convenient to Service: To make servicing easier and analysis work easier, the Song Meters have a
self-diagnostic feature and will automatically log meta-data.
Visit our table or contact Mona Doss at mona@wildlifeacoustics.com to learn how Wildlife
Acoustics range of Song Meter bioacoustics recorders can help you with your research work.
Wildlife Acoustics, Inc. | 3 Clock Tower Place | Suite 210 | Maynard, MA 01754 USA
www.wildlifeacoustics.com • +1.978.369.5225
265
UA2014
Sessions Timetable
266
UA2014
Sessions List
Session
1
2
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Title
Acoustic imaging
Acoustic Monitoring of Marine
Mammals
Acoustic Tomography in
Shallow Seas, Benthic and
Terrestrial Waters
Acoustics in Polar
Environments
Acoustics of Bubbles, Oil and
Gas
Acoustics of marine renewable
energy developments
Advances in Acoustic
Measurement Systems:
Technologies and Applications
Calibration of Sonar and
Hydrophones
Comprehensive Nuclear-TestBan Treaty Monitoring
Distributed Networked
Systems for Surveillance
Experimental and modelling
validation of target strength
measurements
Habitat Mapping: Procedures
and Results
Innovative Approaches for
Characterizing Ocean Bottom
Properties
Modeling Sonar Performance
in Uncertain Environments
Outer Continental Shelf,
Shelfbreak and Canyon
Acoustics
Radiated Noise from Ships and
Production Platforms
Sensitivity of underwater
acoustic observables
Sonar Performance Modeling
and Verification: Applications
to Active and Passive Sonar
Sonar Signal and Information
Processing
Organizer
Date
Lecture
Room
Jiyuan Liu
Purnima Ratilal and
Adam Zielinski
Jean-Pierre Hermand,
Arata Kaneko and
Hiroyuki Hachiya
Jaroslaw Tegowski
and Alexander
Gavrilov
Tim Leighton and Lee
Culver
Stephen Robinson,
Paul Lepper and
Philippe Blondel
Alessandra Tesei
Tuesday 24th
Thursday 26th
C
A
Monday 23rd
A
Thursday 26th
C
Monday 23rd
C
Wednesday
25th
D
Monday 23rd
C
Bo Lövgren and Stefan
Schael
Georgios Haralabus
and Mario Zampolli
Frank Ehlers and Arne
Schulz
Duncan Williams,
David Nunn and Alan
Hunter
Philippe Blondel and
Andrea Caiti
Martin Siderius, Sergio
Jesus, Peter Nielsen,
Jean-Pierre Hermand
and Ross Chapman
Georgios Haralabus
and Chris Strode
Jim Lynch
Wednesday
25th
Tuesday 24th
C
Thursday 26th
A
Wednesday
25th
C
Thursday 26th
B
Monday 23rd
A
Friday 27th
A
Monday 23rd
D
Grazyna Grelowska
Wednesday
25th
Wednesday
25th
Monday 23rd
B
Wednesday
25th
A
Emmanuel Skarsoulis
Michael Ainslie,
Charles Holland, Dale
Ellis and Kevin Heaney
Ryan Goldhahn,
Duncan Williams,
A
B
B
267
UA2014
21
Soundscapes and Measuring
Noise
22
Synthetic Aperture Sonar:
State-of-the-art
Tank Experiments
23
24
25
25
26
27
29
30
31
32
33
34
35
36
268
Three-dimensional sound
propagation models
Towards Automatic Target
Recognition: Detection,
Classification and Modeling (of
Underwater Targets)
Towards Automatic Target
Recognition: Detection,
Classification and Modeling (of
Underwater Targets)
Underwater Acoustic
Measurement Facilities and
Standards
Underwater Acoustic Studies
in Asian Seas
Underwater Communication
and Networking
Underwater Unexploded
Ordnance (UXO) Detection
and Remediation
Unmanned Vehicles (AUV,
USV and Gliders) for
Underwater Acoustic
Surveillance and Monitoring
Vector Sensors: Development
and Applications
Bioacoustics
Underwater Communication
Acoustic Modelling
Acoustic Inversions
Julian Deeks and Peter
Gerstoft
Jennifer Miksis-Olds,
Mark Prior and Kevin
Heaney
Roy-Edgar Hansen
and Daniel Brown
Jean-Pierre Sessarego
and Dominique
Fattaccioli
Michael Porter and
Frederic Sturm
John Fawcett,
Johannes Groen,
Wolfgang Jans and
Yan Pailhas
John Fawcett,
Johannes Groen,
Wolfgang Jans and
Yan Pailhas
Anthony Paolero and
Stephen Robinson
Tuesday 24th
B
Friday 27th
C
Thursday 26th
B
Tuesday 24th
D
Thursday 26th
D
Thursday 26th
D
Monday 23rd
D
Chifang Chen
Thursday 26th
C
Charalampos
Tsimenidis and Oliver
Hinton
Mike Richardson and
Wolfgang Jans
Tuesday 24th
C
Monday 23rd
B
Alain Maguer, Brian
Ferguson and Eric
Delory
Tuesday 24th
A
Tuncay Akal, Sergio
Jesus and Jean-Pierre
Hermand
Tuesday 24th
D
Thursday 26th
Thursday 26th
Tuesday 24th
Friday 27th
Friday 27th
B
D
B
D
B
UA2014
Social Program
The ice breaker party, Sunday 22nd (included in the registration fee)
Location: Conference Venue, Rodos Palace Hotel
Time: 19:00 to 22:00
The conference banquet, Tuesday24th (included in the registration fee)
Location: Conference Venue, Rodos Palace Hotel
Time: 20:00- 24:00
Excursion including lunch, Friday 27th (included in the registration fee)
Location: to be announced
Time: to be announced
Prices for extra tickets
Icebreaker party: 30€ per ticket
Conference banquet: 90€ per ticket
Excursion: 80€ per ticket
Lunch: 30€ per ticket
The accompanying persons can purchase tickets at the registration desk.
269
UA2014
Index of Session Organizers & Presenters
Name
Pages
Abawi Ahmad
Abildgaard Max
Ainslie Michael
Akal Tuncay
Aleksi Ivan
Alhaji Einas
Allotta Benedetto
An Xudong
Austin Melanie
Badiey Mohsen
Baggeroer Arthur
Bahl Rajendar
Balcazar Naysa
Balthasar Mark
Barazzutti Amelie
Barton Robert
Baumgartner Mark
Beamiss Graham
Bemis Karen
Bergès Benoît
Bielanski Jan
Binder Carolyn
Binnerts Bas
Blomberg Ann
Blondel Philippe
Bo Hu
Bohne Tobias
Bonnel Julien
Borejko Piotr
Borodina Elena
Brothers Richard
Brown Daniel
Burenin Alexandr
Buskenes Jo Inge
Buszman Krystian
Butler Michael
Caiti Andrea
Cauchy Pierre
Chalindar Bruno
Chapman Ross
20, 41, 115, 172
8, 203
12, 15, 26, 144, 160
28, 30, 215
35, 154
46, 176
55, 123
23, 63
19, 34, 92, 232
17, 20, 133, 173
31, 150
47, 66
51, 69
43, 110
59, 237
14, 186
25, 210
10, 183
42, 98
13, 84
40, 140
51, 70
38, 93
58, 166
34, 38, 42, 45, 52, 55, 81, 89, 91, 118
28, 216
16, 87
47, 67
24, 174
19, 231
50, 180
58, 165
39, 158
50, 181
33, 104
34, 89
43, 52, 55, 111, 118
21, 205
12, 146
7, 31, 125
270
UA2014
Chen Chi-Fang
Chen Yi
Chiu Linus
Chotiros Nicholas
Colin Mathieu
Cristol Xavier
Crocker Steven
Cruz Erica
Culver Lee
Dakin Tom
Dardis John
De Moustier Christian
Deeks Julian
Delory Eric
Diachok Orest
Diogou Nikoletta
Dobbins Peter
Dogan Hagan
Dol Henry
Dosso Stan
Dziak Robert
Ehlers Frank
Ellis Dale
Elmer Karl-Heinz
Espana Aubrey
Espinosa Victor
Evers Läslo
Fakiris Elias
Fang Yin-Ying
Fattaccioli Dominique
Fawcett John
Ferguson Brian
Ferguson Eric
Fezzani Ridha
Fillinger Laurent
Foglini Federica
Folegot Thomas
Freitag Lee
Gaggero Tomaso
Galindo-Romero Marta
Garrett Joanne
Gavrilov Alexander
Geilhufe Marc
Gemba Kay
Gerard Odile
Gerstoft Peter
53, 56, 189
14, 187
53, 190
15, 148
26, 160
12, 16, 32, 87, 143, 146
7, 10, 127, 184
42, 97
13, 16, 84
33, 104
38, 93
8, 200
31, 35, 39, 150
21, 25, 29, 204, 208
48, 221
61, 246
34, 89
13, 85
37, 113
32, 45, 80, 141
49, 83
39, 43, 109, 158
12, 15, 144
38, 95
41, 116
48, 221, 222
18, 107
52, 118
53, 191
44, 167
46, 50, 176
21, 25, 29, 31, 150, 204
47, 67
25, 208
35, 36, 41, 117, 137, 155
52, 121
26, 159
45, 78
9, 100
19, 230
14, 42, 97, 187
45, 47, 49, 65, 78
50, 180
31, 152
55, 72
7, 31, 35, 39, 126, 150, 151
271
UA2014
Godlewska Malgorzata
Goldhahn Ryan
Grelowska Grazyna
Grimmett Doug
Groen Johannes
Hachiya Hiroyuki
Hai Tran Minh
Hansen Roy-Edgar
Haralabus Georgios
Hazelwood Dick
Heaney Kevin
Hefner Brian
Hermand Jean-Pierre
Hinton Oliver
Hjelmervik Karl Thomas
Hodder Ben
Hodgkiss William
Holden Andrew
Holland Charles
Hope Gaute
Hoppe Lukasz
Hospital-Bravo Raúl
Hovem Jens
Hu Yanhong
Huang Chen-Fen
Huang Jianchun
Hunter Alan
Isakson Marcia
Islam Naveed
Ivansson Sven
Jans Wolfgang
Jeon Jong-Hoon
Jesus Sergio
Jiang Zelin
Jiang Yong-Min
Kaneko Arata
Katsnelson Boris
Kawanisi Kiyosi
Kebkal K.G.
Koelbel Jan
Koessler Matthew
Köhntopp Daniel
Kouzoupis Spyros
Kowalska-Duda Ewa
Kruss Alessandra
Kuperman William
272
52, 120, 121
31, 35, 39, 43, 109, 150
36, 40, 135
39, 157
46, 50, 58, 165, 176
11, 15, 73
30, 198
58, 165
18, 57, 106, 129
38, 95
12, 15, 18, 20, 26, 29, 108, 144, 159, 173
12, 58, 144, 239
7, 11, 15, 28, 30, 73, 125, 215
27, 30, 194
46, 178
37, 113
22, 234
57, 129
12, 15, 144
45, 79
45, 80
36, 137
48, 222
33, 104
11, 74
60, 240
26, 37, 41, 50, 112, 116, 161, 181
22, 235
25, 209
20, 171
8, 46, 50, 176
40, 139
7, 28, 30, 125, 215
23, 62
57, 131
11, 15, 73, 76
19, 22, 231, 233
15, 76
25, 27, 194, 210
8, 202
17, 132
50, 182
61, 248
52, 119
61, 246
37, 113
UA2014
Lan Hualin
Lasmar N.-E.
Le Gall Yann
Leblanc Tara
Leighton Tim
Leissing Thomas
Lepage Kevin
Lepper Paul
Leroy F.
Li Bing
Li Fenghua
Li Min
Li Qihu
Li Yuan
Lianjin Hong
Lin Ju
Lin Ying-Tsong
Lippert Tristan
Liu Jiyuan
Liu Songzuo
Lövgren Bo
Lynch Jim
Madricardo Fantina
Maguer Alain
Mahr Ray
Majewski Piotr
Martins Ricardo
Mashoshin Andrei
Matsumoto Haru
Matveev Anton
Mei Jidan
Michaelis Martin
Miksis-Olds Jennifer
Miyamoto Yoshinori
Mori Kazuyoshi
Moura André
Muniraj Karthikeyan
Murphy Stefan
Muzi Lanfranco
Nadig Clare
Nait-Chabane Ahmed
Nakamura Toshiaki
Nakcha Mohamad
Newhall Bruce
Nichols Stephen
Nicolas Barbara
28, 215
35, 153
15, 149
8, 201
13, 16, 84
61, 244
29, 43, 57, 109, 131, 214
34, 38, 42, 89, 90
25, 211
60, 241, 242
30, 219
33, 104
60, 242
23, 60, 62, 242
28, 217
11, 74
20, 171
19, 232
23, 62
54, 224
33, 103
17, 132
55, 124
21, 25, 29, 204
21, 206
13, 85
29, 213
46, 176
18, 107
10, 184
54, 226
43, 110
26, 29, 159
56, 228
23, 64
36, 137
48, 222
35, 154
7, 125
19, 230
15, 148
61, 247
46, 176
35, 153
18, 106
32, 141
273
UA2014
Nielsen Peter
Nijhof Marten
Nunn David
Ogasawara Hanako
Ouaskit Said
Page Keith
Pailhas Yan
Paolero Anthony
Papadakis Panagiotis
Paris J.F
Passerieux Jean-Michel
Peng Bo
Peng Chiao-Ming
Peng Li
Pereira Shan Victor
Petrov Pavel
Pierce Allan
Porpilho Diego
Porter Michael
Premus Vince
Prior Mark
Prowse David
Puig Vicent
Qiao Gang
Quidu Isabelle
Racca Roberto
Ratilal Purnima
Real Gaultier
Reimann Katja
Ren Qunyan
Richardson Mike
Robinson Stephen
Rose David
Santos Ana Bela
Sasakura Toyoki
Schael Stefan
Schmidt Henrik
Schulz Arne
Serebryany Andrey
Sertlek H.Özkan
Sessarego Jean-Pierre
Shi Jie
Shimura Takuya
Sidelnikov Gleb
Siderius Martin
Skarsoulis Emmanuel
274
7, 125
41, 115, 117
37, 41, 112
61, 245
22, 233
37, 113
46, 50, 176, 177, 179, 182
10, 14, 183
44, 167
27, 197
27, 196
54, 225
12, 146
23, 63
44, 168
20, 172
7, 126
52, 119
20, 24, 171, 175
57, 130
26, 29, 159, 161
37, 112
48, 221
56, 228
46, 178
47, 65
47, 51, 55, 65, 71
44, 57, 129, 168
38, 94
30, 219
8, 200
10, 14, 34, 38, 42, 89, 183, 187
8, 200
59, 236
9, 101
33, 103
29, 213
43, 109
17, 56, 132, 193
12, 29, 145, 163
44, 167, 169
28, 217
27, 195
31, 151
7, 125
32, 141, 142
UA2014
Smith Kevin
Soares Cristiano
Spence Jesse
Steininger Gavin
Strode Chris
Sturm Frédéric
Sun Dajun
Sun Zongxin
Szymczak W.
Taroudakis Michael
Taylor Charles
Tegowski Jaroslaw
Tesei Alessandra
Tingting Teng
Tollefsen Dag
Traverso Federico
Trofimov M.Yu.
Tsimenidis Charalampos
Tsuchiya Takenobu
Viola Salvatore
Wang L.S
Wang Peng
Wang Shiquan
Wang Yuebing
Warner Graham
Wiedmann Karsten
Willcox S.
Williams Duncan
Winkelmann Kay
Wochner Mark
Woock Philipp
Wood Mark
Wood Michael
Wu J.R.
Wu Jeff Chih-Hao
Xenaki Angeliki
Xiao Dong
Xing Siyu
Yack Tina
Yan Sheng
Yang D.S.
Yang Xiuting
Yang Yiing-Jang
Yi Dong Hoon
Zampolli Mario
Zeng Juan
17, 133
34, 91
36, 136
7, 125
57, 129, 131
20, 24, 171, 175
59, 236
60, 240
36, 136
32, 142
37, 113
45, 49, 78
9, 21, 99, 205
35, 155
40, 49, 82, 139
36, 49, 82, 135
24, 174
27, 30, 194, 199
9, 101
9, 99
29, 163
23, 63
14, 186
14, 27, 186, 196
7, 127
54, 224
25, 211
31, 35, 37, 39, 41, 112, 150, 201
8, 201
13, 86
59, 237
21, 206
38, 94
22, 56, 193, 234
53, 192
31, 151
54, 225
30, 198
51, 69
60, 243
28, 215
33, 104
53, 189
51, 70
18, 106
53, 189
275
UA2014
Zhang Chun
Zhang Chuanzheng
Zhang Jiarong
Zhang Lanyue
Zhang Zhibo
Zheng Haibo
Zhou Feng
Zhu Xiao-Hu
Zhu Z.R.
Zielinski Adam
276
55, 72
15, 77
56, 228
61, 244
60, 242
60, 242
54, 226
11, 75
28, 216
27, 47, 51, 55, 65, 68, 194
UA2014
277
UA2014
278

UA2014 - Underwater Acoustics Conference

Transcription

Similar documents

art snorkelling

3_Adventure Playground.indd

SES-300-XXX-ATEX eng

Invitation to seminar and workshop

Paper ICA2016-843

NEW SILENT KoENIgSEgg

the UComms`16 book of abstracts here.

Pantallas Acusticas Pared Verde (EN)

features

Gregariousness and school behaviour of pelagic fish