CrIS Sensor: Generation of RDRs and SDRs

Transcription

CrIS Sensor:
Generation of RDRs and SDRs
R. Glumb & J. Predina
ITT Industries, ITT Aerospace/Communications,
1919 West Cook Road, P.O. Box 3700, Fort Wayne, IN 46801, USA
Contact: ron.glumb@itt.com
Contact: joe.predina@itt.com
Overall Relationship of RDRs, SDRs and EDRs
Sensor Design
• 833 km orbit
• 98.7° Inclination
• CrIS
• ATMS
1.
25
-O
r
RD
Rs
RDR = Raw Data Record ( Uncalibrated )
SDR = Sensor Data Record (Calibrated)
EDR = Environmental Data Record
bi
t
Da
ta
D
um
p
Context of Discussion
Central or
Regional Ground
Stations
2,200 km
Swath
Decode
Spacecraft
Data
RDRs
to
Users
Raw
Uncalibrated
Data
Sensor
Calibration
Algorithms
SDRs
to
Users
Calibrated and
Geolocated
Radiance Data
EDR
Algorithms
EDRs
to
Users
• Temperature Profiles
• Moisture Profiles
• Pressure Profiles
20 minutes
CrIS
• ITT INDUSTRIES •
AER • BOMEM • BALL • DRS
3-2
Sensor Design
Reqmnt ID
Requirement
SDRP907
Spectral Bands
SDRP944
Unapodized Spectral
Resolution
1/(2*max OPD)
SDRP3546
Spectral Uncertainty
SDRP6841
ILS Shape Uncertainty
SDRP3481
Radiometric Uncertainty
SDRP1033
System NEdN
SDRP3696
SDRP3696
FOV
FOV
-
Size (km)
70% width (in/x track)
Shape Match (km)
70% width
50% width
10% width
1% width
SDRP3628
SDRP3630
Scan Extent
SDRP882
SDRP3731
Mapping Uncertainty
LOS Jitter
CrIS
CrIS Science Mission Cardinal Requirements
LWIR
MWIR
650 – 1095 cm
0.625 cm
-1
SWIR
1210 – 1750 cm
-1
1.25 cm
-1
2155 – 2550 cm
-1
2.5 cm
-1
-1
10 ppm (flight unit 1)
5 ppm (subsequent flight units)
0.5% FW HM
0.5% FW HM
0.5% FW HM
0.45%
0.58%
0.77%
See Chart
See Chart
See Chart
Same as LW IR
Same as LW IR
0.3
0.2
0.3
N/A
0.3
0.2
0.3
N/A
11.8/12.7
13.2/14.2
14.9/16.0
16.8/18.0
0.3
0.2
0.3
N/A
30 x-track FORs: +/- 48.333
1.5 km
50 µrad/axis
o
30 x-track FORs: +/- 48.333
1.5 km
50 µrad/axis
3-3
o
30 x-track FORs: +/- 48.333
1.5 km
50 µrad/axis
o
Sensor Design
CrIS Science Mission Cardinal Requirements: NEdN
NEdN (mW/cm²/sr/cm-1)
1
Tscene = 287K
Tscene = 233K
0.1
0.01
0.001
600
1100
1600
Wavenumber (cm-1)
CrIS
3-4
2100
2600
Signal Flow
Earth
Radiance
• CrIS SDR System Is Comprised of
CrIS
Optical
Hardware
CrIS
Signal Processing
Hardware
(RDRs)
Ground
Calibration
Software
CrIS
CrIS System Functional Partitions
for Generation of SDRs
– Optical Processing Hardware
• Converts Scene Radiance to Photons at Detector Surface
• 9 separate FOVs with three colors in each
– Electronic Signal Processing Hardware
• Converts Photons at 27 Detector Surfaces into uncalibrated
sampled data streams out of instrument (RDRs)
– Ground Calibration Software
• Converts Raw Data Records (RDRs) to Calibrated Sensor
Data Records (SDRs)
– Each FOV Geo-located
– 1305 Spectral Channels (colors) per FOV
– Radiometrically calibrated
– Spectrally calibrated
Calibrated
Data
(SDRs)
3-5
Optical Signal Flow: Earth Scene Interrogated with
3x3 FOV Array
Signal Flow
CrIS Field of Regard (FOR)
Definition
• 1 FOR = 9 FOVs
• 1 FOV = 3 IR bands
Single CrIS Scan Line
(full sweep, 30 FORs)
3.33 degree steps
Nadir
48.33 degree
Edge
0.963 degree
(14.0 km)
Anti-sun
S/C
Velocity
3
1.100 degree
1100 km
(16.0 km)
2
1
6
5
4
9
8
7
1.024 degree
(14.9 km)
0.897 degree
(13.0 km)
Three Successive
CrIS Scan Lines
(nadir to edge sweep)
CrIS
3-6
Signal Flow
Instrument Converts Scene Photons to Packetized
Bits (RDRs)
Optical Signal
Flow: Entrance
Pupil to Exit
Pupil
Electrical Signal
Flow: Exit Pupil
to Packetized
Bits
CrIS
3-7
Optical Signal
Flow
Optical Signal Flow: Entrance Pupil to Exit Pupil
• Partially unfolded CrIS optical system shows flow of
signal radiance to detectors.
Interferometer
Cooler
Telescope
Detector
Optics
Scene Radiance
SSM
CrIS
3-8
Optical Signal
Flow
Detector Active
Area Plane
Located at Exit
Pupil
Electrical Signal Flow: Pupil Image Photons to
Packetized Bits
Detector
P/A
Exit Pupil Image
RDR Consists of:
- Engineering
Data
- Science TLM
Data
- Signal ID/QC
Data
ADC
Digital Signal
Processing
RDR
Interferograms
Calibration Data
Downlinked to Earth
CrIS
3-9
Optical Signal
Flow
Interferogram Generated by Interferometer With
+/- 0.8 cm OPD Sweep
Two Sided
Double-sided
Interferograms
Baselined for
CrIS
LW
MW
SW
–0.8
–0.4
–0.2
0.0
0.2
0.4
OPD [cm]
Double-Sided Interferogram Benefits:
• Better phase calibration of instrument
(and consequent treatment of channelization effects)
• Less sensitive to sweep asymmetries of hardware (vs
Single-sided interferograms)
CrIS
3 - 10
0.8
Electrical
Signal Flow
Filters Optimized
for Low Gain and
Phase Distortion
Dedicated A/D
on Each Channel
Minimizes Noise
CrIS Electronic Signal Processing: Key Features
(1 of 2)
• 27 Channel Interferogram Signal Processing
– Anti-alias analog filter
• 5 pole low pass (51.7 kHz)
• 2 pole high pass (300 Hz)
– 14-bit ADC
• 13.4 effective bits on each detector channel @ 128 ksps
– Oversampled to prevent cut-off effects of anti-aliasing filter (gain/phase slope)
from encroaching IR signal passband
• Metrology delay matched sampling, commandable
Digital
Processing
Techniques
Significantly
Reduce Data
Transmission
Rate
Extensive
On-orbit
Programmable
Flexibility
CrIS
– Programmable electronic gain
• Optimizes each CrIS IR channel dynamic range
• 40 dB range, 0.156 dB increments, commandable
– Impulse noise suppression
• Reduces impulse noise energy over 60 times, programmable
– 255 tap decimating digital FIR filter (fully programmable)
• Decimation reduces bandwidth & data rate
• 69 dB stopband, +/-2.5 dB In-band ripple, sharp transition
• 9 channel ASIC implementation for low power & speed
– Bit trim encoding to reduce data rate
• Removes unneeded leading zeros of interferogram data samples
• Implemented by Flight computer, commandable reconfigure
3 - 11
Electrical
Signal Flow
CrIS Formats All
Data into CCSDS
Packets
CrIS Electronic Signal Processing: Key Features
(2 of 2)
• Data transmitted in Packets with Unique IDs
• Science Data Channel (packets per 8 second scan)
–
–
–
–
–
Yields
1.44 Mbps
Data Rate
Packets Tagged
with Unique
APIDs to Speed
Sorting and
Ground
Processing
• Other Data Packets for C&DH, Diagnostics
Up to 128
Unique APID
Packet
Assignments
CrIS
Earth scene packets (810 = 27 detectors x 30 scenes)
ICT calibration packets (54 = 27 detectors x 2 looks)
DS calibration packets (54 = 27 detectors x 2 looks)
Science telemetry packet (1)
Engineering packet (once every 30 scans, 4 minutes)
–
–
–
–
–
–
–
–
Time of Day (TOD) & navigation packet (once/sec)
Housekeeping telemetry (2 kbps)
LEO&A telemetry (0.256 kbps)
Command packets
Test packets
Memory load/memory dump packets
Diagnostic interferogram data packets
Telemetry dwell packets (diagnostic telemetry)
3 - 12
Electrical
Signal Flow
CrIS On-board Signal Processing Builds RDRs
13.4
A/D
PV
HgCdTe
21 038
samples
24
18
Decimate
24
18
Trim....54%
Bit Bit
Trim…63.1%
5-Pole
Butterworth
LPF
51.7 kHz
866
samples
Trim....63%
BitBitTrim…73.4%
Decimate
eff
530
samples
Trim....60%
BitBit
Trim…61.7%
1% Sweep
Velocity Error
FOV #1
255 Tap FIR
BPF
650 - 1095 cm -1
Longwave
Impulse Noise
Suppression
10 cm/sec OPD
Sweep Velocity
202
samples
PV
HgCdTe
2155 -2550 cm -1
Shortwave
IR to Metrology
Delay Matching
PV
HgCdTe
10854
samples
Decimate
20
12.1 - 17.5 kHz
5-Pole
Butterworth
LPF
51.7 kHz
A/D
13.4
5 506
samples
Impulse Noise
Suppressed
Decimate
26
15
Decimate
26
15
21.55 - 25.5 kHz
1550
Laser
Low Data Rate
CrIS
17
17
255 Tap FIR
BPF
Laser Metrology
Driven Sampling
13.4
A/D
Decimate
20
255 Tap FIR
BPF
5-Pole
Butterworth
LPF
51.7 kHz
Impulse Noise
Suppression
1210 - 1750 cm
Midwave
Impulse Noise
Suppression
160 msec
Integration Time
160 kHz
LPF
Delay
Equalize
Sample Pulses
128,906 Hz +/-1% Metrology Reference
3 - 13
2 kbps
TLM
CCSDS Data Packet &
FOV#1
through
#9, TLM,
F
i
li i
d
6.5 - 10.95 kHz
-1
1.5 Mbps
Electrical
Signal Flow
Impulse Noise Detection/Suppression is
Important in a Space Environment
• CrIS Detectors Can Be Subject to Impulse Noise
– Sensor charging/arc discharge
– High energy particle (space radiation environment) bombardment of detector
– Impulse will span two undecimated
interferogram samples
I(x)
– Frequency of occurrence expected
to be very low
Impulse Noise
Interferogram
x
real IGM
• Impulse Noise Clipping Reduces Noise Susceptibility
–
–
–
–
CrIS
Factor of 60 or more suppression improvement
CrIS uses bit trim mask to detect impulse noise prior to digital FIR filter/decimation
Substitute a zero value in place of the measured impulse noise value.
Number of impulse noise hits is counted and reported for each interferogram to
aid in data quality assessment.
3 - 14
255 Tap Digital FIR Filter & Optical Filter Overlay
Long Wave Band
Electrical
Signal Flow
Cascade of Two
Optical Filters
Plus Sharp
Cutoff Digital FIR
Filter Provide
High Out-of-Band
Rejection
LW Band:
650 cm-1 1095 cm-1
1
10
0
10
FIR
FIRFilter
Filter
Passband
PassbandRipple
Ripple
-1
Excellent
Robustness to
EMI Aliasing
Enables Use of
Large
Decimation
Factor
In-band Ripple
Removed by
Calibration
CrIS
Gain
10
Cascaded
Cascaded
Optical
OpticalFilter
Filter
Response
Response
-2
10
FIR
FIRFilter
Filter
Stopband
StopbandRipple
Ripple
-3
10
-4
10
-5
10
0
500
1000
1500
2000
2500
Wavenumber (cm-1)
3 - 15
3000
3500
4000
Electrical
Signal Flow
Filtering and Decimation
Order of
Magnitude
Reduction of
Samples via
Signal
Processing
CrIS Interferogram Measurements
Double Sided IGMs
LW
MW
SW
–0.8
Overscan
Samples Taken
to Fill Digital FIR
Filter Pipeline
127
0.0
0.2
0.4
0.8
LW Band
MW Band
SW Band
± 0.8035 cm
OPD
± 0.4092 cm
OPD
± 0.2015 cm
OPD
26
26
20,736
real
1
CrIS
–0.2
OPD [cm]
24
One Decimated
Overscan
Sample for ZPD
Uncertainty
–0.4
24
20
127
127
10,560
real
FIR Filter
&
Decimate
24
FIR Filter
&
Decimate
20
complex
complex
1
864
3 - 16
20
1
528
127
5,200
127
127
real
FIR Filter
&
Decimate
26
complex
1
1
200
1
Bit Trimming Performed By CrIS Flight Computer
Electrical
Signal Flow
Bits Above Mask
Are Discarded
No Loss of
Information
On-orbit
Programmable
Mask
Interferogram Envelope
407
Bits
Bitsabove
abovemask
mask
are
discarded
are discarded
460
15
[Output Bits/Sample - 1] (complex)
Bit Trimming
Allows CrIS to
Meet Bandwidth
Requirements
(LW example for a pre-trimmed 17 bit word width)
353
10
31
515
299
138
568
729
836
1
866
5
0
-5
-10
Data After Trim:
-15
63.1% in the LW
-0.8
-0.6
-0.4
-0.2
0
0.2
73.4% in the MW
Optical Path Difference (cm)
61.7% in the SW
866 decimated complex samples
CrIS
3 - 17
0.4
0.6
0.8
Electrical
Signal Flow
Space Segment Processing: Functional Flow
SPACE SEGMENT - CrIS Sensor
Detector
DS
ADC
I (x )
Interferogram
Interferometer
(photons)
255 Tap
FIR
I (x )
L(σ )
ES:
Earth Scene
DS:
Deep Space
ICT:
Internal Calibration Target
Filtering &
Decimation
Reduces
data rate
~
I ( x ) complex
Engineering Data
(three different types)
Must be
done first
real
ICT
Observed scene
Impulse Noise
Clipping
Once per 4 minutes
éBit trim format data
élaser frequency info: NNe, NL
éICT emissivity tables & model
éPolarization correction tbl (if
needed)
éILS parameters for 54 channels
éCrIS mounting angles
éLOS angles for each FOV
Signal ID/QC Data
éIGM start time (UTC)
éFOR index (0-31)
éBand ID (LW,MW,SW)
éFOV ID (1-9)
éSweep direction
éZPD min, max & position
éQuality control flags
éImpulse noise count
Bit
Trimming
Downlinked
to Earth
RDR
Packet
Encoding
Interferograms
Calibration
Data
Once per 8 seconds
éLaser diode temperature/current
éICT, BS, Scan mirror
temperatures
éAll optic temperature telemetry
éSSM pointing errors
Science TLM Data
Only Processing of Science Data Shown
CrIS
3 - 18
Electrical
Signal Flow
RDR Content
(1/3)
• Interferogram Data Packets (27 packets/FOR)
Comprises Bulk
of Data
Auxiliary Data
Contained in
Each
Interferogram
Data Packet
Used for
Identification
Purposes
Contaminated
Data Detected by
CrIS Sensor &
Flagged
CrIS
– Interferograms
– Interferogram Identifiers
• Spacecraft ID tag
• CrIS Sensor ID or serial number
• FOR index (1 - 30 = Earth scenes, 0 = ICT, 31 = DS)
• FOV number (1-9)
• Band designator (LW, MW, SW)
• Interferometer sweep direction (forward, reverse)
• UTC stamp (Instant when FOV footprint frozen)
• ZPD magnitude and fringe count
– Data Quality Indicators
• Fringe count error and fail bit trim flags
• Impulse noise count (0-127)
• Invalid interferogram data flag (Saturated channel, failed
detector)
3 - 19
Electrical
Signal Flow
Engineering
Data Embedded
in RDRs
Eliminates Need
for Sensor
Unique
Calibration
Handbooks
Allows Remote
Terminals to
Seamlessly
Synchronize
with Any CrIS
Sensor Downlink
Anytime
CrIS
RDR Content
(2/3)
• Engineering Data Packet (once per 4 minutes)
– Metrology Wavelength Data
• Neon fringe count (from last calibration)
– ICT Calibration Data
• Emissivity versus wavenumber
• ICT radiometric model parameters
– ILS Model Parameters for Each CrIS Detector
• FOV to LOS offset angles
• FOV size (angular)
– Polarization Correction Data vs. Scan Angle & Wavenumber
– Mapping Data
• CrIS to S/C alignment cube
• CrIS LOS to CrIS cube reference angles
• CrIS scanner to interferometer alignment data
– Coefficients to convert data to engineering units
– Bit trim parameters & other format decoding data
3 - 20
Electrical
Signal Flow
RDR Content
(3/3)
• Science Telemetry Data Packet (once per 8 seconds)
Science
Telemetry
Packets Only
Contain
Dynamic Data
Supporting
Science Mission
Calibration and
Geolocation
More Complete
and General
Telemetry/House
keeping Data
Sent to
Spacecraft
Operations
Control Center
Over Different
Channel
CrIS
– Metrology Wavelength Data
• Laser diode case temperature, current and model parameters
– Temperatures
• Beamsplitter, Scan mirror, Scan Baffle, Telescope, Aft optics,
Detector
– SSM servo pointing errors
• From each of 30 previous earth scenes
• In-track
• Cross-track
• Normal Telemetry Data Packets (once per second)
– Rotation of 8 fixed format packets
• 2 kbps maximum data rate
• Recent FT1394 requirement
– Contents
• Temperatures, secondary voltages, PCE status, command
status, heater currents, IFM status, SSM status
3 - 21
SDR System Design Description:
Ground Software Element
SDR Algorithm
Ground Calibration Software Is
Partitioned into 9 Modular Groups
• Initialization
• Radiometric Calibration
– Initialize software
– Initialize RDR reading pointers
–
–
–
–
–
–
–
–
• Data Input
– Low level data handling
– Configuration data handling
– Calibration data handling
• Ingest sensor unique cal data
• Monitor calibration data
• Compute spectrum correction matrix
– Science data handling
• Geolocation
– Map FOV to latitude & longitude
– Calculate view angles/ footprint geometry
• Preprocessing
– Perform bit trim decoding
– Convert interferograms to spectra
Average warm target spectra
Average cold target spectra
Subtract sensor background radiance
Calibrate sensor gain
Remove phase dispersion
Compute ICT radiance
Isolate/reject orthogonal noise
Apply spectrum correction matrix
• Remove ILS errors
• Apply user selectable apodization
• Map channels to fixed wavenumber grid
• Quality Control
–
–
–
–
Identify/exclude bad data
Detect/correct fringe count error
Estimate NEdN (bin by bin)
Flag bad FOVs
• Post-processing
• Spectral Calibration
– Perform spectral calibration
– Compute laser WL from neon lamp
– Compute laser WL from diode parameters
– Select user required spectral bins
– Format data for EDRs
– Archive data
• Data Output
CrIS
3 - 23
Signal Flow Through Ground Calibration Algorithm
SDR Algorithm
~
I ( x)
ES (88%)
~
S es [n]
Scene FCE
Handling
Radiometric
Calibration
(complex)
Compute
Spectrum
L(σ )
Process Calibration References
N ma = 30
~
S (σ )
Health
Monitoring
~
L (σ )
L [r.u.]
Metrology
Monitoring
DS (6%)
~
S ds [n]
Calibration
FCE
Handling
Moving
Average
N ma = 30
ICT (6%)
~
S ict [n]
Calibration
FCE
Handling
Moving
Average
Spectrum Correction
mean cold
reference
~
S ds
Spectral
Resampling
ILS
Correction
mean hot
reference
~
S ict
User Defined
Apodization
L [r.u.]
T ict
σ0
CrIS
3 - 24
σ1
σ
L(σ )
NEdN
CrIS Interferograms to Calibrated Spectra:
Signal Processing Progression
SDR Algorithm
Interferogram
Samples from
CrIS A/D
24
127
LW Band
MW Band
SW Band
± 0.8035 cm
OPD
± 0.4092 cm
OPD
± 0.2015 cm
OPD
26
26
20,736
real
24
20
127
127
10,560
real
FIR Filter
&
Decimate
24
Interferogram
Samples in
RDRs
complex
1
1
1
864
SDR
Algorithm
Spectral
Samples After
SDR Algorithm
Calibration
Spectral
Channels
Retained
76
20
127
5,200
127
real
FIR Filter
&
Decimate
20
FIR Filter
&
Decimate
26
complex
complex
528
1
1
SDR
Algorithm
75
48
47
20
21
433 real
159
Discard
Guard
Band
Discard
Guard
Band
Discard
Guard
Band
433 real
3 - 25
1
SDR
Algorithm
1305 Total CrIS Spectral Channels
CrIS
200
713 real
713 real
127
159
SDR Algorithm
Calibrated
Radiance in 1305
Channels
Noise Estimates
in Each Channel
SDR Content At SDR Algorithm Output
(1 of 2)
• Calibrated Data
– Real part of the spectra after Spectral Correction
• LW 713 bins @ 0.625cm-1 (650-1095)
• MW 433 bins @ 1.25cm-1 (1210-1750)
• SW 159 bins @ 2.5cm-1 (2155-2550)
– Imaginary part of the spectra before Spectral Correction
(LW 713 bins, MW 433 bins, SW 159 bins)
– NEdN estimates (LW 713 bins, MW 433 bins, SW 159 bins)
• Geolocation Data
Mapping Data
CrIS
– Latitude/longitude @ sea level for each FOV center
– Major and minor elliptical footprint size for each FOV
– Elevation & azimuth angle from each FOV center to satellite
3 - 26
SDR Content At SDR Algorithm Output
SDR Algorithm
(2 of 2)
• Identifiers
Identifiers Help
in the Archiving
of Data
– Spacecraft ID, CrIS sensor ID, Sensor flight software version
number, SDR algorithm version number, Apodization tag
– FOR number, FOV number
– Band designator (LW,MW,SW), FOV longitude and latitude,
Slant angle, Viewing angle, Size of FOV on ground
• Quality Control
Quality of Data
Is Assessed and
Tagged
CrIS
–
–
–
–
ZPD reset, Fail bit trim, Impulse noise count (0-127)
Invalid data (RDR and SDR) and invalid geolocation flags
FCE detected and corrected in SDR algorithm
Excess NEdN, Excess Sensor Thermal drift
3 - 27
Web Site
Further Information
• This presentation of CrIS data record generation and
processing is a summary of more detailed information
available at:
http://npoesslib.ipo.noaa.gov
CrIS
3 - 28

CrIS Sensor: Generation of RDRs and SDRs

Transcription

Similar documents

ABS Consulting - SANS Institute

CRIS Radio Air Waves Spring/Summer 2016 newsletter

CRIS AIR WAVES

The Infrared Ozone Retrieval Algorithm for NPOESS-OMPS

S460 Louis Vuitton belt

KELLOGG SWITCHBOARD AND SUPPLY COMPANY

Professional Babbitting Services

Home Theater Color Calibration Spyder3TV™ Offers the

g-nassau-os - 13