Your SlideShare is downloading. ×
0
Onboard Radar Processing Concepts for the DESDynI Mission Yunling Lou, Steve Chien, Duane Clark, Joshua Doubleday, Ron Mue...
Agenda <ul><li>DESDynI mission radar measurements overview </li></ul><ul><li>Benefits of onboard processing technology </l...
DESDynI Mission Radar Measurements Overview July 2010 Biomass Canopy Height 8 Tbit onboard storage 1 Gbps TDRSS link <ul><...
Potential Benefits of Onboard Processing Technology for DESDynI Mission July 2010 <ul><li>Interferograms and Single-look C...
Other Potential Advantages of Onboard Processing for Earth Science Radar Missions July 2010
Proposed Onboard Processing Scenario for DESDynI Mission’s Radar Instrument July 2010 <ul><li>Downlink data volume reducti...
Example Mission Scenario with Autonomous Sciencecraft Experiment Image taken by Spacecraft Event Detection No event Detect...
High Level Architecture of the  Onboard Processor <ul><li>Repetitive, computational-intensive operations are done in the F...
Approach to Onboard Product Development <ul><li>Determine the types of data processing suitable for implementation onboard...
Onboard Product Examples: Forest Fire Extent July 2010 LHH ,  LHV ,  LVV Tracking forest fire extent with UAVSAR L-band po...
Onboard Product Examples: Forest Fuel Load July 2010 LHH ,  LHV ,  LVV UAVSAR data over Kings Canyon
July 2010 Low Fuel Load Forests Distribution of 1-hr Branch Fuel Load <ul><li>Simplified tree crown fuel load model using ...
July 2010 Medium Fuel Load Forests Distribution of 10-hr Branch Fuel Load Onboard Product Examples: UAVSAR Fuel Products
July 2010 High Fuel Load Forests Distribution of 100-hr Branch Fuel Load Onboard Product Examples: UAVSAR Fuel Products
Onboard Product Examples: Earthquake Damage Assessment July 2010 By comparing two post-earthquake amplitude images, we are...
Onboard Product Examples: Glacier Melting July 2010 L-band polarmetric image of the Kangerlugssuaq ice fjord in Eastern Gr...
Input Data: Training, Cross Validation & Evaluation SVM Output Visual Imagery Low statistical correlation with input data ...
Future Work <ul><li>Prototype additional rapid response application products </li></ul><ul><ul><li>Flood extent </li></ul>...
Upcoming SlideShare
Loading in...5
×

FR1.L09.2 - ONBOARD RADAR PROCESSING CONCEPTS FOR THE DESDYNI MISSION

744

Published on

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total Views
744
On Slideshare
0
From Embeds
0
Number of Embeds
0
Actions
Shares
0
Downloads
8
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide
  • 1-hour fuel: fallen needle, leaf needle, and small twigs – source of surface fire
  • 10-hour fuel load: small branches and woody stems. Due to their resistance to drying and greater heat capacity, 10-hour fuels often do not combust in low-intensity surface fires. When moisture is low, however, 10-hour fuels can carry hot fires and help ignite larger (100- and 1000-hour) fuels. Ten-hour fuels are readily consumed when fuel moistures are low.
  • Larger downed woody debris is common 100-hour forest fuels. These fuels take longer to dry, deterring their consumption under most conditions. Likewise, 100-hour fuels are slow to gain moisture, so they can combust after prolonged drought, even with recent precipitation. When 100-hour fuels ignite they can burn for hours, in mixtures of flaming and smoldering combustion. Decay of 100-hour fuels can alter their response and makes them combust more readily than intact fuels.
  • NLCD: water(black), dense veg(dark gray), low veg(light gray), bare/urban(white) Training Samples: ~6000 randomly chosen samples (pixels) from many images in the area. Selected image is just one example image (500x500 pixels). SVM: multi-class strategy: one-vs-one; gaussian radial-basis kernel Inputs: hh, hv, vv (all in db scale, topography &apos;removed&apos;), incidence angle, local incidence, 7x7 pixel average, 7x7 gaussian weighted average, 7x7 variance, hhvv phase, sin(phase), sin(local inc), sin(flat inc), rvi: 8hv / (hh+vv+2hv)
  • Transcript of "FR1.L09.2 - ONBOARD RADAR PROCESSING CONCEPTS FOR THE DESDYNI MISSION"

    1. 1. Onboard Radar Processing Concepts for the DESDynI Mission Yunling Lou, Steve Chien, Duane Clark, Joshua Doubleday, Ron Muellerschoen, and Charles Wang Jet Propulsion Laboratory California Institute of Technology Pasadena, California IGARSS 2010 26-30 July, 2010 Honolulu, Hawaii
    2. 2. Agenda <ul><li>DESDynI mission radar measurements overview </li></ul><ul><li>Benefits of onboard processing technology </li></ul><ul><li>Approach to onboard product development </li></ul><ul><li>Examples of onboard products </li></ul><ul><li>Future work </li></ul>
    3. 3. DESDynI Mission Radar Measurements Overview July 2010 Biomass Canopy Height 8 Tbit onboard storage 1 Gbps TDRSS link <ul><li>Onboard processing (OBP) will enable more frequent ecosystem and cryospheric science observations than currently feasible </li></ul><ul><li>Most disaster response applications would not be possible without onboard processing </li></ul><ul><ul><li>Biomass </li></ul></ul><ul><ul><li>Vegetation Structure </li></ul></ul><ul><ul><li>Vegetation Disturbance </li></ul></ul><ul><ul><li>Sea Ice Thickness </li></ul></ul>Simard et al (2006,2008) <ul><ul><li>Effects of changing climate on habitats and CO 2 </li></ul></ul>Polarimetric SAR <ul><ul><li>Water Resource Management </li></ul></ul><ul><ul><li>Geo-Hazards </li></ul></ul><ul><ul><li>Response of ice sheets to climate change & sea level rise </li></ul></ul><ul><ul><li>Sea Ice and </li></ul></ul><ul><ul><li>Ice Sheet Dynamics </li></ul></ul><ul><ul><li>Changes in Earth’s Surface </li></ul></ul>Pass 2 Pass 1 Repeat Pass InSAR 580 to 1160 Mbps 2000 Mbps Raw Data Rate
    4. 4. Potential Benefits of Onboard Processing Technology for DESDynI Mission July 2010 <ul><li>Interferograms and Single-look Complex (SLC) data can be compressed with ICER, a progressive, wavelet-based image data compressor designed for deep space applications such as the Mars Rover missions. ICER uses only integer math and can be implemented in FPGAs (courtesy: A. Kiely of Jet Propulsion Laboratory) </li></ul>Solid Earth Ecosystem Structure Cryospheric Science Potential Onboard Products Compressed interferogram* Forest biomass Soil moisture Vegetation classification Land use classification Sea ice classification Freeze/thaw maps Potential Downlink Data Volume Reduction Factor 10, but only for limited scenes that can be stored on board 1000 1000 for sea ice (non-interferometric) Potential Disaster Response Products Earthquake Flooding Lava flow Forest fuel load Hurricane damage in forest Flooding Ice melting Ship channel freeze/thawing
    5. 5. Other Potential Advantages of Onboard Processing for Earth Science Radar Missions July 2010
    6. 6. Proposed Onboard Processing Scenario for DESDynI Mission’s Radar Instrument July 2010 <ul><li>Downlink data volume reduction factor is directly proportional to the level of onboard processing involved: </li></ul><ul><ul><ul><li>With SAR image formation followed by image compression technique, a factor of 10 in data volume reduction is expected </li></ul></ul></ul><ul><ul><ul><li>With SAR image formation followed by science data processing for specific feature detection applications, downlink data volume can be reduced by a factor of 1000 </li></ul></ul></ul><ul><li>Ability to generate targeted products onboard will enable rapid response to natural hazards via Direct Broadcast (15 Mbps downlink rate) </li></ul>Onboard Processor Analyze and detect features Reprioritize downlink if needed Modify data acquisitions if needed DESDynI Onboard Autonomy Software Form SAR image Generate onboard product? Compress SAR image in preparation for calibration and post-processing on the ground Generate low resolution product (e.g. sea ice classification at 100 m resolution) Downlink data volume reduced by a factor of 1000 Downlink data volume reduced by a factor of 10 Ground Receiving Station No Yes
    7. 7. Example Mission Scenario with Autonomous Sciencecraft Experiment Image taken by Spacecraft Event Detection No event Detected: Delete Image Event Detected Onboard Science Analysis Track a wide range of science events – floods, volcanoes, cryosphere, clouds,… Key Insight: No need to replicate ground science analysis – just detect activity ASE uses state of the art Machine Learning to detect events in the presence of noise
    8. 8. High Level Architecture of the Onboard Processor <ul><li>Repetitive, computational-intensive operations are done in the FPGA </li></ul><ul><li>Radar parameters that are calculated every range or azimuth line are done in the microprocessor </li></ul><ul><li>Dual-channel processor can handle interferometric or dual-polarized processing </li></ul>July 2010 Product generation ≤ 2.5 Gbps PMC cards for high speed data transfer Fiber Pre-processor (Micro-processor) FPGA Processor 1 FPGA Processor 2 Post-processor (Micro-processor) Ethernet Raw radar data Control interface Control processor SAR image formation & Image compression Custom boards cPCI Chassis RocketIO interfaes
    9. 9. Approach to Onboard Product Development <ul><li>Determine the types of data processing suitable for implementation onboard the spacecraft to achieve the highest data compression gain without sacrificing science objectives </li></ul><ul><ul><li>SAR image formation </li></ul></ul><ul><ul><li>Polarimetric calibration </li></ul></ul><ul><ul><li>Multilooking </li></ul></ul><ul><ul><li>Cross-correlation </li></ul></ul><ul><ul><li>Simple classification techniques </li></ul></ul><ul><li>Adapt complex science algorithms for onboard processing to generate 100 m resolution products such as forest biomass, flood scene map, sea ice classification, and correlation map for change detection </li></ul><ul><li>Identify product applications currently not included in the mission objectives </li></ul>July 2010
    10. 10. Onboard Product Examples: Forest Fire Extent July 2010 LHH , LHV , LVV Tracking forest fire extent with UAVSAR L-band polarimetric data products Fire scars
    11. 11. Onboard Product Examples: Forest Fuel Load July 2010 LHH , LHV , LVV UAVSAR data over Kings Canyon
    12. 12. July 2010 Low Fuel Load Forests Distribution of 1-hr Branch Fuel Load <ul><li>Simplified tree crown fuel load model using LHH and LHV data is based on algorithm developed by Saatchi, et al. </li></ul><ul><li>We collected field data in Kings Canyon to verify the fuel load determination </li></ul>Onboard Product Examples: UAVSAR Fuel Products
    13. 13. July 2010 Medium Fuel Load Forests Distribution of 10-hr Branch Fuel Load Onboard Product Examples: UAVSAR Fuel Products
    14. 14. July 2010 High Fuel Load Forests Distribution of 100-hr Branch Fuel Load Onboard Product Examples: UAVSAR Fuel Products
    15. 15. Onboard Product Examples: Earthquake Damage Assessment July 2010 By comparing two post-earthquake amplitude images, we are able to identify old features (green) that have been removed (perhaps damaged buildings) and new features (red) that have been built (perhaps tent cities) over a two-week period. Amplitude Change Detection with UAVSAR’s 16-day Repeat Pass Data over Port Au Prince Airport, Haiti Jan 27, 2010 Feb 13, 2010
    16. 16. Onboard Product Examples: Glacier Melting July 2010 L-band polarmetric image of the Kangerlugssuaq ice fjord in Eastern Greenland. The grounding line of the glacier is easily identifiable in the image LHH , LHV , LVV Grounding line
    17. 17. Input Data: Training, Cross Validation & Evaluation SVM Output Visual Imagery Low statistical correlation with input data (noisy), high visual feature correlation April 22, 2006 Courtesy Google Earth Labels for training and cross validation statistics. National Land Cover Data 2001, condensed to 4 coarse classes. Polarimetric data, incidence angle, etc: training and final classification input. Support Vector Machine Onboard Product Examples: Vegetation Classification with Support Vector Machine water dense veg. light veg. bare/ urban
    18. 18. Future Work <ul><li>Prototype additional rapid response application products </li></ul><ul><ul><li>Flood extent </li></ul></ul><ul><ul><li>Lava flow </li></ul></ul><ul><ul><li>Hurricane damage </li></ul></ul><ul><ul><li>Ice melting (ship channel and ground thawing) </li></ul></ul><ul><li>Utilize machine learning to classify some of the products for event detection and onboard replanning </li></ul><ul><li>Prototype products for long term monitoring applications </li></ul><ul><ul><li>Vegetation type classification </li></ul></ul><ul><ul><li>Soil moisture monitoring </li></ul></ul>July 2010
    1. A particular slide catching your eye?

      Clipping is a handy way to collect important slides you want to go back to later.

    ×