Synthetic Aperture Radar (SAR) - Geometric Calibration

Contact Persons

David Small

Erich Meier

Keywords

synthetic aperture radar, SAR, geometry, calibration, geolocation, corner reflector, transponder

Quick Reference

Radar works by measuring range timing. The timing needs to be calibrated to ensure that the range measurements made (time or distance) are valid, and not subject to systematic biases, such as uncertainty in the onset of the sampling window start time (SWST). We describe the methodology used at RSL to calibrate and validate the ENVISAT ASAR and other radars. Recently, we used corner reflectors deployed in Switzerland to calibrate the geometry of the TerraSAR-X radar, launched in 2007. At the same time, atomspheric modeling was used to compensate the path delay ("slowing") effect, with the resulting geolocation accuracy of well under 1 m.

Methods

The location of a point on the Earth may be predicted within a given radar image by solving the range and Doppler equations. The accuracy of the timing annotations (both along-track (azimuth) and cross-track (range)) associated with the given image product must be guaranteed to enable non-iterative accurate location prediction.

Artificial targets such as corner reflectors or transponders (see Figure 1) provide strong reference signals that may be used for geometric and radiometric calibration.

Methods

The location of a point on the Earth may be predicted within a given radar image by solving the range and Doppler equations. The accuracy of the timing annotations (both along-track (azimuth) and cross-track (range)) associated with the given image product must be guaranteed to enable non-iterative accurate location prediction.

 

Artificial targets such as corner reflectors or transponders (see Figure 1) provide strong reference signals that may be used for geometric and radiometric calibration.

Figure 1: ESA-ESTEC Active Transponder (left) and Passive Trihedral Corner Reflector (right)

Results

Figure 2 shows the predicted location (blue cross) of a corner reflector in Dübendorf, Switzerland given

• restituted quality state vectors (left),
• precise quality state vectors (middle),
• precise quality state vectors and sampling window start time (SWST) correction applied (right)

Fifteen azimuth and three range looks are taken in the top row, five azimuth and one range look in the middle row. The bottom row shows the image's native single look complex (detected) geometry, one azimuth and one range look respectively.

Figure 2: Predicted and Measured Corner Reflector Locations in ENVISAT-ASAR Image Mode (IM) slant range image

Figure 3 juxtaposes three predictions of a transponder location in an image of the Flevoland area in the Netherlands given three different annotation qualities.

The quality of the prediction improves as more accurate input information (orbital state vectors, radar timing annotations) becomes available.

Figure 3: Predicted and Measured Transponder Locations in ENVISAT-ASAR alternating polarisation (AP) slant range image

Partnerships

Research has been conducted in cooperation with the following organizations:

• European Space Agency
• German Aerospace Centre (DLR), Oberpfaffenhofen, Germany
• Joanneum Research, Graz, Austria
• Canadian Space Agency
• Canada Centre for Remote Sensing

Publications

• SMALL D., SCHUBERT A., ROSICH B., MEIER E. [2007]:
  Geometric and Radiometric Correction of ESA SAR Products, (PDF, 3.5 Mb)
  Proc. of the Envisat Symposium 2007, Montreux, Switzerland, Apr. 23-27, 2007 (ESA SP-636, July 2007). 6p.
• SMALL D., ROSICH B., SCHUBERT A., MEIER E. [2006]:
  ENVISAT ASAR Geometric Validation,
  Proc. of CEOS SAR Cal/Val Workshop, Edinburgh, Scotland, Oct. 3-6, 2006 (in press).
• SMALL D., ROSICH B., SCHUBERT A., MEIER E., NÜESCH D. [2005]:
  Geometric Validation of Low and High-Resolution ASAR Imagery, (PDF, 1.6 Mb)
  Proc. of the 2004 ENVISAT & ERS Symposium, Salzburg, Austria, Sept. 6-10, 2004 (ESA SP-572, April 2005). 9p.
• SMALL D., ROSICH B., MEIER E., NÜESCH D. [2004]:
  Geometric Calibration and Validation of ASAR Imagery (PDF, 3.6 Mb),
  Proc. of CEOS SAR Workshop 2004, Ulm, Germany, May 27-28, 2004.
• SMALL D., HOLZNER J., RAGGAM H., KOSMANN D., SCHUBERT A. [2003]:
  Geometric Performance of ENVISAT ASAR Products (PDF, 1.3 Mb), 
  Proc. of IGARSS'03, Toulouse, France, July 21-25, 2003, pp. 1121-1123.
• SMALL D., SCHUBERT A., KRÜTTLI U., MEIER E., NÜESCH D. [2003]:
  Preliminary Validation of ASAR Geometric Accuracy (PDF, 1.6 Mb), 
  Proc. of ENVISAT Validation Workshop, Frascati, Italy, Dec. 9-13, 2002 (ESA SP-531, August 2003). PDF
• SMALL D., KOSMANN D., HOLZNER J., RAGGAM H., PIRRI M., SCHUBERT A., KRÜTTLI U., HUMMELBRUNNER W., FRANKE M. [2002]:
  ASAR Level 1 Geolocation (PDF, 1.0 Mb), 
  Proc. of ENVISAT Calibration Review, Noordwijk, The Netherlands, Sept. 9-13, 2002. PDF
• SMALL D., BIEGGER S., NÜESCH D. [2000]:
  Automated Tiepoint Retrieval through heteromorphic Image Simulation for Spaceborne SAR Sensors (PDF, 4.3 Mb), 
  Proc. of ERS-ENVISAT Symposium 2000, ESA Publication SP-461, Gothenburg, Sweden, Oct. 16-20, 2000. PDF (PDF, 4.3 Mb)
• SMALL D., HOLECZ F., MEIER E., NUESCH D., and BARMETTLER, A. [1997]:
  Geometric and Radiometric Calibration of RADARSAT Images, CDROM Proc. of Geomatics in the Era of   Radarsat (GER'97), Ottawa, Canada, May 24-30, 1997.  CSA Abstract & PDF --  Local PDF (PDF, 787 Kb)