Optimal block quantization for SAR data

A variable bit rate coding scheme designed for new generation SAR satellites is presented. This coding is based on previous schemes of Block Adaptive Quantizer (BAQ) [1] and is intended as a possible improvement over Sentinel-1 FDBAQ [3,4]. In this project, the number of quantizers is considered a design choice and the optimization is carried out looking for a set of reflectivity classes in which partition all the imaged areas on the Earth. Superior performances to those envisaged for Sentinel-1 FDBAQ are obtained in terms of SNR vs. average rate. Results come from simulations based on a 16-bit mosaic of C band reflectivity obtained by ESRIN from ENVISAT GM data [6] and Wien University database of ENVISAT WSM radar reflectivity [10].

[1]  Jon Hamkins,et al.  Gaussian source coding with spherical codes , 2002, IEEE Trans. Inf. Theory.

[2]  Paul Snoeij,et al.  FDBAQ a novel encoding scheme for Sentinel-1 , 2009, 2009 IEEE International Geoscience and Remote Sensing Symposium.

[3]  Paul Snoeij,et al.  GMES Sentinel-1 FDBAQ performance analysis , 2009, 2009 IEEE Radar Conference.

[4]  Pedro Gonçalves,et al.  Generation of ENVISAT ASAR Mosaics accessible on-line , 2007, 2007 IEEE International Geoscience and Remote Sensing Symposium.

[5]  Ciro Cafforio,et al.  Flexible Dynamic Block Adaptive Quantization for Sentinel-1 SAR Missions , 2010, IEEE Geoscience and Remote Sensing Letters.

[6]  Evert Attema,et al.  Stability of Amazon Backscatter at C-Band: Spaceborne Results from ERS-1/2 and RADARSAT-1. , 2000 .

[7]  Ian G. Cumming,et al.  ENVISAT ASAR data reduction: impact on SAR interferometry , 1998, IEEE Trans. Geosci. Remote. Sens..

[8]  Allen Gersho,et al.  Principles of quantization , 1978 .

[9]  Jorge Nocedal,et al.  A trust region method based on interior point techniques for nonlinear programming , 2000, Math. Program..