Entropy-constrained predictive compression of SAR raw data

Abstract

In this paper we propose to employ entropy-constrained predictive coding for lossy compression of SAR raw data. We exploit the known result that a blockwise normalized SAR raw signal is a Gaussian stationary process in order to design an optimal decorrelator for this signal. The proposed predictive coding algorithm performs entropy-constrained quantization of the prediction error, followed by entropy coding; the algorithm exhibits a number of advantages, and notably a very high performance gain, with respect to other techniques such as FBAQ or methods based on transform coding. Simulation results on real-world SIR-C/X-SAR as well as simulated raw and image data show that the proposed algorithm significantly outperforms FBAQ as to SNR, at a computational cost compatible with modern SAR systems.