ABSTRACT

The essence of remote sensing resides in the acquisition of information about remote targets for further processing. As a high resolution microwave remote sensing instrument, the Synthetic Aperture Radar (SAR) has been more and more widely used. The data compression is one of the most important digital signal processing stage in remote sensing. The traditional compression algorithm is the Block adaptive quantization (BAQ) due to its simplicity in implementation and results. The theoretical foundation of BAQ is the distribution of raw SAR data but in fact, the raw data is not Gaussian distributed especially when there is some saturation with the receiver. In order to overcome this drawback, the authors have studied the correlation between the mean value of the signal and its standard deviation. We also evaluated the correlation between the mean input signal and standard deviation of the output signal from the A/D. Monte-Carlo experiment shows that none of the above two correlations are optimal in the whole data set . Thus, we propose a new algorithm which gives optimum results irrespective of the degree of saturation in whole range of data. Results obtained from simulated data and real data show that the performance of new algorithm is better than conventional BAQ especially when raw data is heavily saturated. The authors have used data received from Indian satellite Chandrayaan-1 and European Space Agency (ESA) in order to carry out the experimental results.