Raster Coded Quantization In Synthetic Aperture Radar Data Compression Applications

The data compression technique is an essential one of Synthetic Aperture Radar (SAR). However, the ratio of the data is greatly dependent on the improved radar's resolution and the increasing swath. Techniques in SAR, especially in space-borne ones, develop toward achieving both raw data acquisition, which is the current state of the art, and on-board Real-time Imaging. Therefore, data compression algorithm will be developed to realize compressing both raw data and SAR images in the future.After reviewing compression algorithms in SAR, the author introduces a new compression algorithm: Trellis Coded Quantization (TCQ). The TCQ algorithm takes advantage of convolution code, and enlarges the Euclidean distances among quantized data by expanding data dimensions.The author puts forward a method of uniform TCQ algorithm to compress SAR raw data and images. According to real-time imaging technique on space-borne SAR, and then designs a multi-state trellis with for compressing tests. The principle of the method lies in: TCQ is a general source-coding algorithm to process various data. However, the BAG, BFPQ algorithms, are developed only in SAR applications. The performance of TCQ seems to be higher SNR than BAQ in compressed data. The design scheme with a multi-state trellis states also shows a better SNR performance than current reported schemes. The author, in the meanwhile, uses TCQ to implement SAR image compressions. The result proves that the compressed images are high PSNR and low distortion characters, which is suitable to be used in space-borne and airborne SAR systems.