Study On High Ratio Compression Algorithms Of SAR Raw Data

As a high-resolution wide area imaging radar, the Synthetic Aperture Radar (SAR) system generates a big amount of data to be transmitted and processed. In the case of satellite with store and forward function, data storage and transmission become a problem since the buffer capacity downlink bandwidth is limited. Therefore, there is a need to reduce the raw data set. And the data compression algorithm is an efficient method to solve this problem. This paper is aimed at study on the SAR raw data compression algorithm.In this paper, the statistics and power characteristic of SAR raw data are analyzed, which are the theoretic foundation of the raw data compression algorithms. After introducing the optimized scale quantization and the rate-distortion theory, the LBG algorithm is described in detail. The traditional compression algorithm of the Block Adaptive Quantization (BAQ) and Vector Quantization (VQ) are reviewed for the future use.Aiming at enhancing the performance of the SAR raw data compression algorithms, the Block Adaptive Vector Quantization (BAVQ) algorithm is improved. The factors that affect the VQ compression performance in BAVQ are analyzed, which are the distribution, correlation and dynamic range of BAQ compressed data, and the Signal to compression Noise Ratio (SNR) of BAQ. Based on these factors, the idea of replacing the traditional block adaptive quantizer by more bits one is proposed. Fast search algorithm for VQ is applied in the improved algorithm, which can reduce the complexity of the improved algorithm effectively, and without reducing its compression performance. According to the different image qualities request of the communication systems, the improved algorithm can compress raw data with different data rate, Using this improved BAVQ algorithm to compress high resolution SAR raw data, larger SNR values than the previous compression algorithms under the samecompression ratio are obtained. The decompressed images reserve most of details on the images resulted from raw data.In order to realize high ratio compression, a new Wavelet Transform coding for SAR raw data compression is proposed. For simplifying the new algorithm, the coding scheme sets all the WT coefficients in the sub-band to zero if which is not essentiality. This new algorithm can compress raw data to 16:1. Using this WT algorithm to compress high resolution SAR raw data, under the large compression ratio, the decompressed images can reserve important details on the images resulted from raw data. This new algorithm is suitable for the communication systems which need higher compression ratio and lower SAR imaging qualities.