SAR原始数据两种量化压缩方式的性能评估  

Performance Evaluation of Two Compression Methods for SAR Raw Data

作　　者：仇晓兰[1,2] 雷斌[1,2] 葛蕴萍[1,2] 胡东辉[1,2] 丁赤飚[2] 

机构地区：[1]中国科学院空间信息处理与应用系统技术重点实验室,北京100190 [2]中国科学院电子学研究所,北京100190

出　　处：《电子与信息学报》2010年第9期2268-2272,共5页Journal of Electronics & Information Technology

基　　金：国家863计划项目(2009AA7010113)资助课题

摘　　要：原始数据压缩是高分辨率SAR需要着重考虑的环节。该文针对目前星载SAR比较常用的4bit量化和8:3bitBAQ两种量化压缩方式,在不同输入功率下对量化信噪比、功率保真能力和目视图像效果这几个方面进行了深入的分析和评估。分析得出,综合各项性能后4bit和8:3bit BAQ的最佳输入功率范围分别为26.1dB～34.9dB和23.5dB～36.5dB,这为星载SAR实际运行中MGC值的设置提供了依据。Raw data compression is an important step for high resolution SAR systems. This paper bases on two data compression methods,the 4 bit quantization and the 8 to 3 bit Block Adaptive Quantization （BAQ）,which are widely used in spaceborne SAR,the evaluations are made on the Signal-to-Noise Ratio （SNR）,the fidelity of the amplitude,and the vision effects of the images under different quantization input power. From the evaluations,the reasonable power regions of the quantization input for 4 bit and 8：3 bit BAQ are provided,which are 26.1 dB～34.9 dB and 23.5 dB～36.5 dB,respectively. This offers a good reference to the Manual Gain Control （MGC） value for the SAR mission.

关 键 词：SAR 数据压缩 分块自适应量化 信噪比 

分 类 号：TN958[电子电信—信号与信息处理]