基于压缩感知和矢量量化的SAR数据级联压缩方法  

A Cascade Compress Method of SAR Data Based on Compressed Sensing and Vector Quantization

作　　者：陈一畅 张群[1] 朱丽莉[1] 顾福飞[1] 李平平 

机构地区：[1]空军工程大学信息与导航学院,西安,710077 [2]解放军93582部队,山西朔州,038301

出　　处：《现代雷达》2013年第10期36-40,共5页Modern Radar

基　　金：国家重点基础研究发展计划(973计划)项目(2010CB731905);国家自然科学基金项目(61201369);广西无线宽带通信与信号处理重点实验室2001年度开放基金项目(21102)

摘　　要：针对合成孔径雷达回波数据量大,不便于传输的问题,提出了一种新的SAR数据压缩方法。首先运用压缩感知(CS)理论对雷达回波信号进行降采样,其次利用块自适应树形矢量量化算法(BATSVQ)对降采样的SAR回波数据进行编码压缩,然后进行传输。在地面接收端或者预警指挥机,通过BATSVQ解码和CS重构算法,高精度地恢复出SAR数据在距离多普勒域的稀疏表示,再利用线调频变标算法实现二维高分辨成像。最后对级联压缩方法的压缩比作了定量分析,新的压缩方法能够在大幅压缩回波数据基础上实现成像。仿真实验验证了该方法的可行性和有效性。A new kind of compression method is proposed to solve the problem of transmitting huge SAR data. Firstly,compressed sensing( CS) is used to down-sample the radar echo signal. Secondly,the down-sampled SAR echo data is compressed and encoded by using the block adaptive tree-structure Vector Quantization( BATSVQ),and then be transferred. Meanwhile,the sparse representation of SAR data in the range-doppler domain is accurately recovered by the algorithms of BATSVQ decoding and CS reconstruction,which is conducted in the receiver or the warning command on the ground. Afterward,the chirp scaling algorithm is executed to achieve 2-d high resolution image. Lastly,the compression ratio of this compression method is quantitative analyzed,which shows that this new method could still imaging on the basis of slash echo data. The effectiveness of the proposed method can be validated by simulation results.

关 键 词：SAR数据 压缩感知 块自适应树形矢量量化算法 压缩比 

分 类 号：TN958[电子电信—信号与信息处理]