基于星载SAR的原始数据压缩算法

The Compression Algorithms on Space Borne SAR Raw Data

星载SAR的原始数据量大,受到星上设备的数据处理能力和下行链路带宽的制约,需要把原始数据进行有效的压缩以便在系统允许的条件下存储和传输。因此必须采取一种能提供较高压缩比和较低编码/解码误差的高效SAR原始数据压缩算法,将SAR的原始数据进行压缩。分块自适应量化(BAQ)算法,其具有计算简单、实施方便的特点。为了进一步提高数据的压缩比和压缩质量,不断有更多的算法被提出,如矢量量化(VQ)、栅格编码矢量量化(TCVQ)等,对BAQ、VQ和TCVQ三种算法加以研究并给出相应的实验结果,最后对以上算法的综合性能做一个评述。

The raw data of the space borne Synthetic Aperture Radar （SAR） becomes more and more large. Limit to the on-board processor and downlink channel, the available capacity of satellite＇s mass data storage medium and downlink is restricted. Therefore, it is highly desirable to compress the data before transmission or storage, in order to meet the system constraint. So a Raw data compression method with high compression rate and lower error rate of coding/decoding need to be employed to compress the SAR raw data. Block-adaptive quantization （BAQ） algorithm is one of the earliest methods for practical use. This method is easy to programming and application. In order to improve of compression rate and compression quality further, some new methods are put forward such as vector qumltization （VQ） and Trellis- Coded Vector Quantization （TCVQ） etc. Considering the actual application and convenience of achieving by hardware, three algorithms such as BAQ, VQ and TCVQ are researched in this paper. Some experimental results are achieved. And lastly, a general comment of the three algorithms mentioned above is drawn.