归一化自适应预测矢量量化算法压缩SAR原始数据

Compression of SAR Raw Data with Normalized Adaptive Predictive Vector Quantization

该文提出归一化自适应预测矢量量化(NAPVQ)算法压缩SAR原始数据。NAPVQ算法先采用矢量线性预测器对输入矢量进行预测,再对原矢量与预测矢量之间的残差矢量进行矢量量化。该算法可视为差分脉冲调制在矢量量化中的拓展,其性能优于块自适应量化(BAVQ)算法以及归一化预测自适应量化(NPAQ)算法。对算法复杂度的进一步分析表明,NAPVQ算法能获得复杂度和性能之间比较合理的折衷,具有实用价值。

This paper presents a new SAR raw data compression algorithm named Normalized Adaptive Predictive Vector Quantization （NAPVQ）. The normalized SAR raw data are firstly processed with vector linear predictor, and then the error vectors are compressed with vector quantizer. Regarded as an extension of the differential pulse code modulation in the vector quantization, the NAPVQ achieves a better performance gain than the Block Adaptive Vector Quantization （BAVQ） and the Normalized Predictive Adaptive Quantization （NPAQ）. With the analyses of algorithm complexity, the proposed algorithm shows a good performance/complexity trade-off and accommodates the requirement in engineering applications.