一种基于内插阵列变换的变速SAR成像算法  

A Imaging Algorithm for Shift-Velocity SAR Based on Interpolated Array Transform

作　　者：彭岁阳[1] 胡卫东[1] 

机构地区：[1]国防科技大学电子科学与工程学院ATR实验室,湖南长沙410073

出　　处：《信号处理》2009年第11期1742-1747,共6页Journal of Signal Processing

摘　　要：在成像平台非匀速运动条件下,SAR数据在方位方向不再是均匀采样,原有的SAR成像算法的适用条件不再满足。本文试图从均匀采样数据重建的角度研究非均匀采样下的SAR模型,借鉴非均匀阵列DOA估计中的内插阵列变换思想,提出了重构虚拟均匀采样数据的算法思想。通过对构造方法需要满足条件的分析,本文推导了基于内插阵列变换的预处理算法。仿真实验表明本文方法能对非均匀采样SAR数据成较清晰的SAR图像。本文针对位置测量存在误差的情况作了性能分析,结果表明该算法在SAR实际测绘精度下仍然具有较好的适应性。Under the non-uniform flying, the SAR data is not uniform sample data in the azimuth direction, and the traditional SAR imaging algorithm could not longer be adapted. We try to research a SAR model under non-uniform sampling condition from the perspective of uniform data reconstruction. Learning from the idea of the interpolated array transform of non-uniform DOA, the algorithm is put forward to construct the virtual sampling data. By the analyses of the conditions which the construct method needs satisfy, the paper brings forward a pretreatment algorithm based on interpolated array transform. The formulation experiment shows that, with the paper＇ s method put forward, fine images can be formed under the non-uniform sampling condition. The capacity of algorithm is analyzed under the condition that the position error existed, and the result shows the algorithm also has fine adaptability under the precision of the real SAR mapping.

关 键 词：非均匀采样 内插阵列变换 数据重构 RD算法 协方差 

分 类 号：TN957.52[电子电信—信号与信息处理]