基于数据统计特性的合成孔径雷达盲成像算法

Synthetic-aperture-radar blind image formation algorithm based on the data statistics

为了解决合成孔径雷达(SAR)运动测量单元精度不高对成像的影响,提出了一种SAR盲成像算法。该算法利用SAR接收数据在距离Doppler域的统计特性,通过对比度最大化方法搜索正确的距离徙动轨迹,实现了在Doppler参数未知情况下的数据校直;还利用了不依赖速度初值的自聚焦方法。给出了实现步骤。星载SAR实测数据实验结果表明,该算法可以在无任何关于SAR平台运动的先验信息的情况下,实现高质量成像。

A blind image formation algorithm for synthetic aperture radar （SAR） was developed to mitigate the image degradation arising from inaccuracies in the motion measurement unit （MMU）. The algorithm aligns the raw data without knowledge of the Doppler parameters by searching for the correct range migration trajectory which has the maximum contrast using the statistical properties of the received raw data in the range-Doppler domain. The algorithm then uses auto-focusing which does not require SAR flying velocity as the initial input. Experiments with real data from a space-borne SAR show that the approach can perform SAR data high-quality image formation processing with no a priori information about the SAR platform motion.