Focusing of synthetic aperture radar images of moving targets using minimum entropy adaptive filters

Synthetic aperture radar (SAR) imaging systems are used for generating high-resolution images for target detection and identification. However, when imaging moving targets, the imaging process is degraded by target motion. These effects generally manifest as a form of smearing in the azimuthal direction. Since the motion parameters are generally not known, it requires a processing technique that can estimate the motion information from the SAR measurements and use these estimates to reconstruct the image. Adaptive filters are well suited to do this. In this study, an adaptive technique based on the minimum entropy filter is developed to minimize the effect of smearing in SAR images of moving targets. The minimum entropy filter utilizes the entropy function as the cost function to be optimized. The use of the minimum entropy filter with simulated SAR measurements of moving targets clearly shows significant improvement in the image quality when the entropy function is optimized. The filters were tested on images of targets with many different motion parameters. The minimum entropy filter consistently produced the best motion compensation. The filters were also tested with field images of ships obtained from a C-band, SAR flown over the coast of Denmark. The filter was able to adaptively reduce the effects of azimuthal smearing.