Deconvolution techniques for non-coherent radar images

This paper presents a new algorithm for deconvolution of low resolution non-coherent radar images. The technique presented is based on an LMS iterative approach, but has important modifications to accommodate the constraints imposed by the absence of phase information in the radar return. The algorithm also operates at a higher resolution than the sample points, in order to compensate for the lack of range and azimuth resolution. This new technique is compared to standard iterative algorithms in both the time and frequency domains. Simulated radar data is used to demonstrate the benefits of the algorithm.