Real-Time Data Processing in Sparse SAR Imaging

Sarsity-driven synthetic aperture radar (SAR) imaging has already shown the superiority in terms of imaging performance improvement and image recovery from down-sampled data. However, conventional observation matrix based approach generally requires inhibitive memory space for the target recovery especially for the large-scale scenes. Although azimuth-range decouple based method significantly reduces the computational cost of sparse SAR imaging, it also needs one or two orders more computing time than typical matched filtering (MF) based algorithm, which does not allow the real-time processing. Therefore, further minimizing the computational cost of sparse SAR imaging is needed. In this paper, we present a novel imaging technique that supports real-time data processing in sparse SAR imaging. With experimental results on real data, the presented method is very useful for achieving desired imaging performance in practical applications, and has been successfully reduced the computational time to the same order as MF based method.