An Improved Compressed Sensing Algorithm and Its Application in SAR Imaging

To realize high-resolution SAR imaging, the amount of the raw data required for imaging is very large. Compressed Sensing (CS) can sample the raw data at a frequency lower than Nyquist frequency and image without losing resolution. In this paper, we propose an improved CS algorithm and employ it in SAR imaging to reduce the amount of data required for imaging. This algorithm improve the reconstruction by decreasing the mutual coherence among the atoms in measurement matrix. By using this method in traditional imaging and SAR imaging, we can achieve at the higher compression ratio as well as the better imaging quality.