Compressed Sensing for Removing Moving Target Artifacts and Reducing Noise in SAR Images

Abstract:
The presence of moving targets corrupts SAR images, since their image is defocused and shifted compared to the image of non moving scatterers. Moreover, thermal noise reduces the contrast of the SAR image. In this paper we introduce a compressed sensing based method that separates SAR-GMTI data in contributions from different velocities and from noise by computing local angle-Doppler diagrams. The SAR images computed from those components representing non moving scatterers have higher SNR and include no moving target artifacts.