Compressive sensing for synthetic aperture radar in fast-time and slow-time domains

Compressive sensing (CS) is a new method to capture and represent compressible signals at a rate significantly below the Nyquist rate. To reduce the high data redundancy among different echoes in synthetic aperture radar (SAR), we apply compressive sensing to SAR in slow-time and fast-time domains. Based on the reflectivity kernel analysis of SAR echoes, We demonstrate that the SAR signals are very sparse, which means CS could be applied to SAR to tremendously reduce the sampling rate. We apply SVD-QR to select a subset of SAR echoes in slow-time domain to reduce the redundancy, and compressive sensing to the selected echoes in fast-time domain. Simulations are performed and our compression ratio could be 47:1 overall.