Compressive support detection in SAR tomography

In this paper, the problem of detecting and localizing multiple scatterers in SAR tomography, starting from compressed measurements is considered. This problem can be addressed as the detection of a sparse signal within the compressed domain and can be approached in the framework of Compressive Sensing (CS) theory. While CS literature has focused on the problem of signal reconstruction, this is frequently not necessary. For instance, in radar systems the main purpose is target detection, which does not necessarily requires a reconstruction of the signal. In this paper, different detectors able of reducing the number of measurements needed for a given detection performance are considered. The detection schemes analyzed are based on support detection techniques, i.e. on the detection of the position of the non-zero elements in the unknown sparse vector, which have already proved to allow a reduction in the number of measurements required for obtaining a reliable solution. Performance evaluation on simulated data is presented.