Detection performance of compressively sampled radar signals

Compressed sensing and sparse reconstruction techniques have been applied to radar signal acquisition, reconstruction of the range-Doppler map, and radar imaging. Most prior work in this area focuses on imaging and reconstruction while little attention has been paid to the effects of these techniques on detection performance. In this paper, we study the detection performance of signals acquired in an undersampled manner via random projections. We compare detection performance for signals acquired via traditional sampling of the matched filter, for correlation detection that operates directly on compressed measurements, and for the matched filter applied to the signal reconstructed via basis pursuit denoising.