Abstract:

Multiband quadrature compressive sampling (MQCS) can perform sub-Landau sampling for multiband analog signals, and acquire the compressive inphase and quadrature components of each subband signal. For multiband radar applications, it is required to solve multiple large-scale sparse reconstruction problems for recovering all subband echoes, which need huge computational cost and thus have poor real-time recovery ability. This paper considers an equal-bandwidth multiband radar for detecting a radar scenario. With the analyses on the MQCS radar, it is found that the radar echoes are not only sparse but also block sparse and then block sparse reconstruction methods can be used to recover the echoes. For real-time or fast reconstruction, a block-sparsity-based segment-sliding reconstruction (BSegSR) scheme is proposed. To improve the reconstruction performance, a two-step block orthogonal matching pursuit (TBOMP) algorithm is developed on the basis of the block orthogonal matching pursuit (BOMP) algorithm. Simulation results validate the effectiveness and the correctness of the BSegSR scheme.