Through-the-wall radar imaging based on modified Bayesian compressive sensing

In this paper, a novel modified complex multi-task Bayesian compressive sensing (MCMT-BCS) algorithm is proposed to acquire high-resolution images in stepped-frequency through-the-wall radar imaging (TWRI) exploiting multipath. Unlike traditional TWRI approaches that assume frequency-independent scattering model, we develop a practical subband scattering model to characterize real-world scattering mechanisms. The target imaging is reformulated as a multi-task sparse signal recovery problem across all frequency subbands as well as multipath modes, where the sparse entries of each task share the same support in the imaged scene. The proposed MCMT-BCS algorithm accounts for both types of coexisting group sparsity to achieve improved high-resolution imaging capability. Simulation results verify the effectiveness of the proposed algorithm.