BCS Based MIMO Sparse Array Synthesis for Near-Field Imaging

This paper proposes a Multiple-Input Multiple -Output (MIMO) sparse array synthesis (SAS) method for wideband near-field imaging based on multi-task Bayesian compressive sensing (MT BCS). By applying the classic SAS model, MT BCS is used to statistically combine the real and imaginary parts of complex excitations with the shared prior to construct the complex excitations. Specifically, MT BCS adds elements with non-zero weights successfully to the model until all elements are included. This greatly reduces computation complexity. Hence, the proposed method exhibits obviously improved synthesizing speed over the classic CS-based MIMO SAS method. Finally, the efficacy of the proposed method is verified through the experimental results.