MSM-FOCUSS for distributed compressive sensing and wideband DOA estimation

In order to recover an ensemble of signals which share a common sparsity pattern while being measured under different sensing matrices, we explore the MSM-FOCUSS algorithm by extending the well-known M-FOCUSS (FOCal Under-determined System Solver) approach from the MMV (Multiple-Measurement-Vectors) to the MSM (Multiple-Sensing-Matrices) scenario. The convergence of the algorithm, the sparsity of the solution and the uniqueness condition for the MSM problem are analyzed. The performance is demonstrated by examples of joint sparse representation in distributed compressive sensing and wideband DOA (Direction-Of-Arrival) estimation.