A Novel Greedy Algorithm for Joint Sparse Recovery Through Information Transfer
Nam Yul Yu · 2018
In multiple measurement vectors (MMV) problems, L measurement vectors of length M are available for recovering jointly sparse signals that have a common support set of size K. This paper proposes a novel greedy algorithm for joint sparse recovery in MMV problems, by exploiting a posteriori probability ratios for every index of sparse input signals. The essence of the algorithm is to transfer the information through iterations, which improves the recovery performance of support indices. Simulation results demonstrate that if M is not too small in K <; M regime, the proposed algorithm can be more reliable and noise-robust than conventional ones, such as simultaneous orthogonal matching pursuit (SOMP), subspace augmented MUSIC (SA-MUSIC), and rank-aware order recursive matching pursuit (RA-ORMP).