A simultaneous sparse approximation approach for DOA estimation

Compressed sensing (CS) is one of the growing fields in electrical engineering and applied mathematics finding a number of applications in array processing as well. Direction-of-arrival (DOA) estimation is one of the main problems in array signal processing. In recent years, some CS techniques have been used to estimate DOA. In this paper, a novel algorithm is developed based on sparse multiple measurement vector model (MMV) to estimate DOAs of far-field and narrowband sources in sensor arrays. The former algorithm exploits singular value decomposition to remove noise from output matrix of array and improve support recovery. In this work, we focus on uniform linear arrays. Simulation results demonstrate the superiority of the proposed algorithm over well-known existing approaches, specifically simultaneous orthogonal matching pursuit (S-OMP), ℓ_2,1 minimization and Root-MUSIC.