Direction Finding with List-Based Orthogonal Matching Pursuit for Sparse Arrays
WESLEY SOUZA LEITE, Rodrigo C. de Lamare · 2021
This work devises a mixed ML-greedy algorithm termed List-Based Maximum Likelihood Orthogonal Matching Pursuit (LBML-OMP) for direction finding with sparse linear arrays. In the proposed technique, a subset of candidate atoms from the dictionary is selected based on the correlation between the residue and the dictionary. After that, LBML-OMP employs a restricted asymptotic maximum likelihood decision rule to properly choose the right atom from this subset. Numerical simulations demonstrate that LBML-OMP outperforms existing algorithms like Orthogonal matching Pursuit (OMP), Iterative Hard Thresholding (IHT), and Spatial Smoothing MUSIC (SS-MUSIC) with sparse linear arrays (SLAs), while requiring a modest increase in computational cost.