DOA estimation of coherent signals based on matrix reconstruction

Abstract

DOA estimation is an important part in the study about target precision locating. For the incoherent source signals, the conventional algorithms such as MUSIC have a good performance in processing for the estimation of the directions-of-arrival (DOA's). But when the signals are correlated, most of the conventional algorithms for incorrelate signals will fail in the estimation of DOA'S. A novel algorithm about the estimation of DOA's for correlated source signals is proposed in this paper. In the proposed algorithm, we reconstruct a new correlation matrix based on the original correlation matrix. The novel algorithm can accurately estimate the DOA's of the coherent incident signals without spatial smoothing processing compared with MMUSIC algorithm. For wide coherent signals, the proposed algorithm still has a good performance compared with ISM and CSM. Simulation results are presented in Section IV and Section V to illustrate the efficiency and accuracy of this method.