Computationally Efficient Forward/backward Averaged DOA Estimation of Coherent Sources in Pairs

DOA estimation of signal sources in multipath environments is a computationally complex problem. In this paper, authors present a computationally efficient DOA estimation algorithm effective against highly correlated or fully coherent sources in pairs. The proposed method is novel in that it applies forward/backward averaging to the signal subspace to de-correlate the signals, unlike existing methods that apply spatial smoothing techniques to the correlation matrix. This significantly reduces the computational complexity and computation time of the algorithm and improves the estimation accuracy making the proposed method amenable to practical hardware implementation. Simulation results are presented to validate the efficacy of the proposed method and a performance comparison is made with Root-MUSIC method.