The Central DOA Estimation Algorithm for Multiple Coherently Distributed Sources

We consider the problem of estimating the central direction of arrival (DOA) of multiple coherently distributed sources. This problem is encountered due to the presence of local scatters in the vicinity of a transmitter or due to signals propagating through a random inhomogeneous medium. Since the integral steering vector of coherently distributed source can be deduced to be a Schur-Hadamard product comprised of point source steering vector and a real vector, a rotational eigenstructure is showed to exist accurately for two identical closely spaced subarrays. And then the central DOA can be obtained analytically without any peak-finding searching, which significantly reduces the computational complexity. In addition, the proposed algorithm needs little prior knowledge about angular signal intensity. The simulation results illustrate the better performance of the proposed method.