Decoupled Nominal 2-D Direction-of -Arrival Estimation Algorithm for Coherently Distributed Source

Abstract

A computationally efficient method for nominal 2-D (azimuth and elevation) direction-of-arrival (DOA) estimation of coherently distributed source impinging on the far field is presented. Since the coherently distributed source is characterized by four parameters, the nominal azimuth DOA, angular spread of the nominal azimuth DOA, the nominal elevation DOA, and angular spread of the nominal elevation DOA, the computational complexity of the parameter estimation is normally high demanding. So a low complexity estimation algorithm is proposed in this paper, the key idea of which is to apply a subspace-based method without eigendecomposition in beamspace and a proposed second-order statistics for estimating the nominal elevation and azimuth DOAs. The proposed decoupled estimation algorithm does not involve any searching. It has a lower computational complexity particularly when the radio of array size to the number of source is large, at the expense of negligible performance loss. Simulation results are included to demonstrate the performance of the proposed technique.