Estimation of two-dimensional direction of arrival under a distributed signal source model

Abstract

Direction of arrival (DOA) estimation is an important issue in array signal processing. Most research on the estimation of DOA has been performed based on the assumption that the signal sources are point sources. It is a reasonable assumption if the signal sources are not distributed and located far enough from the receivers. Under this assumption, array output vector would be a weighted sum of the finite number of steering vectors, with the weighting dependent on the signal sources. In real surroundings, on the other hand, the signals received at an array usually consist not only of a direct path signal, but also of multiple return signals that are coherent, phase-delayed, and amplitude-weighted replicas of the direct path signal. The array output vector is obtained by integrating a steering vector over all direction of arrival with the weighting of a distributed source density function, if the signal sources are distributed, i.e., if the point source assumption is not valid. In this dissertation, we consider a two-dimensional distributed signal source model. To obtain specific and concrete results, we consider a class of the distributed sources, parametric signal sources. In the parametric model, a source is represented by two parameters, the center angle and degree of dispersion. Then, we address the estimation of the elevation and azimuth angles of distributed sources based on the parametric source modeling in the three-dimensional space with two uniform linear arrays and a uniform circular array. We also develop an efficient estimation procedures that can reduce the computational complexity. Some examples are shown to more explicitly see the estimation procedures under the parametric model: results are obtained from computer simulations performed by using MUSIC-based methods.