Deterministic maximum likelihood approach for localization of near-field sources

Abstract

In this paper we proposed deterministic maximum likelihood approach for estimating the direction of arrival and range parameters of the near-field sources. Direct maximum likelihood estimation of near-field source parameters results in complicated multi-parameter optimization problems, we therefore reformulated the estimation problem in terms of actual-data sample, called the incomplete data and a hypothetical data set, called the complete data and then devised the Expectation/Maximization iterative method for obtaining maximum likelihood estimates. The Expectation/Maximization algorithm decomposes the observed data into Its components and then estimates the parameters of each signal component separately providing computationally efficient solution to the resulting optimization problem. The applicability and effectiveness of the proposed algorithm is illustrated by some numerical simulations.