Robust expectation-maximization algorithm for multiple wide-band acoustic source localization in the presence of non-uniform noise variances

Wideband source localization using acoustic sensor networks has been drawing a lot of research interest recently. The maximum-likelihood is the predominant objective which leads to a variety of source localization approaches. In this paper, we would like to combat the source localization problem based on the realistic assumption where the sources are corrupted by the noises with non-uniform spatial variances. We study the respective limitations of two popular source localization methods for solving this problem, namely the SC-ML and AC-ML algorithms and design a new expectation maximization (EM) algorithm. Through Monte Carlo simulations, we demonstrate that our proposed EM algorithm outperforms the SC-ML and AC-ML methods in terms of the localization accuracy.