Distributed expectation-maximization algorithm for DOA estimation in wireless sensor networks

In this paper, we study the problem of using distributed expectation-maximization (EM) for direction-of-arrival (DOA) estimation over wireless sensor networks, where each node is equipped with multiple sensor elements. It is a diffusion adaptation strategy based EM algorithm, which is applicable for practical scenarios where each node is equipped with different number of sensors and the signal-to-noise ratio (SNR) level is different for different nodes. Furthermore, the computational and storage capabilities of different nodes could be different. Performance of the proposed method is evaluated using computer simulations by considering far-field narrow-band source signals.