Direction Finding for Bistatic Mimo Radar Using Em Maximum Likelihood Algorithm

In this paper, we investigate an expectation-maximization (EM) maximum likelihood (ML) algorithm of direction finding (DF) for bistatic multiple-input multiple-output (MIMO) radar, where it is shown that the DF problem can be described as a special case of ML estimation with incomplete data. First, we introduce the signal and the noise models, and derive the ML estimations of the direction parameters. Considering the computational complexity, we make use of the EM algorithm to compute the ML algorithm, referred to EM ML algorithm, which can be applied to the arbitrary antenna geometry and realize the auto-pairing between direction-of-departures (DODs) and direction-of-arrivals (DOAs). Then the initialization is considered. In addition, both the convergence and the Cramer-Rao bound (CRB) analysis are derived. Finally, simulation results demonstrate the potential and asymptotic efficiency of this approach for MIMO radar systems.