DOA estimation based on fourth-order cumulants with unknown mutual coupling

Most of the existing direction of arrival (DOA) estimation algorithms with mutual coupling are generally based on the second-order statistics while assuming that the noise is additive white Gaussian noise. In this paper, a method based on fourth-order cumulants (FOC) is proposed for DOA estimation with uniform linear array (ULA) in the presence of unknown mutual coupling. The method can be applied when non-Gaussian signals coexist with unknown colored Gaussian noise. The unknown mutual coupling can be blindly compensated by the inherent mechanism of the proposed method and the DOA of signals can be accurately estimated without any calibration sources. Since we use the FOC, the Gaussian noise is greatly suppressed and we can obtain a good estimation performance in low SNR condition. Simulation results demonstrate the effectiveness of our method. (C) 2009 Elsevier B.V. All rights reserved.