2D cyclic ESPRIT via Hermitian mapping

To circumvent the rank deficiency from highly correlated or coherent signals of interest (SOIs), the spatial smoothing (SS) methods are perfect solutions, but reduce the effective aperture of the array. And this directly result in a certain performance degradation. Via Hermitian mapping, a novel 2D cyclic ESPRIT algorithm without conjugate centrosymmetric array configuration is presented in this paper, which has closed-form, automatically paired 2-D angle estimates and a better performance in the presence of multipath propagation. Finally, simulation results show the effectiveness of the proposed method.