A real-valued MUSIC algorithm with forward/backward technique

This paper presents a novel real-valued multiple signal classification (RV-MUSIC) algorithm, which tremendously reduces the computation complexity than the complex one by a factor of four. The covariance matrix of the sensor output is firstly transformed into a real-valued covariance matrix via unitary forward/backward (FB) technique. Then the MUSIC algorithm in real domain is used to estimate the direction of arrival (DOA) by utilizing the symmetry properties of array structure. The proposed algorithm can solve problem with multiple incident signals in one-or two-dimensional scenarios. Simulation results show the effectiveness of the proposed methods.