2-D DOA estimation based on an improved PM algorithm with three parallel uniform linear arrays

When parameter P is estimated with conventional propagator method(PM) algorithm for 2-D direction of arrival(DOA） estimation in three parallel uniform linear arrays，there is lots of redundancy in covariance matrix constructed by divided sub-arrays.In order to reduce computational effort，we can eliminate the redundant data in covariance matrix by merging subarrays and estimate parameter P with reconstructed covariance matrix.Then a new parameter P is formed to get an overdetermined nonlinear equations set about the 2D-DOA，and the equations set is dealt with by the nonlinear least square algorithm to gain the elevation angle and the azimuth angle.Simulation results show that the proposed method reduces computational effort and improvs the precision.