Joint sources number detection and DOA estimation under the coexistence of uncorrelated and coherent signals

A joint estimation algorithm of sources number and direction of arrival(DOA) based on matrix reconstruction is proposed for the coexistence of uncorrelated sources and coherent sources. Firstly, we uses the second order statistic of the eigenvalues(SORTE) method and the estimated signal parameter via rotational invariance techniques(ESPRIT) to achieve the uncorrelated number and the estimation of the number of sources in the algorithm. Secondly, the uncorrelated signals are eliminated and a new matrix is constructed based on the spatial difference method. Next, the forward spatial smoothing can be used to realize the covariance matrix recovery. Finally, the number of coherent signals is estimated using the SORTE method, and the DOAs of coherent signals are estimated by root multiple signal classification(root-MUSIC) algorithm. Compared with the previous works, the proposed method can resolve more signals than the array elements with good performance. Moreover, the DOA estimation accuracy of coherent signals can be further improved by the new matrix. Simulation results demonstrate the effectiveness of the proposed method.