Estimating direction of arrival for coherent signals by using projection subspace without source number information

Abstract

Appropriate subspace projection enhances the characteristics of the desired signal power and reduces the interference effect. This paper proposes an algorithm that uses projection subspace to estimate the direction of arrival (DOA) for coherent signals without signal source number information. First, the joint diagonalization structure of a set of Toeplitz matrices constructed from the covariance matrix of the received data is used to estimate the DOA of coherent signal sources and determine the number of signal sources. Next, the projection weights of steering vectors on signal subspaces were replaced with their revised steering vectors, and the spatial spectrum function was reconstructed using both noise and signal subspaces. The spatial spectral peaks were scanned to estimate the angle of arrival of the signals. High-resolution DOA estimations were thus obtained. Finally, computer simulations illustrated the performance of the proposed method and validated the designed procedure.