A low-complexity reduced dimension propagator method for DOA estimation of non-circular signals

Non-circular (NC) signals represent an important class of signals in practical applications, and accurate direction of arrival (DOA) estimation of such signals is of significant importance. In this paper, a low-complexity reduced dimension propagator method (RD-PM) is proposed, which substantially reduces computational complexity by avoiding the two-dimensional spectral search while maintaining estimation accuracy. Specifically, the proposed algorithm first exploits the NC property to extended the received signal. Then, the extended covariance matrix is calculated and partitioned into submatrices. Subsequently, the propagator operator is derived, followed by the formulation of a one-dimensional (1D) spectral search function. Finally, a 1D search is performed to obtain the DOA estimates. Simulation results validate that the proposed algorithm achieves high estimation accuracy while maintaining low computational complexity.