Real-Valued Extended Propagator Method for Estimating DOAs of Noncircular Signals with Polynomial Rooting

A computationally efficient direction-of-arrival (DOA) estimation algorithm for noncircular signals is proposed in this paper. Based on the real-valued extended propagator method and polynomial rooting technique, the algorithm is computationally efficient. Firstly, the real-valued array extension matrix and its covariance matrix are constructed by utilizing the noncircularity of the signals. Secondly, the noise subspace is achieved with extended propagator method without eigendecomposition. Thirdly, the estimated DOAs are obtained by the polynomial rooting method applying to uniform linear arrays. The simulation results indicate that the performance of the algorithm is close to those of Euler-root-MUSIC and NC-root-MUSIC. The analysis of the complexity shows that the algorithm is computationally efficient in comparison to the above mentioned algorithms. The proposed algorithm is of practical value for its good performance and computational efficiency.