Deterministic maximum likelihood method for direction-of-arrival estimation of strictly noncircular signals

In this paper, a noncircular deterministic maximum likelihood (NC-DML) estimator for direction-of-arrival estimation of strictly NC signals is devised. Unlike the conventional DML solution for arbitrary signals, the NC-DML exploits the NC properties of the sources by reconstructing the parameter set, significantly decreasing the number of parameters to be considered. For computing the NC-DML, we present a novel NC alternating projection (NC-AP) approach. The NC-AP solution is carried out based on an augmented virtual array structure. Moreover, it also takes the impact of the initial phase shift of the NC signals into account. Simulation results are included to illustrate the superiority of the proposed method.