PSO algorithm for exact Stochastic ML estimation of DOA for incoherent signals

The performance of Stochastic ML (SML) algorithm of Direction-of-Arrival (DOA) is much more superior to many other algorithms in array signal processing. However, the estimation of SML is a non-linear multi-dimensional optimization problem. Therefore its computational complexity is very high. In this paper, firstly we show exact definition of SML estimation of DOA for incoherent signals and brief description of the conventional solving method, Alternating Minimization (AM) algorithm. Then, we propose to use the Particle Swarm Optimization (PSO) algorithm to solve the estimation of SML. Also in this paper, we propose a method to optimize the inertia factor of PSO. Simulation results show that the computational complexity of the proposed PSO algorithm for SML estimation is much lower than that of the conventional AM algorithm.