Abstract: The update of covariance matrix of samples is studied,and a new method to estimate direction-of-arrival(DOA) for moving sources is proposed.Making use of the maximum likelihood algorithm,this method avoids the decomposition of the covariance matrix which arises repetitiously in subspace tracking types methods.And this method effectively reduces the computational load by locking the targets,reducing the range of search for targets and appling swarm intelligence technologies.Simulation results show that the DOA estimation based on the improved particle swarm algorithm has the ability to track coherent sources and is better for precision and real-time characteristics.