In order to improve the accuracy of node location in wireless sensor network,an intelligent location algorithm based on composite Particle Swarm Optimization (PSO) is proposed for DV-Hop. The problem of uncertainty location is transformed into a multi-dimension optimization problem after the process of maximum likelihood method. Moreover, satisfactory accuracy of location is obtained with the advantage of composite PSO in NP problems. The experiment results show that, proposed algorithm is excellent in decreasing location error, improving accuracy and robustness, compared with traditional methods.