A multi-objective particle swarm optimization algorithm based on data association (DS-MOPSO) is introduced. The Gauss distribution is adopted to ensure the uniform distribution of initial population. The crowding distance and the prior probability are used to calculate the crowing degree of non-dominated solutions in the external archive. The elite particles, which maintain the diversity of solution, are selected by the Sigma value. The space joint probabilistic data association is introduced to generate dynamically the inertia weight of each particle and expand the search area in order to avoid falling into local optimum. Simulation results show that the Pareto solutions obtained by that DS-MOPSO has a good convergence and diversity.