Aiming at improving the multi-target tracking performance of the nonlinear model in the unknown clutter environment, a Gaussian mixture particle probability hypothesis density(GMP-PHD) algorithm based on quasi-Monte-Carlo sampling is proposed in this paper. First of all, this algorithm can track targets steadily in the unknown clutter model by fitting spatial distribution of the unknown clutter intensity with the finite mixture model and estimating the clutter number. Then, the proposed algorithm can improve the tracking accuracy by applying the quasi-Monte-Carlo sampling which makes the best of the uniform distribution property of the low discrepancy points into the GMP-PHD filtering. Simulation results show that the proposed algorithm performs better than the conventional algorithms in the nonlinear unknown clutter environment.