中文摘要：

本文提出了两种新的算法.第一,拟蒙特卡罗概率假设密度（QMC-PHD）滤波,主要思想是利用QMC方法来实现PHD滤波.在仿真实验中可以发现：在目标数目和状态估计方面,新算法比序贯蒙特卡罗概率假设密度（SMC-PHD）滤波器更精确.第二,卷积核拟蒙特卡罗概率假设密度滤波（CKQMC-PHD）,主要思想是基于QMC-PHD滤波的基础之上引入卷积核（CK）的估计算法.当观测噪声变小的时候,CKQMC-PHD滤波还能够很好地估计出目标状态和目标数目,其表现要明显的好于QMC-PHD滤波.仿真实验也证明了CKQMC-PHD滤波的估计效果.

英文摘要：

In this paper,we propose two algorithms.The first one is the quasi-Monte Carlo probability hypothesis density filter（PHD）.Quasi-Monte Carlo is used to implement the PHD filter.In the simulation,we can find that the proposed algorithm is more accurate than the sequential Monte Carlo PHD filter in the estimation of target state and the number of targets.The second algorithm is convolution kernel quasi-Monte Carlo PHD（CKQMC-PHD） filter.The convolution kernel algorithm is used in the QMC-PHD filter.When the observation noise becomes small,CKQMC-PHD filter is very efficient to estimate the state and the number of targets,the behave of the CKQMC-PHD filter is better than that of QMC-PHD filter.The simulation results prove the effect of the CKQMC-PHD filter.