作　　者： ; ;

机构地区： 深圳大学信息工程学院ATR国防科技重点实验室

出　　处： 《信号处理》 2013年第10期1323-1328,共6页

摘　　要： 粒子滤波(PF)技术的研究一直是非线性滤波领域的热点和难点问题,针对非均匀稀疏采样环境下传感器观测的滤波估计问题,提出了一种结合目标运动特性的改进型高斯粒子滤波方法。在该方法中,首先深入分析了传统粒子滤波不能有效对非均匀稀疏采样观测数据进行有效处理的原因,通过引入目标观测、目标观测的有效时间间隔、目标速度等目标特性,综合改善高斯粒子滤波器在时间更新阶段预测粒子和预测协方差估计的准确性,从而提高观测更新阶段重要性密度函数的估计精度,实现对目标状态的精确估计。实验结果表明,对于一维非线性非高斯例子,提出方法要稍好于传统的PF、辅助粒子滤波(APF)和高斯粒子滤波(GPF);而对于实际的非均匀稀疏采样观测样本,提出方法要远好于PF、APF和GPF,能够有效对目标进行状态估计。 The study of the particle filter （PF） has been the hot and difficult problems in the field of nonlinear filtering.To the state estimate of target in Aperiodic Sparseness Sampling Environment,a modified Gaussian particle filtering method based on target motion characteristic is proposed.Firstly,the reasons of the traditional particle filter that can not effectively deal with Aperiodic Sparseness Sampling measurements were analyzed.Secondly,in order to improve the estimated accuracy of the predicted particles and covariance of Gaussian particle filter,the proposed algorithm incorporate target observation,time interval of the target observation and the target speed into the construction of important density function of Gaussian particle filter.Finally,the experimental results show that the performance of the proposed algorithm is slightly better than these of the traditional PF,auxiliary particle filtering（APF） and Gaussian particle filter （GPF） for one-dimensional nonlinear non-Gaussian; but it also shows that the performance of the proposed algorithm is much better than these of the PF,APF and GPF for aperiodic sparseness sampling environment.

关 键 词： 非均匀稀疏采样 高斯粒子滤波 目标运动特性

领　　域： [电子电信] [电子电信]