location: Current position: Home >> Scientific Research >> Paper Publications

一种粒子滤波预处理方法

Hits:

Indexed by:期刊论文

Date of Publication:2012-07-08

Journal:系统仿真学报

Included Journals:Scopus、PKU、ISTIC、CSCD

Volume:24

Issue:7

Page Number:1470-1473

ISSN No.:1004-731X

Key Words:粒子滤波;预处理;权值方差;粒子多样性;实时性

Abstract:粒子滤波(Particle Filter)是一种基于蒙特卡罗(Monte Carlo)的贝叶斯滤波方法,通常的SIR方法存在严重的粒子匮乏现象。用大权值粒子和小权值粒子的组合来取代小权值粒子,可以减小粒子权值方差,增加粒子多样性。仿真结果表明,在状态估计的初期,使得粒子迅速靠近高似然区域,精度得到了大幅度的提高。同时,算法的实时性也有很好的改善,适用于观测噪声和状态噪声较小的情况。

Pre One:Novel Wireless Positioning System for OFDM-Based Cellular Networks

Next One:Robust tracking algorithm for wireless sensor networks based on improved particle filter