基于RBF神经网络的高斯混合近似算法

Journal of Systems Engineering and Electronics ›› 2009, Vol. 31 ›› Issue (10): 2489-2491,2526.

基于RBF神经网络的高斯混合近似算法

樊国创¹, 戴亚平¹, 闫宁²

1. 北京理工大学自动控制系, 北京, 100081;
2. 北京政法职业学院应用法律二系, 北京, 102600

收稿日期:2008-04-11 修回日期:2008-07-01 出版日期:2009-10-20 发布日期:2010-01-03
作者简介:樊国创(1977- ),男,博士研究生,主要研究方向为目标跟踪、数据融合.E-mail:gcfred@gmail.com
基金资助:
中国学位与研究生教育学会“十一五”研究项目资助课题

Gaussian mixture approximation algorithm based on radius basis function neural network

FAN Guo-chuang¹, DAI Ya-ping¹, YAN Ning²

1. Dept. of Automation Control, Beijing Inst. of Technology, Beijing 100081, China;
2. Second Dept. of Applied Law, Beijing Management Coll. of Politics and Law, Beijing 102600, China

Received:2008-04-11 Revised:2008-07-01 Online:2009-10-20 Published:2010-01-03

摘要/Abstract

摘要： 在分析RBF神经网络基本结构的基础上,提出一种基于RBF神经网络求解非高斯概率密度近似为高斯概率密度和的方法.该方法通过选取高斯函数作为神经网络的径向基函数,提取训练好的网络参数,运用这些参数构建混合成分的函数模型.理论分析与仿真证明,与传统采用EM近似算法相比,该算法具有求解跟初值的选取无关、能避免发散、收敛快的特点.

Abstract: A algorithm based on radius basis function(RBF) neural network is presented,in which any nonlinear function can be approximated as a limited Gauss function mixture,on the basis of analysing the structure of RBF neural network.The Gauss function is selected as a radius basis function in the proposed algrithom,and the network parameters to have been trained are drawn and are used to build a mixture function.The results of theoretical analysis and simulation verify that the proposed algorithm is independent of initial values and is convergent rapidly compared with the traditional EM(expectation maximum) algorithm.

中图分类号:

TP274

樊国创, 戴亚平, 闫宁. 基于RBF神经网络的高斯混合近似算法[J]. Journal of Systems Engineering and Electronics, 2009, 31(10): 2489-2491,2526.

FAN Guo-chuang, DAI Ya-ping, YAN Ning. Gaussian mixture approximation algorithm based on radius basis function neural network[J]. Journal of Systems Engineering and Electronics, 2009, 31(10): 2489-2491,2526.

[1]	李佳炜, 江晶, 刘重阳, 吴卫华. 多星载光学传感器系统误差极大似然配准算法[J]. 系统工程与电子技术, 2020, 42(1): 1-9.
[2]	白剑林, 王煜. 一种解决D-S理论证据冲突的有效方法[J]. Journal of Systems Engineering and Electronics, 2009, 31(9): 2106-2109.
[3]	万树平. Vague集在多传感器目标识别中的应用[J]. Journal of Systems Engineering and Electronics, 2009, 31(9): 2067-2070.
[4]	李建文, 郝顺义, 黄国荣. 基于UD分解的改进型强跟踪滤波器[J]. Journal of Systems Engineering and Electronics, 2009, 31(8): 1953-1957.
[5]	万树平. 基于熵权的多传感器目标识别方法[J]. Journal of Systems Engineering and Electronics, 2009, 31(3): 500-502,510.
[6]	王刚, 乔纯捷, 王跃科. 二次正交解调算法及内建自测试的实现[J]. Journal of Systems Engineering and Electronics, 2009, 31(3): 527-530.
[7]	王建文, 税海涛, 马宏绪, 李迅, 刘述田. 优化卡尔曼滤波算法中的目标函数选择[J]. Journal of Systems Engineering and Electronics, 2009, 31(01): 200-203.
[8]	王建文, 税海涛, 马宏绪, 李迅, 刘述田. 优化卡尔曼滤波算法中的目标函数选择[J]. Journal of Systems Engineering and Electronics, 2009, 31(1): 200-203.