Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (22): 80-82.DOI: 10.3778/j.issn.1002-8331.2010.22.025

• 网络、通信、安全 • Previous Articles     Next Articles

Improved neural network model to predict network reliability

WANG Li   

  1. Department of Physics & Electronics Engineering,Jinzhong College,Jinzhong,Shanxi 030600,China
  • Received:2009-03-23 Revised:2009-05-14 Online:2010-08-01 Published:2010-08-01
  • Contact: WANG Li

改进神经网络模型对通信网可靠性的预测

王 丽   

  1. 晋中学院 物理与电子工程学院,山西 晋中 030600
  • 通讯作者: 王 丽

Abstract: An improved PSO algorithm is used to carry on the optimization to the BP neural network weight.This paper takes the LAN/WLAN integration network as background,LAN/WLAN integration network reliability prediction models are established with three methods(BP neural network, BP neural network based on modified PSO,SVM).Compared with the experimental,the validity and superiority of modified PSO-based neural network prediction model have proven.

摘要: 首先利用一种改进后的粒子群算法对BP神经网络权值的选取进行优化,然后以LAN/WLAN集成网络为背景,用三种方法(BP神经网络、改进PSO算法优化后的BP神经网络、SVM)建立了LAN/WLAN集成网络可靠性的预测模型,最后通过实验比较,证明了改进后的神经网络模型预测通信网的可靠性、有效性和优越性。

CLC Number: