粒子群算法优化相空间重构参数的网络流量预测模型  

Network Traffic Prediction Based on Phase Space Reconstruction Optimized By Particle Swarm Optimization Algorithm

作　　者：曾伟[1] 黄亮[2] 

机构地区：[1]华东交通大学信息工程学院,南昌330013 [2]上饶师范学院数学与计算机科学学院,江西上饶334001

出　　处：《计算机测量与控制》2014年第9期3014-3016,3043,共4页Computer Measurement &Control

基　　金：江西省教育厅青年基金项目(GJJ13704)

摘　　要：在网络流量预测过程中,相空间重构参数是影响预测性能的重要方面,传统参数分开优化,为了提高网络流量的预测精度,提出一种粒子群算法优化相空间重构参数的网络流量预测模型(PSO-BPNN);首先将BP神经网络作为学习算法,然后采用粒子群算法对相空间重构参数——延迟时间和嵌入维进行联合优化,并重构网络流量序列,最后以小波BP神经网络建立最优络流量预测模型,并采用仿真实验对模型性能进行分析,结果表明,PSO-BPNN提高了网络流量的预测精度。Parameters of phase space reconstruction are very important in network traffic prediction which is solved separately traditionally. In order to improve the prediction accuracy of network traffic, a novel network traffic prediction model （PSO-BPNN） is proposed in this paper based on particle swarm optimization algorithm and BP neural network. Firstly, BP neural network is taken as perdition algorithm, and the optimal delay time （τ） and embedding dimension （m） are obtained by particle swarm optimization and the network traffic series are reconstructed, finally, network traffic prediction models are established based on reconstruction the network traffic series, and simulation experiments are carried out to test the performance of network traffic prediction model. The results show that PSO-BPNN has improved the prediction accuracy of network traffic.

关 键 词：网络流量 相空间重构 粒子群算法 嵌入维 延迟时间 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]