摘要
网络流量是一种高度自相关、非线性时间序列数据,传统预测方法都是基于线性模型,无法反映网络流量的非线性变化规很,导致预测精度不高。为了提高网络流量的预测精度,在分析网络流量特征的基础上,提出一种基于相空间重构的支持向量机网络流量预测模型。首先利用相空间重构对网络流量原始数据进行重构,捕捉原始数据的多样性,然后将重构的数据利用非线性预测性能优异的支持向量对其进行学习和训练,建立网络流量的最优预测模型。最后以具体的网络流量数据对该模型进行了仿真,仿真结果表明,支持向量机模型比其它网络流量预测模型有较高的预测精度,因此支持向量机模型对于网络流量非线性问题的预测,具有较高的泛化能力和预测精度。
Network traffic is a time series data of highly autocorrelation,nonlinear.However,traditional forecasting methods are based on linear models which can not reflect the non-linear changes in network traffic,and the forecast precision is not high.In order to improve the forecast accuracy of network traffic,based on the analysis of network traffic characteristics,a support vector machine network traffic prediction model is proposed based on re-analysis of phase space.In this model,phase space reconstruction of the network traffic is used to reconstruct the original data and capture the diversity from the raw data,and then the reconstructed data are trained using nonlinear prediction performance of support vector to build optimal network flow forecasting model.The specific network traffic data are used carry out the simulation experiments,and the simulation results show that the support vector machine model has higher prediction accuracy than other network traffic prediction model,So this model can deal with the problems of nonlinear network traffic prediction with higher generalization ability and prediction accuracy.
出处
《计算机仿真》
CSCD
北大核心
2011年第5期159-162,共4页
Computer Simulation
关键词
网络流量
支持向量机
时间序列
预测
Traffic prediction
Support vector machine（SVM）
Time serie
Prediction
