计算机应用与软件2017，Vol.34Issue(10)：185-191,7.DOI:10.3969/j.issn.1000-386x.2017.10.032
基于KPCA优化IHS-RVM的小时间尺度网络流量预测模型
SMALL TIME SCALE NETWORK TRAFFIC PREDICTION MODEL BASED ON KPCA AND OPTIMIZED IHS-RVM
摘要
Abstract
Network traffic time series is high-dimensional,non-linear and time-varying.To deal with low prediction accuracy of traditional time series models,we propose a small time scale network traffic prediction model based on KPCA and optimized IHS-RVM.Firstly,phase space reconstruction of network traffic time series was carried out to explore the relation between input variants and output variants.Then,KPCA was adopted to make a feature extraction of nuclear components of network traffic samples and obtained the valid key information.On that basis,Improved HS (IHS) algorithm was used to determine the embedding dimension and nuclear parameters.Finally,RVM model with hyper parameter optimization was adopted to make the prediction of network traffic.To verify the performance of the model,actual data were collected to make a comparative analysis of performance.Results have shown that,it enjoys better performance than KPCA-IHS-ESN model,KPCA-IHS-SVM model and IHS-RVM model.关键词
小时间尺度/网络流量/改进和声搜索算法/KPCA/RVMKey words
Small time scale/Network traffic/Improved harmony search/KPCA/RVM分类
信息技术与安全科学引用本文复制引用
杨波..基于KPCA优化IHS-RVM的小时间尺度网络流量预测模型[J].计算机应用与软件,2017,34(10):185-191,7.基金项目
云南省科技支撑项目(KKSTJ201358015). （KKSTJ201358015）
