文献类型：期刊文献

中文题名：相空间重构和极限学习机的网络流量预测模型

英文题名：Network Traffic Prediction Based on Phase Space Reconstruction and ELM

作者：袁开银[1];魏彬[1]

第一作者：袁开银

机构：[1]河南财经政法大学现代教育技术中心

第一机构：河南财经政法大学

年份：2018

卷号：25

期号：11

起止页码：2087-2091

中文期刊名：控制工程

外文期刊名：Control Engineering of China

收录：CSTPCD;;北大核心:【北大核心2017】；CSCD:【CSCD_E2017_2018】；

基金：河南省科技厅科技攻关项目(142400411042)

语种：中文

中文关键词：流量预测;相空间重构;混沌特性;极限学习机

外文关键词：Network traffic prediction;phase space reconstruction;chaotic characteristic;extreme learningmachine

摘要：网络流量的预测可以有效降低网络拥塞频率,提高网络的服务质量,针对传统方法无法准确描述网络流量混沌特性的局限性,提出了相空间重构与极限学习机的网络流量预测模型（PHR-ELM）。首先通过相空间重构把网络流量变为有规律数据,然后采用极限学习机实现网络流量的准确预测,最后进行了网络流量预测的仿真测试,结果表明,PHR-ELM可以有效拟合网络流量的混沌变化特性,准确实现了网络流量变化趋势的预测,预测效果要优于传统模型,验证了PHR-ELM的有效性和优越性。
Network traffic prediction can effectively reduce the network congestion rate and improve the quality of network service, traditional methods can not describe the chaotic characteristics of the network traffic, a novel prediction model based on PHR-ELM is proposed in this study. Firstly, network traffic is analyzed and processed by chaos, and phase space reconstruction is used to reconstruct the network traffic data, and secondly, extreme learning machine is used to realize the accurate prediction of network traffic, At last, the simulation test of network traffic prediction is realized, the proposed model can effectively fit chaotic characteristics, and accurately predict the change trend of network traffic, the forecast effect is better than traditional models, the results have verified the validity and superiority.