基于大数据分析的非线性网络流量组合预测模型  

Combined prediction model for nonlinear network flow based on big data analysis

作　　者：许绘香 曹敏 马莹莹 XU Hui-xiang;CAO Min;MA Ying-ying(College of Information Engineering,Zhengzhou Institute of Technology,Zhengzhou 450044,China)

机构地区：[1]郑州工程技术学院信息工程学院,郑州450044

出　　处：《沈阳工业大学学报》2020年第6期670-675,共6页Journal of Shenyang University of Technology

基　　金：河南省科技攻关项目(172102210606);河南省高等学校重点科研项目(17B520040).

摘　　要：针对传统方法不能对网络流量变化特征进行准确描述,并且预测精度较低的问题,提出了基于大数据分析的非线性网络流量组合预测模型.通过对非线性网络流量数据进行有效分解,获得不同尺度的分量,利用混沌理论对多尺度分量进行相空间重构获得流量子序列.构建改进鸟群算法优化模型,并对重构后的网络流量子序列进行预测和组合,获得网络流量预测结果.结果表明,所提模型能够精确地描述网络流量的非线性、周期性以及长相关性等变化特征,具有较高的预测精度.In order to solve the problem that traditional methods can not accurately describe the changing characteristics of network flow and have low prediction accuracy,a combined prediction model for nonlinear network flow based on large data analysis was proposed.Through the effective decomposition of nonlinear network flow data,the components with different scales were obtained.The flow subsequences were obtained through the phase-space reconstruction of components with different scales in terms of chaos theory.An improved bird swarm optimization model was constructed,the reconstructed network flow subsequence was predicted and combined,and the results of prediction model for network flow were attained.The results show that the as-proposed model can accurately describe the non-linear,periodic and long-term correlation characteristics of network flow,and has higher prediction accuracy.

关 键 词：大数据分析 非线性网络 网络流量 组合预测模型 改进鸟群算法 混沌理论 觅食行为 周期性 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]