网络流量的识别和预测系统关键技术研究  

Research on the Key Technologies of Network Traffic Identification and Prediction System

作　　者：袁泽勇 刘波[2] 覃宁[1] 

机构地区：[1]中讯邮电咨询设计院有限公司,四川成都610042 [2]中国移动通信集团四川有限公司,四川成都610041

出　　处：《邮电设计技术》2016年第10期82-87,共6页Designing Techniques of Posts and Telecommunications

摘　　要：近年来用户对于网络的运行和维护提出了新的要求,主要体现在网络流量识别、预测、异常流量的监测和控制。流量识别算法和网络流量模型在网络设计、服务质量、网络管理与监测中起着重要的作用。研究和验证了网络流量的独立小波模型和多重分形小波模型,并对模型参数进行估计,拟合了网络流量,研究了基于小波的网络流量预测方法,并对这种方法进行了改进,用FARIMA模型对单支重构后的序列进行预测。In recent years,the users has raised new claims for the network operation and maintenance,mainly reflected in the network traffic identification,prediction,monitoring and control of abnormal traffic.Traffic identification algorithm and network traffic model play an important role in network design,quality of service,network management and monitoring.It researches and verifies the independent wavelet model and multi fractal wavelet model of the network traffic,estimates the parameters of the models.It studies the network traffic prediction method based on wavelet,improves the method,predicts the sequence of single branch reconstruction with the FARIMA model.

关 键 词：自相似 FARIMA 小波变换 流量预测 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]