基于混沌特性的网络流量预测

Prediction of Network Traffic Flow Based on Chaos Characteristics

高速网络中存在着以自相似为特征的多种业务流量,这种自相似特征和混沌现象的吸引子有着紧密的联系。本文基于混沌时间序列重构相空间理论,根据最大Lyapunov指数,分别采用W o lf原始算法和改进算法,对高速网络中自相似信源的速率进行了预测,并给出了最大可预报时间。仿真结果表明,W o lf改进算法预测精度及可靠性更高。

There are many sorts of the traffic flow of self-similarity characteristics in the high-speed network. This self-similarity keeps in close contact with the attractor of the chaos system based on the theory of phase space reconstruction about chaotic time series and the largest Lyapunov exponents. A predictive rate of self-similar traffic sources is predicated in the high-speed network as well as the maximum predictable time by using Wolf scheme and its improved algorithm. The simulation result shows that the improved Wolf scheme has higher accuracy and reliability.