基于小波的多尺度网络流量预测模型

Multiscale Network Traffic Prediction Model Based on Wavelet

通过把AR IMA线性预测方法引入小波域内,提出一个基于多重分形小波模型的网络流量预测模型,通过对真实网络流量的仿真实验,结果表明该模型能够对网络流量进行比较精确的预测.

Scaling and multiscale behaviours, such as the long-range dependence, the self-similarity, and the multi-fractal have been commonly viewed as the most significant characteristics of the network traffic today. The capacity planning theory for network requires accurate modelling of the incoming traffic, as well as accurate predictions of its future behaviour. The wavelet-based approach is a natural way to provide multiscale prediction to applications. With introduction of the ARIMA linear prediction in the wavelet domain, a novel network traffic prediction model is presented based on the multi-fractal wavelet model, which is shown to be the accurate model of the multiscale network traffic. The simulation results with the real traffic traces show the accuracy and efficiency of the model.