改进的基于时变FARIMA模型和小波变换的网络流量预测算法

网络的流量特性是反映网络实时状态的一个重要特征,对于网络流量的分析、预测一直是该领域的研究热点。传统的基于时间序列模型的方法在计算效率和多尺度分析能力方面存在一定的局限性。本文提出了一种改进的基于小波变换和时变FARIMA模型的流量预测方法,利用小波变换的多尺度分析特性将原有的流量数据进行分解,在使用时变FARIMA模型进行预测,可大大提高算法的执行效率和预测的准确性。最后,本文选取了Bellcore提供的真实的网络流量进行了仿真实验,验证了提出的预测方法的准确性和有效性。

Network traffic prediction based on time-varying FARIMA and wavelet trancformation

Flow characteristics of the network is a reflection of the real-time status of the network is an impor-tant feature for network traffic analysis, forecasting has been a hot topic in the field. Some limitations of traditional methods of time series model based on the presence in the computing efficiency and multi-scale analysis capability. This paper presents an improved wavelet-based and time-varying FARIMA model flow forecasting methods, the use of multi-scale wavelet transform to analyze the characteristics of the original traffic data decomposition, when using varying FARIMA model to predict, can greatly improve the algorithm the efficiency and accuracy of prediction. Fi-nally, select the Bellcore provides real network traffic simulation experiments carried out to verify the accuracy and effectiveness of the proposed prediction method.