基于时序EEMD网络流量预测方法

Prediction Method of Network Traffic Based on Timing EEMD

针对网络资源的分配和网络故障识别的需要,研究网络流量预测模型。为了提高网络流量预测的精度,提出一种基于时序EEMD的网络流量预测方法。首先采用平稳的网络流量原始数据,用改进的经验模式分解算法进行分解,得到若干IMF分量,然后利用短相关模型对各分量建模并合成每个模型生成的流量,得到完整的流量,建立网络流量最优预测模型。最后以真实的实测流量数据对该模型进行仿真。仿真结果表明,基于时序EEMD模型的网络流量预测精度高于其它预测模型,且在非持续性强波动的流量预测中具有较好效果。

Aiming at the need of network resource allocation and network fault identification,this paper studies the network traffic prediction model. In order to improve the accuracy of network traffic prediction,a network traffic prediction method based on timing EEMD is proposed. Firstly,the original data of the network flow are decomposed by the improved empirical mode decomposition algorithm. Some IMF components are obtained. Then,the short correlation model is used to model and synthesize the flow generated by each model to obtain the complete flow. The optimal prediction model of network traffic is established. Finally,the simulation is carried out with the real measured flow data. The simulation results show that the prediction accuracy of network traffic based on time series EEMD model is higher than that of other prediction models,and it has a good effect in the non-persistent strong fluctuation.