相空间重构和支持向量机相融合的网络流量预测

Network traffic prediction model based on phase space reconstruction and support vector regression

针对网络流量非线性、突变性和混沌性特点,利用相空间重构和支持向量机参数的天然联系,提出一种相空间重构和支持向量机相融合的网络流量预测方法。将网络流量预测精度作为建模目标,采用粒子群算法对空间重构和支持向量机参数进行组合优化,建立最优网络流量预测模型。仿真实验结果表明,相对于传统网络流量预测方法,该方法更加能够刻画网络流量复杂的变化特点,有效提高了网络流量的预测精度。

The network traffic has nonlinear, mutation and chaos characteristics, so this paper puts forward a network traffic prediction method based on phase space reconstruction and support vector machine using the relation between phase space reconstruction and support vector machine parameters. The network traffic prediction accuracy is taken as the modeling object, and particle swarm optimization algorithm is used to optimize for spatial reconstruction and parameters of support vector machine to establish the optimal network traffic prediction model. The simulation results show that this proposed method can describe network traffic characteristics of complex changes better compared with the traditional network traffic forecast method, and it effectively improve the prediction accuracy of network traffic.