To improve the network traffic prediction result, the paper put forward a nonlinear combination prediction model of network traffic flow. First, the single models of ARIMA, ARMA, GM (1,1) were used in the prediction of characteristics of network traffic, Then three predicting results were input to a support vector machine for data fusion to obtain the final forecasting result. The simulation results show that, compared with the other network traffic prediction models, the nonlinear combination forecasting model can well reflect the complex changes in network traffic, thereby improves the prediction accuracy of network traffic.%研究网络优化管理问题,网络流量具有趋势性、周期性和非线性,传统预测模型只能反映网络流量变化的片段信息,难以获得高精度的网络流量预测结果.为提高网络流量预测结果,提出一种非线性组合的网络流量预测模型.首先采用单一模型ARIMA、ARMA、GM(1,1)对网络流量各特征进行预测,然后将三种预测结果输入到支持向量机进行融合,得到网络流量的最终预测结果.仿真结果表明,与传统网络流量预测模型相比,改进的非线性组合预测模型能够较好的反映网络流量的复杂变化规律,从而提高了网络流量的预测精度.
展开▼