Elman神经网络的网络流量预测
侯家利
【期刊名称】《计算机仿真》 【年(卷),期】2011(028)007
【摘要】研究网络流量准确预测问题,由于网络流量变化具有非线性、突变性,传统网络流量预测是建立在线性模型的基础上,无法准确描述网络流量变化规律,导致预测精度低.为了提高网络流量的预测精度,提出一种Elman神经网络的网络流量预测模型.根据Elman神经网络良好的时变性捕捉能力和非线性预测能力对网络流量变化规律进行建模和预测.实验结果表明,模型具有良好的预测效果,相对于传统ARIMA模型、BP神经网络模型,Elman神经网络模型预测精度更高,误差更小,说明了改进的优化方法对网络流量预测是有效和可行的.%Network traffic has the non-linear and dynarnlclty, the traditional .network traffic prediction based on linear model cannot describe network traffic change rule, and prediction precision is low. In order to improve the network traffic prediction precision, the paper proposes a network traffic prediction model based on Elman neural network. The model uses Elman neural network' s good ability of shock-capturing and nonlinear prediction to model and predict the network traffic. Experimental results show that the model has good prediction effect, the prediction accuracy is higher, and errors are smaller compared to traditional ARIMA model and the BP nerve network model, which illustrates that the model is feasible and effective to be applied in
Elman神经网络的网络流量预测
