小波神经网络在NGN流量预测中的应用

Application of WNN to NGN Traffic Prediction

基于下一代网络NGN(NextGenerationNetwork)的运行环境,该文提出了一个的基于小波神经网络的IP流量预测方法。在神经网络预测模型中,神经网络中的转移函数使用小波函数来替代,从而建立小波基神经网络;同时,通过使用小波多分辨率方法将原始流量信号分解成不同频率成分的分量信号,然后使用分量信号作为训练样本训练小波基神经网络。通过前述方法建立NGN流量预测模型,并根据实际流量数据预测一天的流量。实验结果表明本方法相较未采用小波的神经网络预测方法,能显著提高流量预测精度。

A model to predict IP traffic in IP-based next generation network is introduced. By using net flow traffic collecting technology, some traffic data for the analysis have been collected from an NGN operator. To build wavelet basis Neural Network （NN）, the sigmoid function is replaced with the wavelet in NN, and wavelet multi-resolution analysis method is used to decompose the traffic signal and then the decomposed component sequences is employed to train the NN. By using the methods, an NGN traffic prediction model is built to predict one day＇s traffic. The experimental results show that the traffic prediction method of wavelet NN（WNN） is more accurate than that without using wavelt in the NGN traffic forecasting.