Abstract: Study of network traffic prediction, network traffic has a sudden, periodic, nonlinear characteristics, the traditional network traffic prediction model unable to establish accurate prediction model, lead to the prediction error, low accuracy of prediction. In order to improve the prediction accuracy of network traffic, put forward a kind of wavelet decomposition and support vector machine network traffic prediction model. Wavelet transform is adopted to network decomposition, the network traffic characteristics of components separated, then by using support vector machine to each component are predicted, the wavelet transform is adopted to predict the results of reconstruction of the components, network traffic has been the final forecasting result. The simulation results show that, compared with other prediction models, improve the prediction accuracy of network traffic, for the prediction of network traffic flow optimization provides reliable basis.