Prediction Model of VBR Video Traffic Based on Wavelet Support Vector Machine

Because VBR video traffic has time-varying and nonlinear,a prediction model of VBR video traffic based on wavelet support vector machine( WSVM) is presented. Firstly,VBR video traffic time series are reconstructed by the phase space reconstruction,and then the reconstructed VBR video traffic time series are input into WSVM to learn and establish the prediction model of VBR video traffic,finally the simulation experiment is carried out to test the performance of the model. The simulation results show that,compared with other models,WSVM has improved the prediction accuracy of VBR video traffic,and it can more accurately describe the complex change rules of VBR video traffic.