Research of a Novel Flash P2P Network Traffic Prediction Algorithm

To improve the quality of service and network performance of the FlashP2P video-on-demand, the prediction FlashP2P network traffic flow is very useful to t control the network video traffic. In this paper, a novel prediction algorithm to forecast the traffic rate of the Flash P2P video is proposed. This method is based on the combination of the local mean decomposition (LMD) and the generalized autoregressive conditional heteroscedasticity (GARCH). LMD is used to decompose the original long-related flow into the summation of the short-related flow. Then, GRACH is utilized to predict the short-related flow. The developed algorithm is tested on a university's campus network. The predicted results show that our proposed method can achieve higher accuracy than those obtained by existing algorithms, such as EMD-ARMA(Empirical Mode Decomposition and Auto-Regressive and Moving Average Model) and WNN(Wavelet Neural Network), while keeping lower computational complexity