Multifractal Analysis and Prediction of VBR Video Traffic

Real-time prediction of video source traffic is an important step in wireless network management tasks such as dynamic bandwidth allocation and end-to-end quality-of-service (QoS) control strategies. However, together with the slowly-decaying auto-covariance function (ACF) and the traffic non-stationarity, it suggests that some conventional prediction tools, which only use short-range dependence (SRD) feature, such as the linear auto-regressive method are not appropriate for prediction at large time scale. In this paper, we use multifractal method to analysis and predict VBR video traffic. We implement our predictor by making predictions in the multifractal domain and solving the prediction of the VBR video traffic through multifractal tree. Because the LRD feature of video traffic is used, the multi-step prediction performance of proposed method is much better than traditional linear methods