H∞-optimal traffic predictor for VBR video

An H-infinity-optimal video traffic predicting algorithm is proposed, which not only minimizes the worst possible effect in the estimation error but also does not require a priori knowledge of the noise source statistics. The proposed method has been applied to modeling and predicting MPEG4 variable bit rate (VBR) video traffic. The low computational complexity and good prediction performance of the H-infinity predictor are encouraging and reflect on the practicability of the proposed approach. Simulation results show that the proposed model is more robust compared with the Kalman predicting model and is able to predict the original traffic more accurately, thus more suitable for real time video traffic prediction and resource reservation.