Bandwidth Prediction for Real Time Video Delivery

 To guarantee quality of service (QoS), the requirements for video transmission such as delay and cell loss rate (CLR) are very stringent. These constraints are difficult to meet if high network utilization is desired. Dynamic bandwidth allocation is thus needed. Video traffic prediction can play an important role in dynamic bandwidth allocation and traffic management in high-speed networks. In this paper, we propose a dynamic bandwidth allocation algorithm based on the predicted relative size change of I frames by employing normalized least mean square predictor. Compared with predicting composite MPEG video traffic levels, the proposed scheme has considerable prediction accuracy. The performance of this scheme is studied using Renegotiated Constant Service Model (RCBR). Compared with CBR service, network utilization can be improved significantly for the same CLR.