Adaptive Online/Offline Smoothing for Streaming Videos over Best-Effort Networks
Chia-Hu Chang, Chu‐Chuan Lee, Pao‐Chi Chang · 2005
The real-time video traffic often exhibits significant burstiness due to natural variations within and between scenes. Unfortunately, the current best-effort Internet cannot offer sufficient quality of service (QoS) guarantees to streaming videos, which degrades the received picture quality significantly. Therefore, an end-to-end strategy is required for delivering video data over Internet. The traffic smoothing technique is one of the approaches for handling fluctuations in bandwidth demands of video traffic. This paper proposes an end-to-end adaptive video streaming mechanism by integrating the TCP-Friendly Rate Control (TFRC) with the online/offline video traffic smoothing algorithms. Using TFRC, the proposed framework can dynamically estimate the current available network bandwidth and then adaptively determine the transmission schedule for a video stream. The adaptation of transmission schedule is based on the current network condition, the available server and client buffer spaces, and the characteristics of video frames. Simulation results show that the proposed framework can effectively reduce the packet loss rate that is resulted from variations in delay and available bandwidth. No packet is timeout in all tested cases when the proposed mechanism is used. More importantly, the proposed system can be applied to both offline and online smoothing cases. I.