Variable-bit-rate(VBR) compressed video can exhibit significant multiple-time-scale bitrate variability, that is why the transmission of compressed video has many complex problems. Bandwidth smoothing techniques can reduce this burstness by prefetching data into the client playback buffer at a series of fixed rates, simplifying the allocation of resources in video servers and the communication network.
In contrast to stored video, live video applications typically have limited knowledge of frame sizes and often require deadlines on the delay between the source and the clients. In this paper we propose a online bandwidth smoothing algorithm for a growing number of streaming video applications such as newscasts, sportscasts and distance learning, where many clients may be willing to tolarate a playback delay of a few seconds in exchange for a smaller bandwidth requirement. Extending techniques for smoothing stored video, we developed online, window-based smoothing algorithm. This algorithm fully utilizes client buffer by keeping previous transmission rate as long as possible at the overlaying parts between smoothing windows, and using min-max strategy, its number of transmission rate change is kept very small.
Simulation with actual MPEG-1 frame traces is employed to evaluate performance of the proposed algorithm. The simulation results show that peak transmission rate and number of transmission rate are significantly reduced, closely approximating the performance of the optimal offline algorithm with relatively small window size(2 - 10 sec), and the algorithm performs better with increase of the client buffer size.
