Dynamic window-based traffic-smoothing for optimal delivery of online VBR media streams

Traffic-smoothing for delivery of online VBR media is one of the most important problems when designing streaming multimedia applications. Given the available client buffer b and playback delay D, Rexford et al. (1997) introduced a window-based approach called SLWIN(k) to smooth online generated traffic for the pre-specified window size W (W<D) and sliding distance k (k⩽W). The required time complexity is O(n*W/k) where n is the number of media frames. Note that, by using the maximum sliding distance W, SLWIN(W) has the minimum computation cost O(n). However, the required peak bandwidth is large. Although SLWIN(I) with the minimum sliding distance 1 can achieve small peak bandwidth, it requires O(n*W) computation cost. As there is a tradeoff between the required peak bandwidth and computation cost, it is hard to decide the best sliding distance k for the conventional static window-sliding SLWLN(R) method. In this paper, a novel dynamic window-sliding scheme is proposed. Our approach can dynamically adjust the sliding distance of window to minimize both the computation cost and the peak bandwidth for online traffic-smoothing. Given the same client buffer, playback delay and window size, our allocated peak bandwidth can be the same as that achieved by SLWIN(I). Besides, our computation cost is the same as that required for SLWIN(W)