Source-rate control is an important issue in video streaming applications, particularly for wireless networks where channel resources are often shared among a variable number of stations using a contention-based access mechanism. In this context, the resulting throughput available for the video server has been demonstrated to be bursty, which is a feature that makes high-quality video streaming quite difficult. On the basis of this observation, we propose a rate control algorithm that works adjusting the rate on a per-window basis to compensate low-throughput periods with high-throughput periods so as to avoid the "saw" effect that is typically observed in frame-based rate control. The time axis is divided into windows of fixed size and rate changes are introduced only at the beginning of each window with the aim of keeping the probability of playback buffer starvation lower than a desired threshold during the entire current window. To achieve this objective, the algorithm makes use of a short-term prediction of the network delay using historical data. Simulations proved the efficiency of algorithm when controlling the starvation probability while avoiding the introduction of sudden changes in the source rate.
This paper addresses the problem of source-rate control in video streaming over wireless networks. The wireless channel resources are assumed to be shared among a variable number of stations using a contention-based access mechanism. In this context, the resulting throughput available for the video server has been proved to be bursty, which is a feature that makes the high-quality video streaming quite difficult. To face the channel burstiness, we propose a solution that focuses on the fluidity of the video playback by controlling the occupancy of the playback buffer so as not to exceed a desired rate of buffer starvation occurrences. While this constraint has been guaranteed, the encoder maximizes the source bit rate. Rate control is performed adaptively on the basis of a per-interval approach, which has the advantage to reduce the fluctuations in the source bit rate, ensuring smooth variations in video quality and avoiding the "saw" effect that is typically observed in frame-based rate control. Because of its low computational complexity, the proposed scheme is attractive for real-time operation. Simulations are carried out to assess the performance of the proposed scheme.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.