Lower bound and optimal scheduling for mean user rebuffering percentage of HTTP progressive download traffic in cellular networks

This work focuses on an analytical investigation a performance of cellular networks for transmission video content. Degradation of video playback is rising, due to increasing a video traffic amount in the modern cellular networks. Stalling of a video stream has the greatest influence to user's personal contentment, which is possible to estimate by rebuffering percentage (percentage of watching time in which user experience is stalling effects). Resource allocation algorithms upgrade at a base station makes possible to decrease degradation effects for video streams. In this article, we propose a lower bound for a mean user's rebuffering percentage of HTTP Progressive Download traffic type in cellular networks in all possible schedulers and optimal scheduling strategy based on it. The performance of the proposed scheduler demonstrates by Key Performance Indicators for video streaming (QoE: Mean Opinion Score and rebuffering percentage) and network (Throughput and Fairness). The obtained results can use as a baseline for performance estimation of existing and new scheduling algorithms for video traffic transmission over cellular networks.