Bitrate adaptation for mobile video streaming based on buffer and channel state

Mobile video streaming is a rising business but poses significant challenges for network and service operators. In mobile networks, time-variant wireless channels often violate the rate requirement of the content, which leads to a stalled video stream. Common HTTP Adaptive Streaming techniques cannot effectively react to channel variation by adapting the video bitrate only to the state of the playback buffer. In this paper, we present an optimal solution that adapts to the current and predicted channel state as well. Our optimal formulation is based on Markov Decision Processes and studied with real content and channel traces from vehicular users. The results clearly show the large benefit of complementing buffer-aware by channel-aware adaptation and prediction. Finally, we present a heuristic that closely achieves optimal service quality at practical computational complexity.