A stochastic optimization-based bitrate adaptation for QoE maximization in wireless video streaming

Emerging wireless access techniques and ubiquitous mobile terminals enable user generated video streaming content to be shared and consumed in a novel paradigm. However, the stochastic and unpredictable nature of shared wireless medium and time varying throughput seriously limits the increasing requirement of Quality of Experience (QoE). In this paper, we investigate the problem of maximizing the user's QoE of viewing video streaming through automatic bitrate adaptation. To this end, we first formulate a stochastic optimization problem by considering the fundamental uncertainty of wireless access networks, i.e., the stochastic throughput. Then, we develop a dynamic bitrate adaptation algorithm based on the Lyapunov optimization framework. Our algorithm makes the video bitrate selection based on the current queue buffer state, and balances the tradeoff between user's QoE and buffer occupation (i.e., memory utilization). Numerical simulation results verify our algorithm outperforms the baseline scheme in providing QoE and bandwidth utilization, and could efficiently balance our desired tradeoff.