Energy-efficient HTTP Adaptive Streaming with Anticipated Channel Throughput Prediction in Wireless Networks

Exploiting predicted channel information and designing energy efficient content delivery protocols has started to draw attention, which is referred to as predictive, anticipatory, or context-aware resource allocation. In this paper, we investigate how predicted user rates can be exploited for streaming on-demand mobile video with dynamic adaptive streaming over HTTP(DASH). Specifically, we propose an edge-cloud assisted framework for prediction based DASH streaming; For optimal prediction scenario, we propose a lightweight algorithm to solve it; For imperfect prediction scenario, we model uncertainty in predicted user rates and propose a chance constraint programming method to dynamically allocate the risks, optimize QoE and system efficiency; For the multi-user scenario, we propose a quality-level-aware throughput gain maximization method to improve the network efficiency, fairness and QoE for all users under different prediction error variances; Simulation studies show that our method has a better performance than traditional methods in terms of average QoE, fairness and energy efficiency.