FEDCVS: a fair and efficient scheduling scheme for dynamic cooperative video streaming on smartphones

As video applications are increasingly popular over smartphones, many cooperative video streaming mechanisms have been proposed. These mechanisms use cellular link as well device-to-device links simultaneously to provide higher quality video streaming to mobile users. However current works solely focus on throughput enhancement in static scenarios. Consequently these mechanisms result in unfairness since smartphones with higher download rate expend more cellular traffic and monetary costs. Additionally, previous works assume a static scenario that all smartpone users start to watch the same video at the same time. Obviously, the static scenario is unrealistic in actual mobile environments. Based on these insights, in this paper, we focus on a more practical dynamic cooperation scenario and propose a scheduling scheme to achieve efficient cooperative video streaming and guarantee fluent user experience. More importantly, the proposed scheduling scheme achieves a significant improvement in fairness among cooperators. Through extensive simulations across a wide range of scenarios, we show that the proposed scheme significantly outperforms other works by 52%, 24% and 27% respectively in terms of fairness, without sacrificing efficiency.