Abstract

Abstract-The rapid growth of mobile video traffic is placing a severe strain on mobile network operators. In this paper, we study the problem of efficient video delivery over the cellular downlink. The key objective is to maximize the Quality of Experience (QoE) of the user, which cannot be entirely captured by traditional QoS metrics like packet loss rate, delay and throughput. In this paper, we consider application level metrics such as the buffering ratio and low bit rate ratio, that are known to closely correlate with user QoE. Our goal is to develop a framework that can guarantee a pre-specified QoE for each user, provided it is feasible under the current channel conditions. We present a two-tiered solution with a standard basestation scheduler that works on a per-packet basis and a Video Management System (VMS) that works at the granularity of thousands of video frames. The video management system uses knowledge of the video playout curves and the average future channel states to develop a scheduling policy that is feasibility optimal, i.e., it can satisfy any set of QoE guarantees that can be satisfied by any other scheduling algorithm. We describe algorithms that maintain a guaranteed buffering ratio, or a guaranteed low bit rate ratio, or both. The algorithms are simple and leverage recent results on real-time scheduling in wireless networks. We evaluate the performance of our algorithms using real video traces and a standard channel model. The VMS ensures that the per-user QoE guarantees are maintained, as compared with a standard PF scheduler that is oblivious to application level QoE requirements.