Two-State Buffer Driven Rate Adaptation Strategy for Improving Video QoE over HTTP

Abstract

With the popularity of smart handheld devices, mobile streaming video has multiplied the global network traffic in recent years. To guarantee the best user experience, dynamic adaptive streaming over HTTP (DASH) is adopted as the de facto industrial technology, which can flexibly select the “proper” bitrate for each next video segment with the varying network throughput. In this paper, we propose a client-side buffer driven rate adaptation strategy which improves the real-time quality of experience (QoE) by answering to users’ differentiated QoE requirements at different buffer states. According to whether the downloaded video content is enough to resist bad network conditions, we define a startup state and a steady state of the buffer and take different rate adaptation goals for each state. For the startup one, filling the buffer as soon as possible helps improve users’ QoE the most. While for the steady state, our strategy aims at providing users with undisturbed viewing experience, which needs major concern over both the video quality and the rebuffering. Simulation results have shown that our strategy can effectively reduce the occurrence of rebufferings caused by the mismatch between requested video rates and varying network bandwidth, and attains standout performance on users’ QoE compared with classical rate adaption methods.