A QoE-Driven Rate Adaptation Approach for Dynamic Adaptive Streaming Over HTTP

In this paper, we aim to optimize the quality of experience (QoE) of Dynamic adaptive streaming over HTTP (DASH) through better bitrate adaptation. To this end, we first propose a quantitative QoE evaluation model based on playback continuity, segment media quality and perceptual quality fluctuations caused by bitrate switching to evaluate users’ subjective perception accurately. Large quantities of subjective mean-opinion-score (MOS) tests show that there exists a good correlation between our model score and subjective perception. Second, we formulate the rate adaptation in DASH services as the optimization problem of multi-stage decision process, and the goal of optimization is to maximize the reward of decision that measured by the QoE model. Numerous experiment results demonstrate that our proposed rate adaptation approach has good performance in different network environments.