QoE-Aware Playback Buffer Level Control Strategy for Adaptive Video Streaming Using MPEG-DASH

Abstract

Adaptive video streaming using Moving Picture Experts Group dynamic adaptive streaming over Hypertext Transfer Protocol (MPEG-DASH) enables adaptive bitrate (ABR) control based on network conditions. Network conditions can be estimated based on the network throughput or playback buffer level. However, the conventional proportional-derivative (PD)-based ABR algorithm that considers both network through-put and playback buffer level can degrade the quality of experience (QoE) when the available bandwidth of each user fluctuates. This study proposes an ABR algorithm that combines multiple bitrate selection methods using PD control and buffer-based throughput estimation, considering buffer-level fluctuations to achieve a stable bitrate and buffer-level control. Simulations confirm that the proposed algorithm improves QoE under multi-user and throughput-varying conditions and does not negatively impact another ABR algorithm in a heterogeneous environment.