Overview of the QoE-aware rate adaptation algorithms used in MPEG DASH

Abstract

Standard called MPEG Dynamic Adaptive Streaming over HTTP (MPEG DASH) was developed in order to ensure that the end user shall be given the best possible quality of requested content under certain network conditions. DASH is used to notify the user that there are several transmission streams with different quality levels available. Bitrate adaptation algorithms used in DASH-based system have developed rapidly over the last four years. This paper gives an overview of the state-of-the-art rate adaptation algorithms. There are several major groups of bitrate adaptation algorithms used in MPEG DASH clients: bandwidth-based algorithms, buffer-based algorithms, and algorithms that take into account both the buffer size and the bandwidth. In recent studies regardless of the approach used in determining the next segment quality level, all algorithms take into account the users' Quality of Experience (QoE). QoE is measured using different metrics and subjective testing. There are different methods of assessing subjective quality like Double Stimulus Impairment Scale (DSIS), Double Stimulus Continuous Quality Scale (DSCQS), SSCQE (Single Stimulus Continuous Quality Evaluation) and ACR (Absolute Category Rating). Although novel adaptation algorithms have improved users' QoE there is still room for improvement.