Throughput-based quality adaptation for DASH in 5G mobile networks

Abstract

Video streaming in mobile networks is currently the most widely used service and its usage is expected to grow exponentially in the next years. Due to the changing conditions of the radio interface, techniques likes Dynamic Adaptive Streaming over HTTP (DASH) allows the user equipment to request the video coding rate that better matches the instantaneous network capacity. There are three types of algorithms to select the appropriate video coding rate based on different types of quality of service metrics: throughput-based, buffer-based and hybrid. In this paper we present three different versions of a throughput-based algorithm, comparing their performance in terms of mean and mode of the video quality index as well as the number of overlapping video chunks. We focus on the end-user quality of experience to evaluate which is the implementation that optimizes the performance.