Modeling and Design of Adaptive Video Streaming Control Systems

Abstract

Adaptive video streaming systems aim at providing the best user experience given the user device and the network available bandwidth. With this purpose, a controller selecting the video bitrate (or level) from a discrete set $\mathcal{L}$ has to be designed. The control goal is to maximize the video bitrate while avoiding playback interruptions and minimizing video bitrate switches. In this paper, we propose a hybrid dynamical system modeling the essential features of an important class of controllers for adaptive video streaming systems. We derive tuning rules to achieve key performance goals by sizing the control system parameters. We show how to: (i) tune the controller parameters to keep the video level switching frequency below a given target; (ii) design the video levels set $\mathcal{L}$ to obtain a performance tradeff between switching frequency and storage costs at the servers; (iii) find the minimum amount of playout buffer that should be stored to avoid rebuffering events with a given probability in case of temporary bandwidth drop. The theoretical results are validated through numerical simulation and experimental evaluation.