Enabling Optimizations Of Video Delivery In HTTP Adaptive Streaming
COM2 Level 4
Executive Classroom, COM2-04-02
Abstract:
Concurrent HTTP adaptive streaming (HAS) players behave selfishly and the resulting competition for shared resources leads to underutilization or oversubscription of the network, presentation quality instability and unfairness among the players, all of which adversely impact the viewer quality of experience (QoE). Empirically, these issues are aggravated when the number of competing players increases. We call these issues the HAS scalability problem.
In this proposal, we design and develop a novel suite of Adaptive Bitrate (ABR) solutions to address the scalability problem of HAS and HAS-like (DASH) video delivery systems. The proposed solutions can be classified into two main categories: centralized (hybrid-based), and distributed (client-based) approaches.
In the centralized approach, we investigate how a central entity like a Software Defined Networking (SDN) controller can assist the video players in their bitrate selection and bandwidth allocation. Such a central entity has a general overview and control over the network, and thus it can facilitate appropriate ABR decisions that are taken by the players.
In the distributed approach, we investigate how distributed rules (e.g., game theory, multisource/multipath mechanisms, etc.) can be leveraged by a client-driven ABR scheme. Thus, a set of video players cooperate and coordinate among themselves to achieve their common objective (i.e., maximizing the viewer QoE) without any implicit or explicit signaling messages that might harm the network efficiency.