Dynamic Peer-Assisted Dash Framework Using Dynamic Updates And Prioritization Scheme

In this paper, we present a dynamic framework for P2P video streaming using dynamic adaptive streaming over HTTP (DASH). DASH is a very popular standard and is expected to dominate the adaptive streaming market. Our proposed framework defines several components; DASH Session Coordinator (DSC), Intelligent Content Server (ICS), and Adaptive Streaming Client (ASC). Each component has a defined set of rules as well as a simple messaging framework. In addition, all components work together to achieve optimized QoE for all clients streaming the same content. The proposed framework is customized for DASH and provides higher priority for P2P streaming over direct streaming thus allowing clients to share downloading content instead of competing for higher bandwidth and consequently improve network utilization and client QoE. Results of our proposed technique showed an increase in client download rate by approximately 60% which is directly related to the QoE perceived by the client. In addition, a reduction of 87% in the number of playback stalls and pauses was achieved due to the elimination of bandwidth competition among clients.