Game theory approach to peer-to-peer video streaming: a comprehensive survey

Abstract

Recent reports and forecasts indicate video is the most important part of the internet traffic. This traffic is the result of increasing video applications, also the high expectations of today users. In fact, streaming HD videos with minimum initial buffering time and interruption forms the QoE backbone. Therefore, resource provisioning for this demands is very challenging and the scalability of these systems depends on spending a high cost on preparation new resources or taking the advantage of users' abilities in the form of a peer-to-peer (P2P) system. Although P2P architecture can significantly improve scalability, it has severe management complexity challenges. The dynamic nature and the autonomy of peers are the prominent reasons for this issue. Because of the strategic context in P2P video streaming and due to the existence of conflicting actions for participant entities in such systems, using game theory has been very interesting as a mathematical tool for modelling and analysing in recent related investigations. Due to the multitude of these methods, the lack of a comprehensive review is intensively palpable. This paper seeks to fill this research gap especially with focus on applying non-cooperative games to P2P video streaming resource allocation.