A Decision Theoretic Approach for In-service QoE Estimation and Prediction of P2P Live Video Streaming Systems Based on User Behavior Modeling and Context Awareness

摘要

Peer-to-Peer (P2P) systems are fast becoming the most popular live video streaming solutions due to their self-scaling properties, ease of deployment, cost effective nature, content addition and content distribution properties. In order to improve the user related and network related quality of these systems, in-service QoE estimation and prediction is required. In this paper, in-service QoE is estimated and predicted based on decision theoretic approach by modeling the problem as a decision network. The results of this new improved approach for QoE estimation and prediction for P2P live video streaming systems shows that the estimated QoE is not only user-centric and context aware but also dynamic in nature as it changes with the changing situation. The proposed approach calculates the QoE quantitatively and as the results indicate can replaced the existing objective and hybrid QoE estimation approaches.