QoE-Based Framework to Optimize User Perceived Video Quality

摘要: Video streaming has become a main contributor in an ever increasing Internet traffic, and meets the users expectation is challenging task for both Network service Provider (NsP) Content (CsP). In this context, new metric called: Quality of Experience (QoE) evolved to measure user satisfaction using video service, it becomes key driver achieving business goal NsP CsP. perspective, we have proposed novel framework that considers QoE adapt quality, named Optimized DASH (OQD). The objective OQD optimize experience, maximize bandwidth usage. A Machine Learning (ML) approach based on GRadient Boosting (GRB) method implemented predict three important network application Influence Factors (QoE IFs). We use Reinforcement (RL) select optimal quality segment, which improves QoE. performance evaluated compared against Greedy adaptive bit-rate terms re-buffering, utilization, average MOS, standard deviation MOS. results clearly show performs well, as user’s perceived regulator overall delivery network.