Quality of Experience Estimation for WebRTC-based Video Streaming

Enabling high Quality of experience (QoE) for video streaming, which accounts for a large and increasing percentage of wireless traffic, anywhere and anytime is a challenging task for relatively new Web Real-Time Communication (WebRTC) protocols as well as for service providers and it triggers broad exploration of this area. In this paper, we present our experience in running experiments for estimation of QoE for a WebRTC video stream in Mobile Broadband (MBB) networks based on Quality of Service (QoS) measurements. For majority of our experiments we exploited the MONROE platform, which is the first open access platform that gives access to several hundreds of nodes for large-scale experiments in commercial MBB in Europe. For analysis of obtained QoS measurements we used a combination of subjective quality assessment and a machine learning approach.