Pippi Longstocking calculus for temporal stimuli pattern on YouTube QoE
Tobias Hoßfeld, Dominik Strohmeier, Alexander Raake, Raimund Schatz · 2013
Video streaming over wireless broadband is currently accounting for half of the world-wide mobile data traffic, and its share is expected to increase even further. As a consequence, mobile operators need to understand how Quality of Experience (QoE) of mobile streamed video is impacted by the network delivery. With the HTTP-based video streaming used by video portals like YouTube or NetFlix, problems or insufficient resources in the networks manifest as stalling events during service consumption on the end user device. In this paper, we discuss the challenges for QoE monitoring of HTTP-based mobile video regarding three aspects. First, the QoE impact of different patterns in terms of stalling event frequency and length is quantified based on a study conducted for YouTube video. Second, we evaluate the prediction power of aggregated measures like the total stalling time. Third, we discuss the resulting consequences and remaining challenges for mobile video QoE modeling, monitoring and provisioning.