A Simple Model for Predicting the Number and Duration of Rebuffering Events for YouTube Flows

In this paper, we propose a simple model for predicting the number of rebuffering events and their duration in progressive downloads from YouTube. These metrics are necessary to predict the quality perceived by YouTube users. The proposed rebuffering model is based on two thresholds of the amount of data stored by the player buffer: the first threshold is extracted from the results of previous studies, and the second is derived from the experimental results presented in this paper. The proposed model can be easily implemented in simulation tools and we present an example of its use in a Long-Term Evolution simulator in which the mentioned quality metrics have been estimated for different users.