Métodos de modelamiento y predicción de tráfico orientados a plataformas de transmisión de video e IPTV usando series de tiempo1

This paper, intends to review the most important techniques that allow per- forming statistic video traffic modeling and forecasting, focusing in time series models, so we can identify which models are better to describe the representative IPTV stochastic traffic. For this purpose, we make a short introduction to time series analysis, and a review of the state of the art on video traffic modeling over data networks. From this research we conclude that, of all the available models to describe and forecast network traffic, the more appropriate to use within IPTV systems are ARIMA time series models, from which SARIMA model are the best option.