Statistical analysis and evaluation of modelling techniques for self-similar video source traffic

Video is expected to be a significant part of information in future mobile communication systems. An accurate and efficient video source traffic characterisation is therefore essential in order to assess new traffic handling algorithms and appropriately dimension the supporting communication networks. Modelling VBR traffic as a stochastic process is often challenging since such traffic exhibits the co-existence of long-term and short-term dependencies. To highlight such issues, in this paper, three video traffic-modelling techniques are presented and analysed in terms of first- and second-order statistics. A comparative evaluation of these models has been carried out based on simulation.