Measurement Modelling and Simulation of Videoconference Traffic from VBR Video Encoders

we contribute measurement and modelling results on discrete-event queueing studies on the network performance of videoconference traffic encoded by variable bit rate video encoders over IP networks. The statistical analysis and modelling of extensive data compressed during realistic talking heads communication by the encoders of the videoconferencing tool ViC, namely, NV, NVDCT, H.261, H.263, H.263+, BVC and CellB (the traffic patterns of which, to the best knowledge of these authors, haven’t been studied in their entirety) suggest that although the fit of the actual frame-size sequence histogram with a Gamma probability distribution function, using the method of moments, is not accurate, a careful choice of the autocorrelation parameter by calculating the autocorrelation decay not at the initial lags, but rather through taking the long-term trends into account, can lead to a conservative with respect to queueing and simple generalization of the short range dependent DAR and GBAR models. By comparing the buffer overflow estimation given by extensive simulation runs of the models against the one given by the actual videoconference data, all being included as the video source of a single-server queueing system implemented in ns-2 simulator, we confirm our claims and the importance of the long-term trends of the autocorrelation function on queuing performance.

Videoconference traffic (especially H.26x encoded) is expected to account for large portions of the multimedia traffic in future heterogeneous networks (wire, wireless and satellite). The videoconference traffic models for these networks must cover a wide range of traffic types and characteristics because the type of the terminals will range from a single home or mobile user (low video bit rate) to a terminal connected to a backbone network (high video bit rate). Furthermore, successful videoconference traffic modelling can lead to a more economical network usage, leading to lower communication costs and a more affordable and of higher quality service to the end-users. However, the variation of the videoconference communication parameters (which in most cases are the target video bit rate, target frame rate and the used encoder), the differences in the video coding algorithm implementations and finally the variety of the visual contents (

Free download research paper