The extended On/Off process for modeling traffic in high-speed communication networks

High-speed network traffic is impulsive and exhibits long-range dependence. While the latter characteristic has been studied extensively, the former has received much less attention. The On/Off model is a well known model for capturing the long-range dependence of traffic traffic. In this paper we propose an extension to the On/Off model, which allows the model to also capture the traffic impulsiveness. We provide queuing analysis of the proposed model, which along with numerical results suggests that the traffic marginal distribution may have significant impact on networking engineering.