The Impact of Non-Gaussian Distribution Traffic on Network Performance

Recent extensive measurements of real-life traffic demonstrate that theprobability density function of the traffic is non-Gaussian. Ifa traffic model does not capture this characteristics, any analytical orsimulation results will not be accurate. In this work, we study theimpact of non-Gaussian traffic on network performance, and present anapproach that can accurately model the marginal distribution ofreal-life traffic. Both the long- and short-range autocorrelations arealso accounted. We show that the removal of non-Gaussian componentsof the process does not change its correlation structure, and wevalidate our promising procedure by simulations.