A new fractal point process for modeling self-similar traffic

Packet traffic from various sources-ethernet LAN, VBR video, CCSN-has been shown to exhibit self-similarity and related to properties of long range correlation, slowly decaying variances and fractal dimensions. This discovery has motivated research into unconventional traffic models such as fractional Gaussian noise, fractional ARIMA, and chaotic maps. Queuing analyses based on these models have shown unusual results that cannot be explained by traditional models. In this paper, we present a novel fractal point process for modeling self-similar traffic, which is a general fractional-binomial-noise driven Poisson process (FNBDP). This model is presented to determine the important statistic, peak-to-mean ratio, without losing fractional property. We firstly demonstrate the fractional behavior of the general FBNDP and then compare its parameters to statistics collected from real traffic