/PPBP-ns3

Poisson Pareto Burst Process application for ns-3 discrete-event network simulator.

Primary LanguageC++

Poisson Pareto Burst Process (PPBP) ns-3 traffic generator

Motivation

This repository is an extension to the PPBP ns-3 generator designed by Doreid Ammar. Unfortunately, their version is not compatible with newer releases of ns-3(ns-3.15 and later). This is mainly because of the shift in the random variable creation in the newer versions of ns-3. In ns-3.14 and earlier, ns-3 simulations used a different wrapper class called ns3::RandomVariable. As of ns-3.15, this class has been replaced by ns3::RandomVariableStream; the underlying pseudo-random number generator has not changed. Most of the ns-3 random variable classes have changed from ExponentialVariable to ExponentialRandomVariable etc. including modified attributes.

Poisson Pareto Burst Process (PPBP) Primer

The key principles of Poisson Pareto burst process are as follows:

  • Burst arrivals follow a Poisson distribution

  • The burst lengths follow a Pareto distribution. For large burst lengths, overlapping traffic aggregate to form a long-range dependent traffic model.

  • Each burst is modeled by a constant bit rate.

Instructions

References

  • Doreid Ammar's PPBP traffic generator for older versions of ns-3.

  • D. Ammar, T. Begin and I. Guerin Lassous, "A new tool for generating realistic Internet traffic in NS-3", Proc. of International ICST Conference on Simulation Tools and Techniques (SIMUTools), Barcelona, Spain, March 21-25, 2011, Poster.

  • M. Zukerman, T. D. Neame and R. G. Addie, "Internet Traffic Modeling and Future Technology Implications", Proc. of IEEE INFOCOM 2003, San Francisco, USA, April 2003.