Modeling Network Traffic
Mohsen Guizani, Ammar Rayes, Bilal Muhammad Khan, Ala Al‐Fuqaha · 2010
Network traffic modeling is essential for theoretical capacity and performance representation as well as for simulation models. This chapter presents several traffic models that are used by simulation researchers and practitioners to model network traffic loads. It also describes the most commonly used global optimization techniques to solve optimization problems. These optimization techniques are useful in modeling and solving complex physical problems. Traffic models can be generally divided into two main categories: models which exhibit long-range dependencies or self-similarities; and Markov models that exhibit only short-range dependence. The chapter studies two traffic models that fall into the second category, namely constant bit rate and variable bit rate. It also considers one traffic source that falls into the second category, namely the Pareto traffic. Evaluating the performance of a mobile ad hoc network depends greatly on the mobility model used. Controlled Vocabulary Terms computer networks; hidden Markov models; mobile ad hoc networks; modelling; optimisation; Pareto distribution