Traffic Models for the OPNET Simulator of the ALAX

Abstract

Traffic modeling is an important component in the design of high speed and high bandwidth environments. Here we present the traffic models we developed in OPNET, which were currently being used to validate and refine the design of an ATM-LAN Access switch. These models attempt to capture the correlation and burstiness of the traffic that the switch is expected to carry. This burstiness and correlation in the traffic manifests itself in terms of packet delays and buffer overflow. Traffic models mainly fall into two categories -(1)Short Range Dependent models and (2) Long Range Dependent models. These two differ dramatically in terms of the behavior of the correlation function. Traditionally designers have been using Short Range Dependent models. However recent investigations suggest that traffic sources such as Variable Bit Rate Video and Ethernet traffic are better represented by Long Range Dependent models. Here we describe both types of models. The Long Range Dependent source is based on the M/G/infinity model. Two Short Range Models were developedne is a Markov Modulated Poisson Process and the other is based on an underlying Autoregressive model. This report describes all these models in sufficient detail along with the relevant implementation issues.<P> <BR>