Traffic modelling of asynchronous bufferless optical packet switched networks

Abstract

Analytical models based on stochastic processes have been widely employed in order to assess the performance of optical packet switched (OPS) networks. A crucial issue with this approach is how to model packet arrivals to an output port in an optical packet switch, and how to evaluate the blocking probability. This article presents various Markovian arrival models for asynchronous bufferless OPS networks. In addition to the well-known Erlang and Engset arrival models, we present two novel Engset based arrival models, i.e. the Engset Asymmetric arrival model and the Engset Non-looping arrival model. We consider optical packet switches with and without wavelength conversion. Analytical expressions for the time-, call-, and traffic congestion are derived for each arrival model. A numerical evaluation shows that the choice of arrival model and performance metric significantly influences the observed blocking probabilities under certain parameter settings. The major aim with this article is to contribute to establish a theoretical framework for teletraffic analysis in asynchronous bufferless OPS networks.