Modelling of hyper-exponential ON/OFF traffic streams by a four-state Markov Modulated Rate Process (MMRP) is considered as a means of obtaining a sufficient(v accurate, yet computationally tractable analytical model for ATM traffic. A general formulation of the proposed model along with the associated multiplexing analysis is first developed. Then, a special time-reversible version of the model that arises naturally in many practical configurations of interest is analysed, which enables the modelling and the parameter assessment of the ON and OFF phases independently as two-stage, Erlang-type processes. Application examples with comparison of analysis, simulation and experimental results derived on an ATM testbed are given.
Abstract. A number of experiments were designed and performed on the EXPLOIT ATM testbed, aiming at validating selected results on statistical multiplexing of bursty traffic. The influence of the burst-size and silence-duration distributions on the multiplexer's performance, the performance experienced by the individual classes in a heterogeneous traffic environment and the performance in the case of heterogeneous traffic featuring different time scales are the main issues investigated.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.