A HYBRID ANALYSIS SIMULATION FOR ATM PERFORMANCE WITH APPLICATION TO QUALITY-OF-SERVICE OF CBR TRAFFIC/

A discrete time queueing model for the performance of an ATM system is analyzed using matrix analytic methods. Time is segmented into slots with each slot equal to the transmission time of one ATM cell. The ATM system is modeled as a single server queue with Markovian arrivals an d service time equal to one slot. The arrival process includes as a sp ecial case the superposition of on-off sources, possibly heterogeneous . The queueing model is of the ''M/G/1 type''. By exploiting the struc ture of the ''M/G/1 type'' Markov chain, the complexity of the solutio n to the problem is reduced to only the inversion of a 2x2 matrix irre spective of the size of the Markov chain. This simplification allows u s to investigate Constant Bit Rate (CBR) traffic performance (or quali ty-of-service) issues through a hybrid analysis/simulation approach. S pecifically, we analyze the impact of on-off background traffic on the probability of consecutive cell losses, cell delay variation, and tra ffic shaper or playback buffer underflow and overflow probabilities of CBR traffic sources.