Priority queue with two-state Markov-modulated arrivals

Abstract

The authors consider a single server priority queueing system with two priority classes of customers. The work is motivated for performance analysis of multiplexing video and data in communication networks, where videos at cell level arrive according to a two-state Markov-modulated Poisson process (MMPP) and data arrive according to a Poisson process. With the supplementary variable method, the authors explicitly obtain the steady-state joint generating function of the queue lengths of each class. They also investigate the queue lengths of each class by a technique of Markov-modulated fluid flow (MMFF). In comparison with the MMPP model, the MMFF model provides computationally simple procedures and results. Numerical results show that performance measures obtained with the MMFF model are quite close to those with the MMPP model.