QUEUEING ANALYSIS FOR TRAFFIC CONTROL WITH COMBINED CONTROL OF DYNAMIC MMPP ARRIVALS AND TOKEN RATES

ABSTRACT. We analyze the queueing model for leaky bucket (LB) scheme with dynamic ar-rivals and token rates. In other words, in our LB scheme the arrivals and token rates are changed according to the buffer occupancy. In telecommunication networks, the LB scheme has been used as a policing function to prevent congestion. By considering bursty and correlated prop-erties of input traffic, the arrivals are assumed to follow a Markov-modulated Poisson process (MMPP). We derive the distribution of system state, and obtain the loss probability and the mean waiting time. The analysis is done by using the embedded Markov chain and supple-mentary variable method. We also present some numerical examples to show the effect of our proposed model. 1