Exact Sampling of Stationary and Time-Reversed Queues

We provide the first algorithm that, under minimal assumptions, allows simulation of the stationary waiting-time sequence of a single-server queue backward in time, jointly with the input processes of the queue (interarrival and service times). The single-server queue is useful in applications of Dominated Coupling from the Past (DCFTP), which is a well-known protocol for simulation without bias from steady-state distributions. Our algorithm terminates in finite time, assuming only finite mean of the interarrival and service times. To simulate the single-server queue in stationarity until the first idle period in finite expected termination time, we require the existence of finite variance. This requirement is also necessary for such idle time (which is a natural coalescence time in DCFTP applications) to have finite mean. Thus, in this sense, our algorithm is applicable under minimal assumptions.

[1]  R. Durrett.  Probability: Theory and Examples , 1993 .

[2]  Sandeep Juneja,et al.  State-independent Importance Sampling for Random Walks with Regularly Varying Increments , 2012, 1206.3390.

[3]  Jose Blanchet,et al.  Steady-state simulation of reflected Brownian motion and related stochastic networks , 2012, 1202.2062.

[4]  Wilfrid S. Kendall,et al.  Perfect Simulation for the Area-Interaction Point Process , 1998 .

[5]  Karl Sigman,et al.  Exact simulation of the stationary distribution of the FIFO M/G/c queue: the general case for ρ , 2011, Queueing Systems.

[6]  P. Glynn,et al.  Simulating the maximum of a random walk , 2000 .

[7]  David Bruce Wilson,et al.  Exact sampling with coupled Markov chains and applications to statistical mechanics , 1996, Random Struct. Algorithms.

[8]  John Frank Charles Kingman,et al.  The single server queue in heavy traffic , 1961, Mathematical Proceedings of the Cambridge Philosophical Society.

[9]  Serguei Foss,et al.  Convergence rates in monotone separable stochastic networks , 2006, Queueing Syst. Theory Appl..

[10]  Upendra Dave,et al.  Applied Probability and Queues , 1987 .

[11]  Stephen B. Connor,et al.  Perfect simulation of M/G/c queues , 2014, Advances in Applied Probability.

[12]  J. Blanchet,et al.  On exact sampling of stochastic perpetuities , 2011, Journal of Applied Probability.

[13]  Jing Dong,et al.  Sampling point processes on stable unbounded regions and exact simulation of queues , 2012, Proceedings Title: Proceedings of the 2012 Winter Simulation Conference (WSC).

[14]  Karl Sigman,et al.  Exact simulation of the stationary distribution of the FIFO M/G/c queue , 2011, Journal of Applied Probability.