Mean packet delay analysis for the selective repeat automatic repeat request protocol with correlated arrivals and deterministic and nondeterministic acknowledgement delays

Abstract

An analytical approach for analyzing the mean packet delay in a selective repeat (SR) automatic repeat request (ARQ) error‐control system is presented. The system is modeled as a discrete time queue with an infinite buffer. Transmission errors in the forward channel are assumed to occur independently, and packet arrivals may have different interarrival time distributions. Bernoulli arrivals in independent random‐size batches and nonindependent arrivals such as that modeled by a two‐state Markov process are investigated. Comparisons of numerical predictions with simulation results are shown to highlight the efficiency of the proposed analytical approach in particular in the case of correlated arrivals and long nondeterministic acknowledgement delays where methods previously proposed cannot be applied. The case of Poisson arrivals and deterministic acknowledgement delay is also discussed in appendix C in order to enable the application of the proposed approach to a wide class of network scenarios and validate the analytical predictions by comparisons with numerical results previously appeared in the literature.