Analysis of adaptive GBN schemes in a Gilbert–Elliott channel and optimisation of system parameters

Abstract

In this article, an adaptive go-back-N ARQ scheme is analyzed and optimized in a time-varying channel environment with unreliable feedback. Two-state Markov chains, known as Gilbert–Elliott channel models, are assumed for both forward and return channels. The adaptivity of the scheme is based on varying the number of copies of a packet that are sent at each transmission attempt. The scheme has two design parameters. They are associated with detecting changes in the forward channel state. The performance of the adaptive scheme is measured by its average throughput, which is upper-bounded by the throughput of the so-called ideal adaptive scheme. This upper bound is very useful in the optimization of the design parameters. The effect of round-trip delay and the forward channel variation rate on the optimal parameter values is studied, as well as the effect of the return channel characteristics on both the performance of the scheme and on the optimization results.