A Two-Dimensional Markov Model for Cross-Layer Design in AMC/ARQ-Based Wireless Networks

Nowadays, there are lots of research efforts focusing on cross-layer designs combining adaptive modulation and coding (AMC) with an automatic repeat request (ARQ) protocol. One of the main drawbacks of these research works is that they rely on first-order amplitude-based finite-state Markov chains (AFSMC) to model the wireless fading channel. Furthermore, most of the analytical models used in these works present several deficiencies that could compromise the design of cross-layer protocols. To overcome the aforementioned problems, in this paper we propose a novel cross-layer analytical framework that is based on both the use of a first-order two-dimensional Markov model and a judicious implementation of the AMC threshold searching algorithm used in the transmission mode (TM) selection. This threshold search is designed independently from the channel model and is able to discriminate between useful and useless TMs. To exploit the joint impact on QoS performance measures of both AMC and ARQ we present a formulation of an AMC/ARQ cross-layer design as a constrained optimization problem.