Markov Chain-based Optimization of Multihop IEEE 802.15.4 Wireless Sensor Networks

Abstract

In this work, we propose an optimization framework for the IEEE 802.15.4 medium access control (MAC) protocol. More precisely, we derive a theoretical tool providing reliable guidelines for tuning the parameters of the MAC protocol. The presented tool could be used in two different directions: (i) for fixed network topology, it might be of interest to determine the MAC protocol configuration able to guarantee the best performance according to some quality of service metric; (ii) for fixed parameters of the MAC protocol, it might be of interest to determine the optimal network topology. Both these situations appear in practical situations. In particular, the first scenario happens in all cases where one cannot decide a-priori the node positions, because of some randomness or for some physical constraints, so that the only degree of freedom is given by the MAC protocol itself. The second set of scenarios occurs in the circumstances where one can decide the nodes displacement, thus introducing more degrees of freedom, but making also the problem solution more complicated. The proposed optimization tool applies some classical operative research instruments to a recently proposed Markov-Chain based model that has shown to be suitable for the performance analysis of a generic Cluster-Tree (CT) multihop IEEE 802.15.4 network. We will also show that this tool could be effectively used in a real scenario with a low-cost low-energy hardware platform.