Modeling multi-hop CSMA-based networks through Semi-Markov chains

In this paper we address multi-hop Carrier Sense Multiple Access (CSMA)-based networks and we propose a novel approach based on a semi-Markov chain analysis, decoupling node and network levels. We model node states through a Semi-Markov Process and, on the other hand, we define a finite state transition diagram to describe the network states. The latter is used to derive the parameters included in the Semi-Markov Chain that depend on the network status, as for example, the probability of finding the channel busy. In particular, we consider a multi-hop wireless network, where nodes are deployed over a straight line. Each node sends its data to the next node in the line, to reach the destination located at the end. Nodes use a CSMA-based protocol, where priority in the access to the channel is given to nodes closer to the destination. We mathematically derive the network performance, in terms of throughput and energy consumption. The model is validated through comparison with simulations.