How to correctly use the protocol interference model for multi-hop wireless networks

This paper tries to reconcile the tension between physical model and protocol model that have been used to characterize interference relationship in a multi-hop wireless network. The physical model (a.k.a. SINR model) is widely considered as a reference model for physical layer behavior but its application in multi-hop wireless networks is limited by its complexity. On the other hand, the protocol model (a.k.a. unified disk graph model) is simple but there have been doubts on its validity. This paper explores the following fundamental question: How to correctly use the protocol interference model? We show that in general, solutions obtained under the protocol model may be infeasible in practice and thus, results based on blind use of protocol model can be misleading. We propose a novel concept called "reality check" and present a method of using protocol model with reality check for wireless networks. Subsequently, we show that by appropriate setting of the interference range in the protocol model, it is possible to narrow the solution gap between the two models. Our simulation results confirm that this gap is indeed small (or even negligible). Thus, our methodology of joint reality check and interference range setting retains the protocol model as a viable approach to analyze multi-hop wireless networks.