IoT node selection in Opportunistic Networks: Implementation of fuzzy-based simulation systems and testbed

Abstract

Opportunistic Network is a type of Ad-hoc networks, where nodes may be connected temporarily. Contacts between nodes are intermittent and an end-to-end link may never be established. These networks enable communication when conventional network infrastructures cannot be implemented. Each day, numerous nodes become part of the Internet, extending the connectivity to everyday objects. This network, made up of diverse nodes which connect to each other, is called Internet of Things (IoT). An IoT network is heterogeneous, which means that when a task needs a certain action, there are some IoT nodes who may complete it better than the others. Choosing which nodes are better fitted, is the main challenge we have considered in this paper. To make a proper decision on which IoT nodes are better than others, we have proposed two systems based on Fuzzy Logic: IoT Node Selection System 1 (INSS1) and IoT Node Selection System 2 (INSS2). For INSS1 we use three input parameters: Node’s Distance to Task (NDT), Node’s Remaining Energy (NRE), Node’s Buffer Occupancy. For INSS2, we add a new parameter Node Inter Contact Time (NICT) and together with 3 parameters of INSS1, there are four input parameters. The simulation results show that INSS2 makes a better IoT node selection, but is more complex than INSS1, because of higher number of rules. We implemented a testbed and compared simulation results with experimental results. The testbed results give a better insight than the simulation results due to the fact of being implemented on a real environment.