Application-Aware LoRaWAN Gateway Placement in Massive IoT Water Distribution Networks

The application of the Internet of Things (IoT) for environmental sustainability in the context of smart Water Distribution Systems (WDSs) is gaining more and more interest with the aim to preserve one of our most precious resources. Indeed, effective water monitoring is fundamental to identify faults, to optimize resource allocation, and to ensure a reliable delivery of clean water. Leveraging the low-power wide area network (LPWAN) communication technologies like LoRaWAN (Long Range Wide Area Network), this paper presents an innovative, context-sensitive approach to enhance the positioning of gateway concentrators in high-performance WDNs. Our approach is based on graph theory, which enables us to rank graph nodes according to their centrality, together with graph signal processing, by modeling pressure values as signals on graphs. By focusing on strategic locations that maximize network coverage and minimize infrastructure redundancy, we demonstrate that our method significantly reduces the required number of gateways and, consequently, the overall energy consumption of the system. We test our method on a synthetic but realistic scenario of massive IoT networks equipped by hydraulic and radio properties. We focus on a massive IoT in WDSs, utilizing simulation tools such as EPANET and NS-3, and the application of graph signal processing rooted in graph theory to enhance packet delivery ratio and energy efficiency.