Improving Energy Efficiency and Reliability by Optimizing Gateway Placement in Dense LoRaWAN Networks

Energy efficiency has become a top priority for NextG networks, driven by the need to reduce costs, comply with regulations, and mitigate the environmental impact of telecommunications infrastructure. LoRaWAN (Long Range Wide Area Network) plays a significant role in NextG networks due to its unique features involving low power consumption, cost-effectiveness, and broader coverage. As technology advances, there is an exponential rise in network traffic, resulting in increased interference within networks and reduced coverage. By strategically positioning gateways, network operators can ensure that the network meets the requirements of various applications and environmental conditions while managing deployment costs effectively. This paper presents two algorithms to improve the gateway placement strategy in LoRaWAN. The First uses the analysis of the connected components (CC-Place), and the second is based on Simulated annealing (SA-Place). We show that these approaches are energy-efficient for denser networks and those with broader coverage. Moreover, our results also show that our proposed algorithms make these networks more reliable by reducing the collision probabilities. We compare these algorithms with Graph-based gateway placement in [1] and intelligent, fuzzy C-Means (FCM) cluster-based algorithm in [2]. We evaluate the number of gateways each algorithm uses and their collision probabilities. We show that our algorithms propose fewer gateways, thus lowering the initial network cost.