QoS-aware Optimal Deployment of LoRa Gateways in UAV-enabled LoRaWANs

Employing flying gateways such as Unmanned Aerial Vehicles (UAVs) is an attractive approach to providing fast and effective densification for wireless access networks. Flying gateways as UAVs can solve performance issues of different applications, including those involving Internet of Things (IoT) devices. On the other hand, traditional mobile network operators need solutions for integrating IoT technologies such as Long Range Wide Area Network (LoRaWAN) with their 3rd Generation Partnership Project (3GPP) infrastructure. In this paper, we have associated Quality of Service (QoS) parameters from the non-3GPP Long Range (LoRa) technology to the 3GPP-defined network slicing. To ensure the QoS of the IoT devices, i.e., the slices QoS requirements, we have formulated an optimization problem to obtain the minimum number of UAVs and their positions and consider the interference reduction. We also have introduced three optimization strategies for the problem: (i) bi-objective focusing on minimizing the number of UAVs, (ii) bi-objective focusing on the distribution of devices between Spreading Factor (SF) configurations, and (iii) mono-objective to minimize the number of UAVs, used as a baseline. We have evaluated our proposal through analytical modeling and simulations using Network Simulator 3 (ns-3), in which we confirm the QoS assurance and interference reduction.