Dynamic resources allocation in non-3GPP IoT networks involving UAVs

In this work, we investigate how to minimize the number of gateways deployed in Unmanned Aerial Vehicles (UAVs) needed to meet the demand for non-3GPP Internet of Things (IoT) devices, seeking to improve the Quality of Service (QoS), keeping a balance between delay and data rate. Gateways deployed in UAVs add the mobility flexibility of UAVs, which paves the way for meeting emergency demand increments. The 5 ^th Generation Networks (5G) and Beyond 5 ^th Generation Networks (B5G) systems incorporated access to IoT devices via non-3GPP access, opening up new integration possibilities. Furthermore, Low Power Wide Area Network (LPWAN) networks, especially Long Range Wide Area Network (LoRaWAN), allow access over long distances with reduced energy consumption. In this scenario, our work proposes a Mixed Integer Linear Programming (MILP) optimization model to minimize the number of UAVs that meet the increment of emergency demand, comply with limits of QoS, and maintain the compromise between delay and data rate. Simulation results show that the proposed model significantly reduces the number of gateways, maintains optimal levels of QoS, and maintains the compromise between delay and data rate compared to the presented baselines.