UAVs Relay in Emergency Communications With Strict Requirements on Quality of Information

Abstract

Mobile Ad-hoc Networks (MANETs) are increasingly being used to provide communications in emergencies. Maintaining the required quality of service of information including its timely delivery, in MANETs, is extremely challenging under dynamic conditions such as a changing bushfire scenario. Nodes get destroyed or their link qualities degrade leading to sub-optimal paths, reduced network performance, and disconnected end nodes. To remedy partitioning in ground-based MANETs and to improve their communication performance, aUtonomous Aerial Vehicles (UAVs) have emerged as a promising relay solution. This paper presents U-QoIT, a UAV deployment algorithm for connecting the disconnected end nodes in a dynamic bushfire scenario. It aims to meet the Quality of Information requirements, with a minimal number of UAVs deployed. U-QoIT takes the UAV deployment problem as a combinatorial search problem and reduces the deployment complexity by applying a reduction by minimization approach. This is done to deploy fewer UAVs without compromising connectivity. Extensive simulation results depict that U-QoIT significantly improves user usability by 48% compared to the best-performing baseline approach, of Dynamic Service Area (DSA) based UAV deployment, at the expense of a negligible increase in computational time. It achieves a 38% drop in packet latency and retains its out-performance in a highly mobile network. In a sparse network, it improves user usability by 106%. Furthermore, it needs fewer UAVs, compared to the baseline UAV deployment methods, without compromising its performance gains.