Towards a robust FANET: Distributed node importance estimation-based connectivity maintenance for UAV swarms

The emergence of unmanned aerial vehicle (UAV) swarms has brought a variety of critical challenges to solve. Among them, connectivity maintenance has been the subject of numerous investigations. Especially in a network whose topology dynamically changes, such as the flying ad hoc network (FANET), the connectivity is prone to be lost when topology changes. Besides, the network connectivity tends to deteriorate when node failure occurs. In this paper, a distributed connectivity maintenance framework that considers the vulnerability of nodes and the robustness of a network, is proposed. Firstly, a distributed neighbor selection rule to determine the neighbor set of each node in the network is designed. Secondly, a distributed connectivity maintenance strategy to evaluate the impact of the removal of a node is developed. Based on the perturbation of algebraic connectivity (AC), a distributed node importance evaluation algorithm is designed to approximate the AC measure of the resulting network. Finally, numerical simulation results are presented to validate the effectiveness of the proposed framework.