A Distributed Critical Node Detection Algorithm for UAV Swarm Networks

Unmanned aerial vehicle (UAV) swarm can be widely used in cooperative detection and strike across wide geographic areas. To support the collaboration of multiple UAVs, a connected network topology is essential. Thus, critical nodes, whose removal will disconnect the network into two or more separate components, will play an important role in UAV swarm networks. This paper investigate the distributed critical node detection algorithm for UAV swarm networks. Firstly, we analyze the logical relationship between critical node, robustness, and k-hop subgraph. Guided by the analysis, we propose a distributed critical node detection algorithm by using the geometrical relationship of neighbors and the k-hop local topology. Abundant simulation results indicate that the proposed algorithms can strike a good balance between accuracy and overhead and greatly outperform the global detection algorithms especially in large-scale networks. Besides, the simulation study provides some useful design guidance for robust topology control of UAV swarm networks.