Research on fast link prediction of combat network based on CROTS
Ruihua Ding, Bin Wang, Kebin Chen, Jing Kang, Kang Ouyang · 2024
Through the analysis of combat networks, commanders can effectively understand the connections between combat nodes, thereby taking the initiative on the battlefield. However, due to interference from the battlefield environment and electromagnetic environment, commanders may not be able to obtain all information, resulting in incomplete topology of the combat network. Therefore, link prediction is needed to predict the possibility of links between nodes in the network. Based on CROTS (cross-domain passive teacher-student learning), this paper proposes a fast, efficient, and learnable link prediction algorithm, which can greatly improve the prediction speed. This prediction algorithm is studied through experiments and specific examples. This paper proves the effectiveness and feasibility of this method.