Algorithm for Link Prediction in a Self-Regulating Network with Adaptive Topology Based on Graph Theory and Machine Learning

This article presents a functional graph model of a network with adaptive topology, where the network nodes represent the graph vertices, and data exchange between the nodes is represented as edges. The dynamic nature of network interaction complicates the solution of the problem of monitoring and controlling the operation of a network with adaptive topology, which should be done in order to ensure a guaranteed correct network interaction. The importance of solving such a problem is justified by the creation of modern information and cyber-physical systems, which are based on networks with adaptive topology. The dynamic nature of links between nodes, on the one hand, makes it possible to provide self-regulation of the network and, on the other hand, significantly complicates the control over the network operation, because it is impossible to identify a single pattern of network interaction. On the basis of the developed model of the network with adaptive topology, a graph algorithm for link prediction is proposed, which is extended to the case of peer-to-peer networks. The algorithm is based on the significant parameters of network nodes characterizing both their physical characteristics (signal level and battery charge) and their characteristics as objects of network interaction (characteristics of the centrality of graph nodes). The correctness and adequacy of the developed algorithm is confirmed by the experimental results on modeling a peer-to-peer network with adaptive topology and its self-regulation when different nodes are removed.