A software model for the formation of network nodes virtual clusters in mobile transport networks

A. A. Bryukhovetskiy ; D. V. Moiseev

Taurida Journal of Computer Science Theory and Mathematics, no. 2 (2023), pp. 50-59

Résumé

Ensuring the security of nodes in mobile networks, creating a universal model for detecting vulnerabilities in vehicle interfaces operating in an external stochastic, highly dynamic environment and in conditions of a lack of a priori information about the behavior of devices is associated with insoluble problems for constructing formal models and methods for determining reliable characteristics of devices. The advent of intelligent transport systems such as intelligent driving, collaborative travel and mobile payments will contribute to the rapid development of the intelligent transport industry. The Automotive Peer To Peer Network (VANET) is considered the main system for deploying applications based on it that contribute to improving the quality of driving, inform drivers about road hazards, etc. In practice, solving problems related to processing the characteristics of highly dynamic objects in a stochastic environment requires large computational costs. As a rule, with the practical interest of obtaining reliable estimates, researchers are faced with the dimensions of problems that belong to the class of NP-hard ones. Solving them is difficult in real time, as it requires a complete search of options. For most practical problems this is unacceptable due to the large dimension, limited time and insufficient resources. The proposed approach is aimed at increasing the reliability of decisions made in a stochastic highly dynamic environment characterized by a rapidly changing network topology, its mobility, spatial density, and node localization. The method is focused on the formation of clusters with high stability, stable communication, and low message transmission latency. Cluster coordinators are selected based on the trust level of the vehicle. The article presents a software model for the formation of virtual clusters of network nodes, which allows you to switch from processing a rapidly changing network topology structure to a quasi-permanent one. One of the variants of the network architecture, a graph model of virtual clusters, is presented, an algorithm for the formation of minimal coverages is given. The results of modeling of the program system are presented. The model is implemented by means of the language JAVA 17, OS Windows 10, occupied volume 176 KB.

Détail

Keywords: decentralized processing, virtual clusters, graph model, adjacency matrix, set coverage.