INFORMATION SECURITY OF CYBER-PHYSICAL SYSTEMS: AUTHENTICATION OF DIGITAL TWINS

Abstract

Formulation of the problem: the main catalysts for the development of cyber-physical systems are currently the growth of artificial intelligence and the creation of digital twins that have a complex vertical structure and exchange data for joint learning. At the same time, the empowerment of digital twins as data owners can lead to critical consequences in the field of ensuring the security of data systems. The development of evolutionary methods for ensuring information security, and in particular, methods for authenticating digital twins, is a fundamental issue on the way to the development of cyber-physical systems. Objective: analysis of aspects and principles of building a system and the process of authenticating digital twins in dynamic and scalable cyber-physical systems, organizing the process under study, indicators of its effectiveness and criteria for their evaluation. Methods used: system analysis, tuple algebra, methods for designing and evaluating the efficiency of complex systems. Novelty: the use of a multi-agent structure of the digital twin authentication system, which makes it possible to achieve guaranteed awareness of the security status of the system as a whole and respond appropriately in case of compromising events. The implementation of intelligent authentication management is proposed to be carried out using the application capabilities of tuple algebra, which takes into account differences in the structures of traditional and intelligent systems, as well as the difficulties of parallelization in distributed systems. To increase the stability of the multi-agent authentication system, the possibility of using cryptocode protocols is considered, which makes it possible to ensure the restoration of reliable authentication data in case of failures. Result: substantiation of new principles and technological solutions in the field of high-level design of cyber- physical systems.