APPROACH TO CLASSIFICATION OF THE INFORMATION SECURITY STATE OF ELEMENTS FOR CYBER-PHYSICAL SYSTEMS BY APPLYING SIDE ELECTROMAGNETIC RADIATION

• Published in: Nauchno-tekhnicheskiĭ vestnik informat͡s︡ionnykh tekhnologiĭ, mekhaniki i optiki Vol. 18; no. 1; p. 98
• Main Authors: Semenov, V V, Lebedev, I S, Sukhoparov, M E
• Format: Journal Article
• Language: Russian
• Published: Saint Petersburg St. Petersburg National Research University of Information Technologies, Mechanics and Optics 01.01.2018
• Subjects: Anomalies; Complex systems; Cyber-physical systems; Data analysis; Data transfer (computers); Decision making; Electromagnetic radiation; Electronic components; Electronic devices; Machine learning; Monitoring; Quality assessment; Statistical analysis
• ISSN: 2226-1494; 2500-0373
• DOI: 10.17586/2226-1494-2018-18-1-98-105

We consider problematic issues of information security for cyber-physical systems. The analysis of the characteristics of autonomous objects has been carried out. The paper presents a model for the information security monitoring system based on the characteristics obtained as a result of the electromagnetic radiation analysis of the electronic components of the autonomous devices in cyber-physical systems. A typical scheme for the system state determination is shown. For reasons of the device features providing the infrastructure, information security state assessment is aimed at the analysis of the system normal functioning, rather than the search for signatures and characteristics of anomalies during various types of information attacks. An experiment is described providing statistical information on the operation of remote devices of cyber-physical systems where data accumulation for decision-making occurs by statistical information comparison. The experiment results on information impact on a typical system are presented. The proposed approach for the statistical data analysis of autonomous devices based on a naive Bayesian classifier can be used to determine the information security states. A special feature of our approach is the ability for quick adaptation and application of various mathematical apparatus, and machine learning methods to achieve given quality of probabilistic assessment of information security state. The implementation of this type of monitoring does not require the development of complex system applications. It allows for implementation of various system building architectures processing the autonomous object on-board or data transfer and state calculation on external computer nodes of monitoring and control systems.