Home // CYBER 2023, The Eighth International Conference on Cyber-Technologies and Cyber-Systems // View article

Cyber Situational Awareness of Critical Infrastructure Security Threats

Authors:
Fatemeh Movafagh
Uwe Glässer

Keywords: Cyber-physical systems; supervisory control systems; online threat detection; infrastructure interdependencies; machine learning; anomaly detection; dynamic attack scoring.

Abstract:
The rising frequency and sophistication of cyberattacks pose a notorious threat to critical infrastructures, heavily reliant on industrial control systems for advanced automation. To explore this evolving challenge systematically, a robust cyber situational awareness framework is essential. Our paper adopts a dual approach, focusing on both the broader scope of threat mitigation and remediation to understand the breadth of the problem and on online intrusion detection applied to supervisory control data to comprehend its depth. The methodical framework and analytic model we propose here are tailored to cyber-physical systems used for industrial control and operational technology. By acknowledging transitional vulnerabilities in these systems, we stress the necessity of proactive measures to mitigate the risk of widespread cascading and escalating infrastructure failures. At the core of our contribution lies GenericAttackTracker, a novel analytic framework for online anomaly detection, which combines dynamic attack scoring with Bayesian inference to fuse results from supervisory control data analysis with real-time contextual information into actionable threat intelligence. By leveraging the abstract semantic properties of Heterogeneous Information Network Analysis for structural analysis and of Abstract State Machines for deriving executable abstract models of complex distributed systems, our framework supports a system of systems view of critical infrastructures and facilitates the daunting task of dynamically analyzing their intricate interdependencies.

Pages: 53 to 61

Copyright: Copyright (c) IARIA, 2023

Publication date: September 25, 2023

Published in: conference

ISSN: 2519-8599

ISBN: 978-1-68558-113-8

Location: Porto, Portugal

Dates: from September 25, 2023 to September 29, 2023