Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

This thesis by Chih-Yuan Lin explores network-based anomaly detection methods for Supervisory Control and Data Acquisition (SCADA) systems, which are crucial for monitoring and controlling infrastructure like power plants and water facilities. With the increasing integration of open protocols and internet connectivity, SCADA systems face heightened vulnerability to sophisticated cyber-attacks, including zero-day exploits. The work focuses on enhancing SCADA security through anomaly detection that models normal traffic patterns and identifies deviations. The study categorizes SCADA traffic into polling and spontaneous types, proposing statistical and machine learning approaches to detect anomalies, achieving high detection rates with minimal false positives. Aimed at researchers and professionals in cybersecurity and SCADA systems, the thesis contributes to developing resilient monitoring systems in critical infrastructures.

Intrusion detection systems (Computer security) --- Supervisory control systems.