Traffic Characterization of Event-Triggered Multiagent Systems Under FDI Attacks

Area of Science:

• Control Systems Engineering
• Cybersecurity
• Networked Systems

Background:

• Event-triggered systems offer efficiency but are vulnerable to cyberattacks.
• False data injection (FDI) attacks pose a significant threat to the stability and reliability of multiagent systems (MASs).

Purpose of the Study:

• To investigate the impact of multiplicative FDI attacks on the triggering behaviors of periodic event-triggered MASs.
• To develop a traffic model for characterizing triggering behaviors and analyzing attack effects on minimum interevent time (MIET).
• To propose a method for selecting sampling periods for anomaly detection under FDI attacks.

Main Methods:

• Developed an abstraction-based traffic model to characterize triggering behaviors, including MIET and interevent time (IET) transitions.
• Analyzed the influence of FDI attacks on MIET.
• Designed a behavior-based anomaly detection algorithm using the traffic model.

Main Results:

• The proposed traffic model effectively characterizes triggering behaviors under various initial states.
• Quantified the impact of FDI attacks on MIET.
• Demonstrated the effectiveness of the anomaly detection algorithm in identifying anomalous triggering behaviors caused by attacks.

Conclusions:

• The study provides a framework for understanding and mitigating FDI attacks in event-triggered MASs.
• The developed anomaly detection method is effective for practical applications in securing MASs.