NN-Based Event-Triggered Protocol for NCSs Under DoS and Unknown Deception Attacks

Area of Science:

• Control Systems Engineering
• Cybersecurity
• Artificial Intelligence

Background:

• Networked control systems (NCSs) are vulnerable to cyber attacks like Denial-of-Service (DoS) and Deception Attacks (DAs).
• Resource constraints in NCSs necessitate efficient control protocols.
• Ensuring system stability under adversarial conditions is critical for reliable operation.

Purpose of the Study:

• To develop a robust control strategy for NCSs against combined DoS and DA.
• To enhance the resilience and stability of NCSs using intelligent control methods.
• To address resource limitations while maintaining system security.

Main Methods:

• A neural network (NN)-based resilient event-triggered control protocol (RETCP) was designed.
• NNs were utilized to neutralize and approximate malicious data from DAs.
• A novel predictor was developed to compensate for signal loss during DoS attacks.

Main Results:

• The proposed NN-based RETCP effectively mitigates the impact of DoS and DA.
• The developed predictor enhances NCS tolerance to complex cyber threats.
• The intelligent control protocols achieve input-to-state stability (ISS) under combined attacks.

Conclusions:

• The NN-based controller ensures ISS for NCSs facing sophisticated cyber attacks.
• The proposed RETCP offers a viable solution for securing NCSs in challenging environments.
• Experimental validation on an uncrewed ground vehicle (UGV) confirms the protocol's efficacy.