Asynchronous Control of Cyber-Physical Systems With Quantized Measurements and Stochastic Multimode Attacks.

Area of Science:

• Control Systems Engineering
• Cyber-Physical Systems Security
• Stochastic Processes

Background:

• Cyber-physical systems (CPS) face security challenges due to open networks and band-limited communication.
• Attacks on CPS can be multi-modal and stealthy, making detection difficult.
• Observer-dependent asynchronous control is crucial for system stability under adversarial conditions.

Purpose of the Study:

• To develop a robust control strategy for CPS against observer-dependent asynchronous attacks.
• To address the challenge of stealthy, multi-mode attacks in networked systems.
• To design a controller that operates effectively despite information loss and unknown attack modes.

Main Methods:

• Quantization of measured outputs for band-limited channels.
• A multichannel transmission framework with semi-Markov chain dynamics for attack modes.
• A dual-layer stochastic process (hidden semi-Markov jump mode) to model stealthy, multi-mode attacks.
• Design of an observed-mode-dependent controller using emission probabilities.

Main Results:

• The proposed controller effectively stabilizes CPS under stealthy, multi-mode attacks.
• The methodology accommodates data quantization and information loss.
• Simulations demonstrate the controller's feasibility on unmanned ground vehicle and mass-spring-damper systems.

Conclusions:

• The developed observer-dependent asynchronous control strategy enhances CPS resilience against sophisticated attacks.
• The hidden semi-Markov jump mode framework provides a robust approach for analyzing complex attack scenarios.
• The research offers a practical solution for securing CPS in adversarial environments.