Observer-Based Asynchronous Stabilization for Networked Systems With Multichannel Attacks and Applications.

Area of Science:

• Control Systems Engineering
• Network Security
• Stochastic Systems

Background:

• Networked systems (NS) are vulnerable to multichannel attacks.
• Asynchronous phenomena, like controller-mode/attack-mode mismatches, complicate stabilization.
• Accurate modeling of complex, time-varying attack behaviors is challenging.

Purpose of the Study:

• To develop an observer-based asynchronous stabilization method for NS under multichannel attacks.
• To address the challenge of inaccessible actual attack modes.
• To establish a stability criterion for systems under complex attack scenarios.

Main Methods:

• Utilized a piecewise homogeneous semi-Markov chain (SMC) model modulated by a superstratum Markov chain to capture attack dynamics.
• Designed an observer-based mode switching delay technique to handle unknown attack modes.
• Employed a Lyapunov function dependent on observed modes, piecewise homogeneous variables, and ET for stability analysis.
• Applied matrix decoupling and convexification to reduce computational complexity.

Main Results:

• Established a sufficient criterion for ς-error mean-square stability under random multichannel Denial-of-Service (DoS) attacks.
• Demonstrated reduced computational complexity through matrix decoupling and convexification.
• Validated the proposed method's effectiveness via two practical simulation cases.

Conclusions:

• The proposed observer-based asynchronous stabilization method effectively ensures system stability under multichannel attacks.
• The advanced SMC model accurately depicts complex attack behaviors.
• The technique provides a robust solution for securing networked systems against sophisticated cyber threats.