Decentralized optimal H_∞ fusion estimation for multi-sensor networked systems with two-channel hybrid attacks.

Area of Science:

• Control Systems Engineering
• Networked Systems Security
• Estimation Theory

Background:

• Networked systems are vulnerable to hybrid attacks (deception and denial-of-service) on communication channels.
• Ensuring reliable estimation in the presence of such attacks is critical for system performance and security.

Purpose of the Study:

• To develop a decentralized optimal H_∞ fusion estimation method for multi-sensor networked systems under hybrid attacks.
• To enhance the robustness and performance of estimation algorithms in insecure communication environments.

Main Methods:

• A novel augmentation method is used to model hybrid attacks using Bernoulli distributed random variables.
• Sufficient conditions for H_∞ performance and exponential mean-square stability are derived.
• Linear Matrix Inequality (LMI) techniques are employed to design the optimal fusion filter.

Main Results:

• The proposed method effectively handles simultaneous stochastic deception and denial-of-service attacks.
• The derived conditions guarantee the stability and H_∞ performance of the estimation error systems.
• The designed decentralized optimal H_∞ fusion filter achieves superior performance compared to local filters.

Conclusions:

• The developed decentralized optimal H_∞ fusion estimation strategy is effective against hybrid network attacks.
• The approach provides a robust solution for secure estimation in multi-sensor networked systems.
• Validation on a civil aircraft system confirms the practical applicability and effectiveness of the proposed algorithms.