Partial-encryption-decryption-based secure state estimation of singularly perturbed complex networks: A Paillier...

Area of Science:

• Control Systems Engineering
• Network Security
• Information Theory

Background:

• Secure state estimation is crucial for complex networks transmitting data over open channels.
• Existing methods face challenges in balancing data security with computational efficiency.
• Perturbations in estimator gains and measurement quantization can degrade estimation performance.

Purpose of the Study:

• To develop a secure and resilient state estimation technique for discrete-time singularly perturbed complex networks.
• To propose a novel Paillier-based partial encryption-decryption (PED) mechanism for enhanced data protection.
• To design an estimator that robustly handles gain perturbations and quantization errors.

Main Methods:

• Integration of probabilistic measurement quantization with Paillier homomorphic encryption for partial encryption-decryption (PED).
• Development of a group-based round-robin protocol for selecting measurement subsets for encryption.
• Application of Lyapunov stability theory to derive conditions for estimation error boundedness.
• Characterization of estimator gains using matrix inequalities for a tractable design.

Main Results:

• The proposed PED mechanism achieves a tradeoff between data security and computational efficiency.
• Sufficient conditions are derived to guarantee exponential ultimate boundedness of estimation errors in the mean-square sense.
• A computationally tractable procedure for designing the resilient state estimator is presented.

Conclusions:

• The novel secure state estimation scheme effectively enhances data security while maintaining computational efficiency.
• The designed estimator demonstrates robustness against estimator gain perturbations and quantization errors.
• Numerical validation confirms the effectiveness and resilience of the proposed approach for complex networks.