Distributed H_∞ filtering of replay attacks over sensor networks.

Area of Science:

• Control Systems Engineering
• Network Security
• Nonlinear System Analysis

Background:

• Distributed H∞ filtering is crucial for discrete-time nonlinear systems.
• Sensor networks are vulnerable to replay attacks, compromising system integrity.
• Existing methods often lack robustness against sophisticated, time-varying attack strategies.

Purpose of the Study:

• To develop a secure distributed H∞ filtering approach for discrete-time nonlinear systems under replay attacks.
• To model the temporal behavior of replay attacks using an indicator variable and a novel pattern.
• To derive conditions for guaranteed H∞ performance despite adversarial actions.

Main Methods:

• Introduced an indicator variable to model replay attack occurrences.
• Developed a time-varying attack pattern based on three parameters.
• Transformed the filter dynamics into a switching system with time-varying delays.
• Utilized switching system theory to derive H∞ performance conditions.
• Employed matrix inequalities to compute filter gains.

Main Results:

• A sufficient condition for guaranteed H∞ performance was derived, revealing tolerant attack conditions (duration and proportion).
• The developed filtering strategy effectively mitigates the impact of replay attacks.
• Applicable filter gains were successfully calculated using matrix inequality solutions.

Conclusions:

• The proposed secure filtering strategy is effective for discrete-time nonlinear systems facing replay attacks in sensor networks.
• The method provides a robust framework for analyzing and guaranteeing system performance under specific attack constraints.
• The illustrative example confirms the practical availability and efficacy of the developed secure filtering approach.