Secure Fusion Estimation for Bandwidth Constrained Cyber-Physical Systems Under Replay Attacks.

Area of Science:

• Cyber-Physical Systems (CPSs)
• Control Theory
• Information Fusion

Background:

• State estimation is critical for CPS monitoring and supervision.
• Security and estimation performance are key concerns in CPS.
• Multisensor Information Fusion Estimation (MIFE) offers improved accuracy and robustness against attacks.

Purpose of the Study:

• To investigate secure distributed Kalman fusion estimation for CPSs under replay attacks.
• To address challenges posed by bandwidth-constrained communication channels.
• To develop a robust estimation framework from a defender's perspective.

Main Methods:

• A novel mathematical model incorporating a compensation strategy for replay attacks and bandwidth constraints.
• Design of a recursive distributed Kalman fusion estimator (DKFE) using linear minimum variance principles.
• Development of data compression and compensation algorithms tailored to different communication frameworks.

Main Results:

• The proposed DKFE achieves desired performance under bandwidth constraints.
• Attack-dependent and bandwidth-dependent conditions for ensuring estimator security were derived.
• The effectiveness of the developed methods was demonstrated through an illustrative example.

Conclusions:

• The proposed methods provide a secure and effective solution for distributed Kalman fusion estimation in CPSs.
• The framework enhances reliability and robustness against replay attacks and communication limitations.
• This work contributes to the secure operation and supervision of cyber-physical systems.