Joint Estimation of States, Sparse Sensor Attacks, and Unknown Inputs in Discrete-Time Cyber--Physical Systems

Area of Science:

• Control Systems Engineering
• Cybersecurity
• Networked Systems

Background:

• Cyber-physical systems (CPSs) are vulnerable to sparse attacks and unknown disturbances, compromising their operational integrity.
• Secure state estimation (SSE) is crucial for maintaining CPS stability and performance under adversarial conditions.
• Existing methods often struggle with dynamic attacks and real-time estimation accuracy.

Purpose of the Study:

• To develop a robust SSE method for discrete-time CPSs resilient to sparse attacks and unknown disturbances.
• To achieve disturbance decoupling and accurate real-time state estimation.
• To provide a theoretical framework and practical validation for the proposed estimation techniques.

Main Methods:

• Constructing an augmented system incorporating disturbances and attacks into the state vector.
• Deriving a necessary and sufficient condition for the augmented system's observability.
• Designing a Luenberger-like observer with a projection operator for sparse attack estimation.
• Developing a state reconstruction method leveraging augmented system observability.

Main Results:

• The proposed observer achieves asymptotic estimation of system states, disturbances, and attack signals.
• A condition for augmented system observability is established.
• The method effectively estimates current-time system states, outperforming existing approaches under dynamic sparse attacks.
• Demonstrated effectiveness and superiority through two illustrative examples.

Conclusions:

• The developed SSE method provides robust and accurate real-time state estimation for discrete-time CPSs.
• The approach effectively decouples disturbances and handles sparse, dynamically switched attacks.
• The findings offer significant advancements in securing CPSs against sophisticated cyber threats.