Stealthy false data injection attacks with resource constraints against multi-sensor estimation systems.

Area of Science:

• Cyber-Physical Systems
• Control Theory
• Network Security

Background:

• Cyber-physical systems (CPS) rely on accurate sensor data for reliable operation.
• Limited resources in CPS can be exploited by adversaries to degrade performance.
• False data injection (FDI) attacks pose a significant threat to CPS estimation accuracy.

Purpose of the Study:

• To investigate methods for maximally degrading the estimation performance of a cyber-physical system under limited resources.
• To develop a stealthy false data injection (FDI) attack scheme targeting partial sensor channels.
• To derive an optimal attack strategy for maximizing estimation errors.

Main Methods:

• Formulating the stealthy attack as a constrained optimization problem.
• Deriving the attack stealthiness condition and compromised estimation error covariance.
• Developing an explicit solution for the optimal attack strategy.
• Analyzing the relationship between estimation error and attacked sensors.

Main Results:

• An explicit and proven solution for the optimal stealthy FDI attack strategy is provided.
• The study identifies the critical sensor channels to attack for maximum performance degradation.
• Numerical simulations validate the theoretical analysis and the effectiveness of the proposed attack.

Conclusions:

• The proposed FDI attack effectively degrades CPS estimation performance by strategically targeting sensors.
• The derived sensor selection principle guides optimal attack vector selection under resource constraints.
• This research highlights vulnerabilities in multi-sensor CPS estimation against sophisticated FDI attacks.