Optimal Stealthy Attacks With Historical Data in Cyber-Physical Systems Under Periodic Detection

Area of Science:

• Cyber-Physical Systems Security
• Control Theory
• Optimization Techniques

Background:

• Cyber-physical systems (CPSs) are vulnerable to sophisticated cyber attacks.
• Mismatched control and detection cycles in CPSs pose unique challenges for designing effective defense and attack strategies.
• Existing stealthy attack strategies often involve complex numerical methods.

Purpose of the Study:

• To formulate and design novel stealthy attack strategies for CPSs with mismatched control and detection cycles.
• To develop a method that enforces strict stealthiness at detection instants and applies energy constraints at non-detection instants.
• To simplify the optimization problem for deriving optimal attack parameters, avoiding iterative procedures.

Main Methods:

• Formulation of a linear attack strategy based on historical innovation sequences.
• Exploitation of relationships among stealthiness-related parameters to reformulate a nonconvex optimization problem.
• Derivation of closed-form optimal attack parameters through a single equality constraint reformulation.

Main Results:

• The proposed method successfully formulates stealthy attack strategies for CPSs with mismatched cycles.
• The reformulated optimization problem allows for direct derivation of closed-form optimal attack parameters.
• Achieved degradation in state-estimation performance surpasses that of existing strategies.

Conclusions:

• The developed approach offers an effective and computationally efficient method for designing stealthy attacks in CPSs.
• The strategy's ability to outperform existing methods highlights its potential for security research and vulnerability analysis.
• Numerical simulations confirm the effectiveness and advantages of the proposed attack formulation.