Secure Estimation With Privacy Protection.

Area of Science:

• Control Systems Engineering
• Information Security
• Computer Science

Background:

• State estimation systems often face challenges in protecting sensitive user action data.
• Existing optimal estimators may be computationally infeasible for real-world privacy-preserving applications.

Purpose of the Study:

• To develop a privacy-preserving mechanism (PPM) for state estimation systems.
• To design a computationally efficient suboptimal estimator (SE) that maintains user privacy and estimation performance.
• To establish guidelines for balancing privacy and performance based on user needs.

Main Methods:

• Proposed a privacy-preserving mechanism (PPM) to obscure sensitive user actions.
• Derived an optimal estimator (OE) and subsequently designed a computationally efficient suboptimal estimator (SE).
• Utilized a privacy-preserving optimization problem to define performance-privacy tradeoffs.

Main Results:

• The proposed suboptimal estimator (SE) is proven to be stable.
• The SE effectively satisfies user requirements for both privacy protection and estimation accuracy.
• A framework for customizing the privacy-performance balance was established.

Conclusions:

• The developed suboptimal estimator offers a practical solution for privacy-preserving state estimation.
• The study provides a method to quantitatively manage the tradeoff between data privacy and estimation performance.
• Illustrated examples validate the theoretical findings and the effectiveness of the proposed approach.