Dual-differential privacy-preserving under switching control: A novel approach to improve privacy

초록

Differential privacy methods have received attention for addressing data privacy in dynamic systems. Designing a novel differential privacy mechanism to achieve higher privacy and exploring the influence of dynamic system properties on privacy are vital issues. This paper investigates a novel differential privacy-preserving mechanism under switching control. System mode switching is developed to improve data privacy, and controllers are designed to weaken the effects of privacy noise on system stability. (i) A dual-differential privacy-preserving (DDPP) mechanism combining global privacy and local privacy is applied to improve the data privacy of dynamic systems. (ii) Implementing local privacy in the model output signal provides higher data availability. Then, more accurate closed-loop models can be obtained than global privacy, and control accuracy can be improved. (iii) A mode-switching law is presented to achieve higher privacy. Moreover, dual-dwell time bounds that ensure the privacy and stability of the privacy-preserving system under switching control are presented. Finally, a numerical simulation is provided to verify the effectiveness of the proposed method. (c) 2025 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

키워드