Differential privacy EV charging data release based on variable window.

Area of Science:

• Computer Science
• Electrical Engineering
• Cybersecurity

Background:

• Vehicle-to-Grid (V2G) networks generate valuable real-time data for mining.
• Direct data publication in V2G networks poses significant user privacy risks.
• Existing data release models often struggle to balance privacy protection with data utility.

Purpose of the Study:

• To propose a robust privacy protection mechanism for V2G data release.
• To ensure user privacy is maintained without compromising data utility.
• To develop an adaptive method for real-time V2G data privacy.

Main Methods:

• Implementation of a differential privacy mechanism tailored for V2G data.
• Definition and application of a variable sliding window for dynamic data adjustment.
• Consideration of sampling intervals and window proportions for privacy budget allocation.

Main Results:

• Experimental validation on real-world datasets demonstrates the effectiveness of the proposed method.
• The differential privacy approach significantly enhances data utility compared to baseline methods.
• The variable sliding window effectively balances privacy and utility.

Conclusions:

• The proposed differential privacy mechanism offers superior performance in V2G data release.
• The adaptive variable sliding window is key to improving data utility under privacy constraints.
• This research provides a practical solution for secure and useful V2G data sharing.