Node Location Privacy Protection Based on Differentially Private Grids in Industrial Wireless Sensor Networks.

Area of Science:

• Computer Science
• Data Privacy
• Network Security

Background:

• Wireless sensor networks (WSNs) are integral to Industry 4.0, enabling data collection and analysis via cloud platforms.
• Data analysis in WSNs raises privacy concerns, particularly regarding sensitive location information.
• Existing differentially private methods lack detailed analysis of data domain partitioning granularity.

Purpose of the Study:

• To propose a novel differentially private data domain partitioning model for WSNs.
• To enhance the accuracy of geospatial data release in privacy-preserving systems.
• To address limitations in existing methods concerning partition granularity and error analysis.

Main Methods:

• Developed a data domain partitioning model considering noise and non-uniformity errors.
• Introduced a uniform grid release method based on the proposed partitioning model.
• Implemented a cell merging technique to further minimize errors before noise addition.

Main Results:

• The proposed method demonstrates significantly improved query accuracy compared to existing approaches.
• The data domain partitioning model provides a more accurate selection of grid sizes.
• Cell merging effectively reduces overall errors in differentially private data release.

Conclusions:

• The novel partitioning model and uniform grid release method enhance differential privacy for WSN geospatial data.
• The approach offers a more accurate and effective solution for privacy-preserving data analysis in industrial applications.
• This work contributes to more secure and reliable data utilization in the era of Industry 4.0.