A novel ε-sensitive correlation indistinguishable scheme for publishing location data.

Area of Science:

• Computer Science
• Information Security
• Mobile Computing

Background:

• Location-based services (LBS) generate extensive data, offering benefits but posing privacy risks.
• Existing k-anonymity methods are insufficient due to background knowledge enabling location inference.
• Sensitive correlations in location data can be exploited, compromising user privacy.

Purpose of the Study:

• To propose a novel privacy protection scheme for location data that addresses the limitations of existing methods.
• To ensure correlation indistinguishability for location data, safeguarding against sensitive information leakage.
• To develop a privacy-preserving method that maintains the utility of published location data.

Main Methods:

• Utilized entropy to determine location centroids and construct Voronoi diagrams.
• Transformed untreated location data coordinates to their respective cell centroids.
• Employed ε-sensitive correlation privacy metrics for rigorous mathematical justification of data transformation.

Main Results:

• The proposed scheme effectively destroys sensitive correlations in location data.
• Security analysis confirms the enhanced privacy capabilities of the algorithm.
• Comparative experiments validate the algorithm's advantage in location data privacy protection and data availability.

Conclusions:

• The ε-sensitive correlation privacy protection scheme offers superior privacy compared to existing methods.
• The integration of Voronoi diagrams and entropy provides a robust framework for location data anonymization.
• The proposed method balances privacy preservation with the practical utility of location-based data.