Location Protection Mechanism in Sparse Mobile CrowdSensing

As Internet of Things technology advances quickly, Sparse Mobile CrowdSensing has been used as a productive way to collect and process data. Leveraging the real time capabilities and wide coverage of Internet of Things communication, Sparse Mobile CrowdSensing utilizes distributed sensing nodes to collect and process real-time data, thereby supporting diverse intelligent services. In practical scenarios, users are often required to upload data carrying location information to support subsequent data completion. However, such data can expose sensitive user information, posing significant privacy risks. To mitigate these risks, location obfuscation techniques are employed to reduce the likelihood of privacy leakage. Nevertheless, these privacy-preserving measures can significantly degrade the quality of data completion. To address this challenge, we propose a scheme that integrates data correction to balance privacy protection and data completion quality. Specifically, our approach obfuscates location data using geographic indistinguishability and leverages historical data to obtain the data regularity, enabling modifications to the obfuscated data to ensure its usability. Our mechanism performs better than other baseline methods in terms of ensuring data completion accuracy while preserving a certain degree of privacy, according to evaluation on the real datasets.