An Effective Location Privacy-Preserving K-anonymity Scheme in Location Based Services

With the development of global positioning systems, communication technologies, the Internet and mobile terminals, location-based services (LBS) have become increasingly widely used. LBS brings great convenience to people's daily life and social activities. At the same time, people pay more attention to the problem of sensitive information leakage when using LBS. Users need to provide their accurate location information when sending queries to the LBS server. User location privacy is extremely vulnerable to serious threats during the exchange of services, and these location-related queries may cause serious privacy issues. In this paper, considering that the opponent may use auxiliary information such as data analysis and crawlers to determine the approximate location, we propose an effective location privacy-preserving k-anonymity scheme (PPKS) to generate virtual locations based on the probability density function in mathematical statistics to realize the k-anonymity of users with privacy zone awareness in LBS. We first determine the origin of the region and then calculate the probability density function of the X-axis and the Y-axis. Subsequently, the k-1 positions are distributed to the X-axis and Y-axis in proportion, respectively. Finally, according to the proportions, the k-1 virtual positions are generated. The results of security analysis and experimental evaluation show that the solution can significantly improve the level of privacy and anonymity.