A Hilbert-based framework for preserving privacy in location-based services

摘要

Preserving user's privacy has recently drawn special attention in the field of location-based services and many techniques such as k-anonymity or obfuscation have been suggested to protect user's privacy. All of these traditional techniques are, however, geometry-based and separated from the database level. This separation causes the query processing to involve in two phases, querying the database to retrieve the exact locations of users and then modifying them to decrease the quality of this information. This two-phase process is time-consuming due to the number of disk accesses required to retrieve the user's exact location. Also, these geometry-based techniques cannot guarantee location privacy when the adversary gains knowledge about the geography of the obfuscated region. We address these problems by proposing Hilbert-based framework for preserving user's privacy and B~(ob)-tree for supporting geographic-aware obfuscation. Experiments and analyses show that this framework provides a significant improvement over the algorithm separated from the database level for both query processing time and location privacy protection.