Utilizing Spatio-temporal Data Index for Location Privacy Protection

Situation management systems are developing quickly nowadays. Among their vast applications, disaster management, homeland security and traffic management are of the most important ones. In domains above, the locations of people inside the observed areas are great sources of information. While we can put the accuracy of the location information first in emergency domains, such as disaster management, we need to care about users´ locations privacy in civil domains, such as traffic management. Thus, privacy-preserving techniques have gained special attention from researchers, such as obfuscation-based or k-anonymity. While existing solutions have integrated obfuscation-based into the indexed Spatio-temporal data to gain performance boost, the lack of reciprocity in these solutions grants the attackers ability to infer the exact users location by using some tricks, e.g. Query Sampling Attacks. Meanwhile, an important property of Hilbert Cloaking algorithm is reciprocity. This property ensures that any user in the k-Anonymizing Spatial Region has the same blurred region, thus the disadvantage of the above solutions is overcome. In this paper, we propose a new solution that combines Hilbert Cloaking algorithm and obfuscation-based technique to increase the privacy protection. Our solution is also integrated into Spatio-temporal data index structure so that it can be used for commercial Database Management Systems.