FADBM: Frequency-Aware Dummy-Based Method in Long-Term Location Privacy Protection
Chapter
Accepted version
Åpne
Permanent lenke
http://hdl.handle.net/11250/2642467Utgivelsesdato
2020Metadata
Vis full innførselSamlinger
Originalversjon
https://doi.org/10.1109/ICPADS47876.2019.00060Sammendrag
With the rapid usage of location-based services (LBSs), protection of location privacy has become a significant concern. Existing dummy-based methods mainly consider generating dummies in continuous queries, which neglects the fact that the user would launch queries in a frequent region(e.g., home), resulting in privacy disclosure in a long-term period (i.e., more than 30 days). To solve this problem, we propose a method which generates dummy frequent regions, and make dummies locate in these dummy regions as far as possible. Compared with other methods, evaluation based on real-world dataset shows that the proposed method can reduce the ratio of recognized dummies and restored trajectory in a long-term situation.