|
Rapid advances in location based applications are leading to increased concern about location privacy. Current mobile operating systems only provide users with rudimentary location access controls - either to block or allow location access - which are inadequate and inefficient in mitigating privacy threats. Most existing location obfuscation mechanisms are based on syntactic privacy models that do not consider mobility and are hence vulnerable to inference attacks. The goal of this project is to develop rigorous and customizable privacy notions and location perturbation mechanisms for individual location sharing in location based applications, and study their impact in the context of location based queries and geospatial crowdsourcing. Building privacy mechanisms that can be directly used by mobile users will minimize unnecessary disclosure of their precise locations and trajectories to location based applications and other possibly malicious third parties. By addressing the spatiotemporal privacy barrier, this project promises significant impact in enabling and promoting adoption of location based applications while ensuring individual privacy. The project also includes a set of integrated educational activities including hackathon events, involvement of graduate and undergraduate students, and encouragement of participation by women and minorities.
The project has several objectives: 1) Develop rigorous privacy notions based on differential privacy and its variants for location protection while accounting for spatiotemporal correlations of a user's locations through mobility modeling, and to investigate optimal and heuristic perturbation mechanisms to achieve privacy guarantees with optimal utility; 2) Extend privacy solutions for protecting a trajectory or a sequence of time-location pairs; and 3) Develop customizable privacy solutions for location and trajectory protection through policy graphs, considering user preferences, spatiotemporal semantics, and privacy-utility tradeoffs. Privacy-utility tradeoffs using perturbed locations are studied in the context of location based applications such as predictive range queries and k-nearest neighbor queries as well as geospatial crowdsourcing.
More information on this project can be found at http://www.mathcs.emory.edu/aims/stprivacy
|
TWC: Small: Rigorous and Customizable Spatiotemporal Privacy for Location Based Applications