Ph.D. Preliminary Exam - Ge Xu

Date: 28 Jan, 2009
Time: 8:00 AM
Location: 223 Atanasoff Hall
Topic: Location Cloaking for Location Privacy and Safety Protection
Major Professor(s): Ying Cai


Abstract:

Many applications today rely on location information of users or devices. Without appropriate measures, disclosing such information can present heightened privacy and security risks. A person's whereabouts, for example, may imply sensitive private information such as medical condition, lifestyle, or political association. Location information also has the potential to allow an adversary to physically locate a subject, which is particularly concerned in digital battlefields where the exposure of one's location can lead to fatal consequences. Our research has two goals: (1) Location privacy protection in location based services (LBSs), which aims to prevent adversary from deriving who's where at what time; (2) Location safety protection in ad hoc networks, which aims to prevent adversary locating an entity of interest. In our preliminary work, for the former, we propose using historical location data to protect users' location privacy. When requesting a service, a user's location is revealed as a spatial region that has been visited by a number of users. An adversary may be able to identify all these people, but will not know who was there when the service was requested. For the latter, we define the safety level of a spatial region as the reciprocal of the density of nodes inside, and the location information revealed by a node must be a region with a safety level high enough. For both of the two research problems, we develop effective location cloaking algorithms which enable the users or devices disclose their location with reduced resolution for applications while ensuring such information cannot be used by adversaries to compromise their privacy or safety. In our future work, we plan to make some enhancements based on our current location cloaking techniques. For the location privacy protection, we will investigate the problem of location cloaking against some new attack models; For the location safety protection, we will investigate the impact of the cloaked location on existing geographic routing protocols and if new routing techniques can be developed for better performance by taking this factor into consideration in the design.