Client-centric Privacy Protection For Location-based Services Public
Contenu téléchargeableTélécharger le fichier PDF
This dissertation addresses issues related to privacy protection in Location-based Services (LBS). LBS has become an immensely valuable source of real-time information and guidance. Nonetheless, the potential abuse of users' sensitive personal data by an LBS server itself is evolving into a serious concern. Privacy concerns in LBS exist on two fronts, namely i) Location Privacy, i.e. protection of location information and ii) Query Privacy, i.e. protection of service attribute(s). Although distinct, location privacy and query privacy are very closely related, which results in even greater challenges to privacy protection.Most existing research in this field either requires a trusted third-party (anonymizer) or uses oblivious protocols that are computationally and communication-wise expensive. The design of privacy-preserving techniques described in this dissertation is principled on not requiring a trusted third-party, i.e. they are completely client-centric, while being highly efficient in terms of time and space complexities.In particular, this dissertation introduces four client-centric privacy protection systems to address four different types of privacy concerns, respectively: i) CAP, for location privacy protection in snapshot LBS queries, ii) BACK-TRACK, for location privacy protection in continuous LBS queries, iii) DUMMY-Q, for query privacy protection in snapshot and continuous LBS queries, and iv) Digital Marauder's Map - a system studying threats posed to location privacy by a local adversary. For each system, theoretical analysis and extensive experimental studies are presented to demonstrate its effectiveness on privacy protection, its ability to maintain the utility of LBS, and its efficiency in terms of time and space.
|La vignette||Titre||Date de téléchargement||Visibilité||actes|