Degree Type

Dissertation

Date of Award

2010

Degree Name

Doctor of Philosophy

Department

Computer Science

First Advisor

Ying Cai

Abstract

Many applications today rely on location information, yet disclosing such information can present heightened privacy and safety risks. A person's whereabouts, for example, may reveal sensitive private information such as health condition and lifestyle. Location information also has the potential to allow an adversary to physically locate and destroy a subject, which is particularly concerned in digital battlefields.

This research investigates two problems. The first one is location privacy protection in location-based services. Our goal is to provide a desired level of guarantee that the location data collected by the service providers cannot be correlated with restricted spaces such as home and office to derive who's where at what time. We propose 1) leveraging historical location samples for location depersonalization and 2) allowing a user to express her location privacy requirement by identifying a spatial region. With these two ideas in place, we develop a suite of techniques for location-privacy aware uses of location-based services, which can be either sporadic or continuous. An experimental system has been implemented with these techniques. The second problem investigated in this research is location safety protection in ad hoc networks. Unlike location privacy intrusion, the adversary here is not interested in finding the individual identities of the nodes in a spatial region, but simply wants to locate and destroy them. We define the safety level of a spatial region as the inverse of its node density and develop a suite of techniques for location safety-aware cloaking and routing. These schemes allow nodes to disclose their location as accurately as possible, while preventing such information from being used to identify any region with a safety level lower than a required threshold. The performance of the proposed techniques is evaluated through analysis and simulation.

Copyright Owner

Ge Xu

Language

en

Date Available

2012-04-30

File Format

application/pdf

File Size

100 pages

Share

COinS