COLLABORATIVE RESEARCH: CT-ISG - A Context-Aware Approach to the Design and Evaluation of Privacy Preservation Techniques in Location-Based Services

基于位置的服务（LBS）带来的隐私保护挑战对用户和服务提供商都至关重要。本项目致力于设计和评估LBS中的隐私保护技术。该项目与现有研究的重要区别在于它强调了请求上下文的作用。上下文是指攻击者可能使用的外部信息/知识，以及请求本身，以获取用户隐私信息。例如，利用用户在特定时间的大致位置的外部知识，攻击者可以挑出特定LBS请求的用户，从而将请求中的私人信息链接到用户。就其性质而言，请求与请求之间的上下文变化以及不同的上下文可能需要不同的隐私保护技术。因此，该项目的技术目标是（1）系统地分类隐私上下文，（2）分析现有的防御策略，（3）设计和评估新的防御策略。从教育的角度来看，该项目涉及研究生，使他们接触到前沿研究，并将研究成果纳入课堂。此外，该项目还为两个机构更广泛的研究人员之间的积极合作提供了一个平台。该项目的结果将对保护用户隐私和人们愿意采用LBS来改善他们的生活和工作条件产生积极影响。研究结果将通过技术报告、会议出版物和期刊以及http://csis.gmu.edu/NSFLBSprivacy和http://www.cs.uvm.edu/~xywang/NSFLBSprivacy网站传播。

Privacy protection challenges arising from location-based services (LBS) are critical to users as well as service providers. This project concentrates on designing and evaluating privacy protection techniques in LBS. The important departure of this project from the existing research is in its emphasis of the role of request contexts. A context refers to the external information/knowledge that the attacker may use, together with the requests themselves, to gain user private information. For example, with the external knowledge of a user's approximate location at a particular time, the attacker may single out the user of a particular LBS request and thus link the private information in the request to the user. By its nature, context changes from requests to requests and different contexts may call for different privacy protection techniques. The technical objectives of the project are therefore to (1) systematically categorize privacy contexts, (2) analyze existing defense strategies, and (3) design and evaluate new defense strategies. From an educational perspective, the project involves graduate students and exposes them to leading-edge researches, and incorporates research results into classrooms. In addition, the project provides a platform for active collaborations among a broader set of researchers of the two institutions. The results from this project will have a positive impact on protecting user privacy and on people's willingness to adopt LBS in enhancing their living and working conditions. Research results will be disseminated through technical reports, publications at conferences and journals, and the websites at http://csis.gmu.edu/NSFLBSprivacy and http://www.cs.uvm.edu/~xywang/NSFLBSprivacy.