位置服务中的查询隐私度量框架研究  

A Quantifying Framework of Query Privacy in Location-Based Service

作　　者：张学军[1,2] 桂小林[1] 冯志超[1] 田丰[1] 余思[1] 赵建强[1] 

机构地区：[1]西安交通大学电子与信息工程学院,西安710049 [2]兰州交通大学电子与信息工程学院,兰州730070

出　　处：《西安交通大学学报》2014年第2期8-13,37,共7页Journal of Xi'an Jiaotong University

基　　金：国家自然科学基金资助项目(61172090;61163010);国家高技术研究发展计划重大专项资助项目(2012ZX03002001);教育部高等学校博士学科点专项科研基金资助项目(20120201110013);陕西省科技发展计划"十五"攻关资助项目(2012K06-30)

摘　　要：针对查询隐私度量机制存在过高评价用户隐私保护水平的问题,提出一个泛化的查询隐私度量框架。该框架将影响用户查询隐私的各种因素放在一起统一考虑,形式化定义用户、攻击者、隐私保护机制和隐私度量指标,提供了一种融合攻击者背景知识和推理能力的系统的隐私度量方法,可在相同条件下正确度量多种查询隐私保护机制的有效性,帮助用户选取合适的隐私需求以获得隐私保护和服务质量之间的平衡。在由Thomas Brinkhoff路网数据生成器生成的数据集上进行的模拟实验验证了该框架的有效性和准确性。A generalized quantification framework for evaluating user query privacy is proposed to address the overestimation of user＇s privacy protection level offered by query privacy protection mechanism in location-based services. The framework is to comprehensively take the various elements and relations that influence the query privacy of mobile users into account together and to formally define the mobile user, adversary, privacy protection mechanism and privacy metrics which reflect the user＇s privacy and service quality requirements. Moreover, the framework provides a systematic approach with integrating adversary＇s background knowledge available to and reasoning abilities in privacy quantification. The method can correctly evaluate the effectiveness of various query privacy protection mechanisms under the same conditions, and can help user to select appropriate privacy requirements to achieve a right tradeoff between privacy protection and service quality. The effectiveness and accuracy of the framework are verified through a set of experiments on datasets generated by network-based generator of moving objects proposed by Thomas Brinkhoff.

关 键 词：基于位置的服务 查询隐私 隐私度量 隐私保护机制 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]