摘要
针对已有位置k匿名方法不能很好地权衡位置隐私和服务可用性之间的关系、易于遭受背景知识推理攻击、资源开销大的问题,提出了一种抵御推理攻击的服务相似性位置k匿名隐私保护方法。通过引入服务相似性生成标签相似地图,并根据服务相似度将用户所在分区与其他分区合并,生成满足用户服务质量需求的匿名候选区。为抵御推理攻击,在选择k匿名集时使用位置熵对攻击者的背景知识进行量化以生成熵最大的k匿名集。在生成k匿名集时,采用贪心策略并在匿名集随机选取一个位置点请求服务以降低资源开销。安全性分析和实验结果表明:相较于已有位置k匿名方法,在相同条件下,所提方法的隐私保护度和服务可用性平均提高了48.49%和60.01%,时间开销和通信开销平均降低了86.12%和66.57%。所提方法在隐私保护度、服务可用性和资源开销之间取得了更好的权衡,且可有效抵御背景知识推理攻击。
Aiming at the problem that existing location k anonymity methods cannot provide good trade-offs between location privacy and location based service(LBS)utility,which are vulnerable to background knowledge inference attacks and leading to an overhead of resources cost,a service similarity location k anonymity privacy protection scheme against background knowledge inference attacks is proposed.Introducing definition service similarity to generate a label similar map and merging the user partition with other partitions according to service similarity,an anonymous candidate region that satisfies the user service quality requirement is generated.To resist the inference attacks,the attacker background knowledge is quantified by location entropy as selecting k anonymous set to generate the k anonymous set with the largest entropy value.During generating k anonymous set,a greedy strategy is adopted.To reduce the over-expenditure of resources,a location point in the k anonymous set is randomly selected to request services.The security analysis and experimental results show that compared with the existing location k anonymity methods,the privacy protection and LBS utility of the proposed method increase by 48.49%and 60.01%on average,respectively,and the time overhead and communication overhead reduced by 86.12%and 66.57%on average,respectively,under the same conditions.Therefore,the better trade-offs among privacy protection,LBS utility and resource cost are realized by this scheme,which effectively resists background knowledge inference attacks.
作者
闫光辉
刘婷
张学军
蔡刚
何福存
李俊成
YAN Guanghui;LIU Ting;ZHANG Xuejun;CAI Gang;HE Fucun;LI Juncheng(School of Electronic and Information Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2020年第1期8-18,共11页
Journal of Xi'an Jiaotong University
基金
国家自然科学基金资助项目(61762058,61662066)
兰州交通大学“百名青年优秀人才培养计划”基金资助项目
关键词
位置隐私
服务相似性
位置熵
背景知识
k匿名
贪心策略
location privacy
service similarity
location entropy
background knowledge
k anonymity
greedy strategy
