重复攻击下差分隐私保护参数ε的选取方法

doi:10.19678/j.issn.1000-3428.0048453

计算机工程 ›› 2018, Vol. 44 ›› Issue (7): 145-149. doi: 10.19678/j.issn.1000-3428.0048453

重复攻击下差分隐私保护参数ε的选取方法

郝晨艳^1a,2,3,彭长根 ^1b,2,3,4,张盼盼 ^1a,2,3

1.贵州大学 a.数学与统计学院;b.计算机科学与技术学院,贵阳 550025; 2.贵州省公共大数据重点实验室,贵阳 550025; 3.密码学与数据安全研究所,贵阳 550025; 4.广东省信息安全技术重点实验室,广州 510006

收稿日期:2017-08-24 出版日期:2018-07-15 发布日期:2018-07-15
作者简介:郝晨艳(1990—),女,硕士,主研方向为信息安全、可信计算;彭长根,教授、博士、博士生导师;张盼盼,硕士。
基金资助:
国家自然科学基金(61662009,61262073,61363068);国家密码发展基金(MMJJ20170129);广东省信息安全技术重点实验室开放课题(GDXXAQ2016-04);贵州省科技计划项目(［2016］1023,［2017］1045)。

Selection Method of Differential Privacy Protection Parameter ε Under Repeated Attack

HAO Chenyan ^1a,2,3,PENG Changgen ^1b,2,3,4,ZHANG Panpan ^1a,2,3

1a.College of Mathematics and Statistics;1b.College of Computer Science and Technology,Guizhou University,Guiyang 550025,China; 2.Guizhou Provincial Key Laboratory of Public Big Data,Guiyang 550025,China; 3.Institute of Cryptography and Data Security,Guiyang 550025,China; 4.Guangdong Provincial Key Laboratory of Information Security Technology,Guangzhou 510006,China

Received:2017-08-24 Online:2018-07-15 Published:2018-07-15

摘要/Abstract

摘要：

现有差分隐私保护参数ε的选取仅考虑在给定攻击模型下,对攻击者进行一次攻击的情形,但在实际中往往涉及攻击者的连续多次攻击。为此,提出一种重复攻击下实现差分隐私保护技术安全性的攻击模型。通过对重复攻击的结果与噪音分布函数的特点进行分析,获取噪音落入容错区间的概率,结合排列组合方法,推断出攻击者攻击成功的概率,进而得到参数ε的选取范围。实验结果表明,只要给定攻击次数、容错区间和攻击者成功的概率,该模型就可以找出合适的参数ε。

关键词: 差分隐私, 重复攻击, 随机算法, 攻击模型, 数据隐私

Abstract:

The published literature on the selection of differential privacy protection parameter ε focuses on only one time attack of data-sets by attackers,while in practice it frequently involves the attacker’s repeated attacks.In view of this,a measure model under repeated attacks is proposed.The probability that the noise falls into the fault-tolerant interval is obtained by analyzing the results of the repeated attack and the characteristics of noise distribution function.Combining with the method of permutation and combination,it infers the attacker’s probability of successful attack,further obtains the selection range of the parameter ε.Experimental results show that,when given the number of attacks,the fault-tolerance interval and the probability of the attacker’s success,the proposed model can always find the appropriate parameter ε.

Key words: differential privacy, repeated attack, random algorithm, attack model, data privacy

中图分类号:

TP311

郝晨艳,彭长根,张盼盼. 重复攻击下差分隐私保护参数ε的选取方法[J]. 计算机工程, 2018, 44(7): 145-149.

HAO Chenyan,PENG Changgen,ZHANG Panpan. Selection Method of Differential Privacy Protection Parameter ε Under Repeated Attack[J]. Computer Engineering, 2018, 44(7): 145-149.

https://www.ecice06.com/CN/Y2018/V44/I7/145

参考文献

［1］DWORK C.Differential privacy:a survey of results［C］//Proceedings of International Conference on Theory and Applications of Models of Computation.Berlin,Germany:Springer-Verlag,2008:1-19.
［2］DALENIUS T.Towards a methodology for statistical disclosure control［J］.Statistik Tidskrift,1997,15(2):429-444.
［3］SWEENEY L.K-anonymity:a model for protecting privacy［J］.International Journal of Uncertainty,Fuzziness and Knowledge-based Systems,2002,10(5):557-570.
［4］MACHANAVAJJHALA A,GEHRKE J,KIFER D,et al.l-diversity:privacy beyond k-anonymity［C］//Proceedings of the 22nd International Conference on Data Engineering.Washington D.C.,USA:IEEE Press,2006:24-24.
［5］LI Ninghui,LI Tiancheng,VENKATASUBRMANIAN S.T-closeness:privacy beyond k-anonymity and l-diversity［C］//Proceedings of the 23rd IEEE International Conference on Data Engineering.Washington D.C.,USA:IEEE Press,2007:106-115.
［6］WONG R C W,LI Jiuyong,FU A W C,et al.(α,k)-anonymity:an enhanced k-anonymity model for privacy preserving data publishing［C］//Proceedings of the 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York,USA:ACM Press,2006:754-759.
［7］ZHANG Qing,KOUDAS N,SRIVASTAVA D,et al.Aggregate query answering on anonymized tables［C］//Proceedings of the 23rd IEEE International Conference on Data Engineering.Washington D.C.,USA:IEEE Press,2007:116-125.
［8］KELLY D J,RAINES R A,GRIMAILA M R,et al.A survey of state-of-the-art in anonymity metrics［C］//Proceedings of the 1st ACM Workshop on Network Data Anonymization.New York,USA:ACM Press,2008:31-40.
［9］彭长根,丁红发,朱义杰,等.隐私保护的信息熵模型及其度量方法［J］.软件学报,2016,27(8):1891-1903.
［10］DWORK C.Differential privacy［C］//Proceedings of the 33rd International Colloquium on Automata,Languages and Programming.Washington D.C.,USA:IEEE Press,2006:1-12.
［11］DWORK C,MCSHERRY F,NISSIM K,et al.Calibrating noise to sensitivity in private data analysis［M］.Berlin,Germany:Springer-Verlag,2006.
［12］MCSHERRY F,TALWAR K.Mechanism design via differential privacy［C］//Proceedings of the 48th Annual IEEE Symposium on Foundations of Computer Science.Washington D.C.,USA:IEEE Press,2007:94-103.
［13］ROTH A,ROUGHGARDEN T.Interactive privacy via the median mechanism［C］//Proceedings of ACM Symposium on Theory of Computing.New York,USA:ACM Press,2010:765-774.
［14］LI Chao,HAY M,RASTOGI V,et al.Optimizing linear counting queries under differential privacy［C］//Proceedings of the 29th ACM SIGMOD-SIGACT-SIGART Symposim on Principles of Database Systems.New York,USA:ACM Press,2010:123-134.
［15］XIAO Xiaokui,WANG Guozhang,GEHRKE J.Differential privacy via wavelet transforms［J］.IEEE Transactions on Knowledge and Data Engineering,2011,23(8):1200-1214.
［16］HAY M,RASTOGI V,MIKLAU G,et al.Boosting the accuracy of differentially private histograms through consistency［J］.Proceedings of the VLDB Endowment,2010,3(1-2):1021-1032.
［17］YUAN Ganzhao,ZHANG Zhenjie,WINSLETT M,et al.Low-rank mechanism:optimizing batch queries under differential privacy［J］.Proceedings of the VLDB Endowment,2012,5(11):1352-1363.
［18］LEE J,CLIFTON C.How much is enough? choosing e for differential privacy［J］.Information Security,2011,7001:325-340.
［19］HSU J,GABOARDI M,HAEBERLEN A,et al.Differential privacy:an economic method for choosing epsilon［C］//Proceedings of the 27th IEEE Computer Security Foundations Symposium.Washington D.C.,USA:IEEE Press,2014:398-410.
［20］何贤芒,王晓阳,陈华辉,等.差分隐私保护参数 ε 的选取研究［J］.通信学报,2015,36(12):124-130.
［21］DWORK C,ROTH A.The algorithmic foundations of differential privacy［M］.Berlin,Germany:Springer-Verlag,2014.

[1]	卢晓天, 朴春慧, 杨兴雨, 白英杰. 基于贝叶斯网络的差分隐私高维数据发布技术研究[J]. 计算机工程, 2024, 50(5): 167-181.
[2]	张晓均, 李兴鹏, 唐伟, 郝云溥, 薛婧婷. 云-边融合的可验证隐私保护跨域联邦学习方案[J]. 计算机工程, 2024, 50(3): 148-155.
[3]	张攀峰, 吴丹华, 董明刚. 基于粒子群优化的差分隐私深度学习模型[J]. 计算机工程, 2023, 49(9): 144-157.
[4]	王群, 李馥娟, 倪雪莉, 夏玲玲, 梁广俊. 区块链数据形成与隐私威胁[J]. 计算机工程, 2023, 49(8): 1-12.
[5]	吴茂强, 黄旭民, 康嘉文, 余荣. 面向车路协同推断的差分隐私保护研究[J]. 计算机工程, 2022, 48(7): 29-35.
[6]	金媛媛, 倪志伟, 朱旭辉, 陈恒恒, 陈千. 基于本地差分隐私的空间数据自适应划分算法[J]. 计算机工程, 2022, 48(5): 136-144.
[7]	王元庆, 刘百祥. 一种去中心化的隐私保护匿名问卷方案[J]. 计算机工程, 2021, 47(6): 123-131.
[8]	李杨, 陈子彬, 谢光强. 一种基于ExtraTrees的差分隐私保护算法[J]. 计算机工程, 2020, 46(2): 134-140.
[9]	李远航, 陈先来, 刘莉, 安莹, 李忠民. 面向差分隐私保护的随机森林算法[J]. 计算机工程, 2020, 46(1): 93-101.
[10]	王丹,龙士工. 权重社交网络隐私保护中的差分隐私算法[J]. 计算机工程, 2019, 45(4): 114-118.
[11]	郝志峰,王日宇,蔡瑞初,温雯. 基于贝叶斯网络与语义树的隐私数据发布方法[J]. 计算机工程, 2019, 45(4): 124-129.
[12]	李洪成,吴晓平. 支持告警序列差分隐私保护的网络入侵关联方法[J]. 计算机工程, 2018, 44(5): 128-132.
[13]	孙奎,张志勇,赵长伟. 一种增强的差分隐私数据发布算法[J]. 计算机工程, 2017, 43(4): 160-165.
[14]	唐昌龙,刘吉强. 基于数据分配策略的数据泄漏检测研究[J]. 计算机工程, 2015, 41(4): 140-144.
[15]	张晶,李艳. 基于动态博弈的粗糙网络安全分析模型[J]. 计算机工程, 2015, 41(4): 129-134.

选择文件类型/文献管理软件名称

选择包含的内容

重复攻击下差分隐私保护参数ε的选取方法

Selection Method of Differential Privacy Protection Parameter ε Under Repeated Attack

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

重复攻击下差分隐私保护参数ε的选取方法

Selection Method of Differential Privacy Protection Parameter ε Under Repeated Attack

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价