通信学报 ›› 2019, Vol. 40 ›› Issue (4): 83-94.doi: 10.11959/j.issn.1000-436x.2019077

• 学术论文 • 上一篇    下一篇

基于多方安全计算的属性泛化mix-zone

王斌1,2,张磊2(),张国印1   

  1. 1 哈尔滨工程大学计算机科学与技术学院,黑龙江 哈尔滨 150001
    2 佳木斯大学信息电子技术学院,黑龙江 佳木斯 154007
  • 修回日期:2019-02-04 出版日期:2019-04-25 发布日期:2019-05-05
  • 作者简介:王斌(1979-),男,黑龙江安达人,哈尔滨工程大学博士生,佳木斯大学副教授,主要研究方向为机器学习、隐私保护。|张磊(1982-),男,黑龙江绥化人,博士,佳木斯大学副教授,主要研究方向为信息安全、隐私保护。|张国印(1962-),男,博士,山东黄县人,哈尔滨工程大学教授、博士生导师,主要研究方向为嵌入式系统与体系结构、网络技术与信息安全。
  • 基金资助:
    黑龙江省自然科学基金资助项目(F2015022);黑龙江省普通本科高等学校青年创新人才培养计划基金资助项目(UNPYSCT-2017149);国家级大学生创新创业训练基金资助项目(201810222033);黑龙江省省属本科高校基本科研业务费基金资助项目

Attribute generalization mix-zone based on multiple secure computation

Bin WANG1,2,Lei ZHANG2(),Guoyin ZHANG1   

  1. 1 College of Computer Science and Technology,Harbin Engineering University,Harbin 150001,China
    2 College of Information and Electronic Technology,Jiamusi University,Jiamusi 154007,China
  • Revised:2019-02-04 Online:2019-04-25 Published:2019-05-05
  • Supported by:
    The Natural Science Foundation of Heilongjiang Province(F2015022);The University Nursing Program for Young Scholars with Creative Talents in Heilongjiang Province(UNPYSCT-2017149);The National Undergraduate Innovation and Entrepreneurship Training Program of China(201810222033);Basic Scientific Research Service Fee Project of Heilongjiang Provincial Undergraduate Universities

摘要:

针对路网环境下mix-zone无法有效地实现属性进行隐藏或泛化和抵御伪装攻击的问题,基于属性泛化和同态加密,提出了一种秘态属性泛化的隐私保护方法。该方法通过同态加密,实现了秘密出价选择计算代理、秘密计算相似属性,并以相似属性完成属性泛化的整体处理。通过属性泛化,解决了mix-zone可被攻击者利用属性追踪的问题,同时秘密计算的属性处理不会泄露任何信息给参与者,也防止伪装攻击者获得mix-zone中各用户的隐私信息。最后,通过安全性分析和实验验证分别在理论和实践这2个方面对所提算法的优势加以分析和比较。

关键词: 路网环境, 隐私保护, 属性泛化, 同态加密, 安全计算

Abstract:

In order to cope with the problems of resist the attack of user tracking with attributes and resist the disguised attack in road networks,based on the conception of attribute generalization and the methods of homomorphic encryption,a privacy protection method to complete the calculation of attribute generalization in private state and achieve attribute generalization was proposed.The proposed method utilized the homomorphic encryption to achieve secret bidding selection with an agent,and then utilize secure multi-party computation to calculate the similar attributes and achieve the attribute generalization in the end.With the help of fully attribute generalization,this method can solve the problem of users can be tracked by the potential attributes,and at the same time the privacy calculation also does not reveal any information to participants,so this method can prevent the disguised attacker that obtains user’s information in mix-zone.At last,in order to demonstrate the superiority of the proposed method in both of academically and practicality,the security analysis and experimental verification are given,and the procedure of formulation verification and the result of experiment further substantiate the superiorities of the proposed method.

Key words: road network, privacy protection, attribute generalization, homomorphic encryption, secure computation

中图分类号: 

  • TP311