基于SMC的分布式隐私保护数据发布研究  

Research on distributed privacy preserving data publishing based on SMC

作　　者：方炜炜[1] 周长胜[1] 贾艳萍[1] 刘亚辉[1] 

机构地区：[1]北京信息科技大学信息中心,北京100192

出　　处：《系统工程与电子技术》2012年第11期2390-2395,共6页Systems Engineering and Electronics

基　　金：国家自然科学基金重点项目(61175048);国家自然科学基金(60875029)资助课题

摘　　要：针对垂直分布式存储结构的隐私保护数据发布问题,基于元组等价群的概念给出全局k-匿名化的定义和充要条件,采用集合多项式表示方法求解出全局元组ID等价群;并基于多方安全计算的同态加密协议构建了具有隐私性、准确性和公平性的分布式隐私保护k-匿名模型,从而实现了各微数据提供方不泄露本地隐私信息的前提下由半可信第三发布方发布出可供统计分析和数据挖掘等需求的真实有效数据集。实验结果表明,该模型具有很好的安全性、准确性和适用性。To solve the privacy-preserving data-publishing problem in context of vertical distribution, the definition of global/e-anonymity and its necessary and sufficient conditions are given based on the concept of equivalent groups of tuples, the global equivalent groups of tuples are solved by using a polynomial representation method, a distributed privacy-preserving k-anonymity model which has the character of privacy, accuracy and fairness by applying the set polynomial indication method and homomorphic encryption protocol based on secure multi-party computation is proposed. The model can help each data owner to securely publish real and affect data set for statistical analyzing and data mining, without revealing sensitive original information. Experiments demonstrate that this model can provide the good capability of security, accuracv and efficiency.

关 键 词：隐私保护数据发布 多方安全计算 匿名 同态加密 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]