基于差分隐私的大数据安全访问权限认证仿真  

Big Data Security Access Authentication Simulation Based on Differential Privacy

作　　者：刘东[1] 任海玲 LIU Dong;REN Hai-ling(College of Science,Ningxia Medical University,Yinchuan Ningxia 750004,China;Ningxia Medical University,Yinchuan Ningxia 750004,China)

机构地区：[1]宁夏医科大学理学院,宁夏银川750004 [2]宁夏医科大学,宁夏银川750004

出　　处：《计算机仿真》2021年第8期421-424,486,共5页Computer Simulation

基　　金：宁夏医科大学校级重点项目,项目名称:基于大数据的慢性病管理及其影响因素研究(XZ2020008);宁夏自然科学基金项目,项目名称:基于深度学习融合智能优化算法的医学图像处理技术研究(2020AAC03169)。

摘　　要：在使用大数据查询重要信息时,为了防止隐私泄露,需要通过安全防护手段保护数据信息安全,提出基于差分隐私的大数据安全访问权限认证仿真方法。利用添加拉普拉斯生成噪声的方式,统计起始数据,并转换扰动,达到保护隐藏信息的效果。对大数据拥有者或生产者贴上标识,并对访问大数据的对象贴上标签,通过正确设置大数据所有者的读授权条件与写授权条件,在完成数据传输的同时使访问权限也获得认证,最终实现安全防护。经过仿真证明,所提方法对于大数据的安全认证时间较短,安全性较高,全面验证了上述方法的实际应用价值。In order to prevent privacy leakage during the big data query of important information,this paper reports a simulation method of big data security access authentication based on differential privacy.First of all,the noise was generated by adding Laplacian.Meanwhile,the initial data was counted and transformed into disturbance for improving the security of hidden information.Secondly,big data owners,producers and objects accessing big data were given identification.Then,the read and write authorization conditions of big data owners were set correctly.Then,the data transmission was completed and the access rights were authenticated.Eventually,security protection was achieved.The simulation results show that this method has short big data security authentication time,high security and excellent application value.

关 键 词：差分隐私 大数据 安全访问 标识标签 权限认证 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]