Journal of Computer Applications ›› 2024, Vol. 44 ›› Issue (9): 2777-2784.DOI: 10.11772/j.issn.1001-9081.2023091296

• Cyber security • Previous Articles     Next Articles

Federated spatial data publication method with differential privacy and secure aggregation

Zhizheng ZHANG1,2, Xiaojian ZHANG1,2, Junqing WANG1,2(), Guanghui FENG3   

  1. 1.School of Computer and Information Engineering,Henan University of Economics and Law,Zhengzhou Henan 450046,China
    2.Zhengzhou Key Laboratory of Artificial Intelligence Security and Privacy Protection (Henan University of Economics and Law),Zhengzhou Henan 450046,China
    3.School of Computer Science and Cyber Engineering,Guangzhou University,Guangzhou Guangdong 510006,China
  • Received:2023-09-20 Revised:2023-12-26 Accepted:2023-12-29 Online:2024-02-22 Published:2024-09-10
  • Contact: Junqing WANG
  • About author:ZHANG Zhizheng, born in 1997, M. S. candidate. His research interests include differential privacy, federated analysis.
    ZHANG Xiaojian, born in 1980, Ph. D., professor. His research interests include differential privacy, machine learning, graph data management.
    FENG Guanghui, born in 1982, Ph. D. candidate. His research interests include privacy protection, differential privacy.
  • Supported by:
    National Natural Science Foundation of China(62072156)

结合差分隐私与安全聚集的联邦空间数据发布方法

张治政1,2, 张啸剑1,2, 王俊清1,2(), 冯光辉3   

  1. 1.河南财经政法大学 计算机与信息工程学院, 郑州 450046
    2.郑州市人工智能安全与隐私保护重点实验室(河南财经政法大学), 郑州 450046
    3.广州大学 计算机科学与网络工程学院, 广州 510006
  • 通讯作者: 王俊清
  • 作者简介:张治政(1997—),男,河南信阳人,硕士研究生,主要研究方向:差分隐私、联邦分析
    张啸剑(1980—),男,河南郑州人,教授,博士,CCF会员,主要研究方向:差分隐私、机器学习、图数据管理
    王俊清(1999—),男,河南驻马店人,硕士研究生,主要研究方向:隐私保护、差分隐私
    冯光辉(1982—),男,河南驻马店人,博士研究生,主要研究方向:隐私保护、差分隐私。
  • 基金资助:
    国家自然科学基金资助项目(62072156)

Abstract:

Aiming at the problems of federated spatial data isolation, spatial data indexing, and privacy of publishing spatial data, a Federated Spatial data Publishing (FSP) method based on dynamic quad-tree was proposed. Firstly, in each iteration of the FSP method, quad-tree replica was shared by the server with each client in the round, and each client encoded its own location data using the quad-tree replica, and discrete noise was generated through Polya distribution for locally perturbing the encoding results. Secondly, local masks were generated through LWE (Learning With Error) to encrypt the noisy results. Thirdly, the reported values from each client in the iteration were combined by the aggregator to perform secure aggregation and mask elimination. Then the aggregated results were sent to the server. The quad-tree structure was pruned by the server dynamically in a bottom-up way based on the collected encoding vectors and noise variance. Experimental results on four spatial datasets Beijing, Checkin, NYC, and Landmark show that the FSP method not only ensures client privacy, but also reduces the Mean Squared Error (MSE) in federated spatial data publication by 3.80%, 2.96%, 7.51% and 14.13% at a privacy budget of 1.8, respectively, compared to the existing better federated spatial data publication method AHH (Adaptive Hierarchical Histograms). This indicates that the FSP method achieves higher precision than similar methods in federated spatial data publishing.

Key words: federated analytics, distributed differential privacy, secure aggregation, spatial data publication, quad-tree

摘要:

针对联邦空间数据的数据孤岛问题、空间数据索引问题以及发布联邦空间数据存在的隐私问题,提出基于动态四分树的联邦空间数据发布(FSP)方法。首先,在FSP方法的每轮迭代中,服务端把四分树副本共享给该轮中每个客户端,每个客户端利用四分树副本编码自身位置数据,利用Polya分布产生离散噪声在本地扰动编码结果;其次,结合容错学习(LWE)生成本地掩码对噪声结果进行加密;再次,安全聚集端结合该轮迭代中每个客户端的报告值,执行安全聚集与消除掩码操作,然后把聚集结果发送给服务端;最后,服务端结合收集的编码向量与噪声方差自底向上地动态修剪四分树结构。在Beijing、Checkin、NYC和Landmark 4个空间数据集上的实验结果表明,FSP方法在保证客户端隐私的同时,与已有的较好的联邦空间数据发布方法AHH(Adaptive Hierarchical Histograms)相比,在隐私预算为1.8时,FSP的均方误差(MSE)分别降低了3.80%、2.96%、7.51%和14.13%。可见使用FSP方法进行联邦空间数据发布的精度优于同类方法。

关键词: 联邦分析, 分布式差分隐私, 安全聚集, 空间数据发布, 四分树

CLC Number: