面向网络服务最小相似度安全计算的无界函数加密  

Unbounded function encryption for minimum similarity privacy computation in web services

作　　者：解君睿 陈振华[1,2] XIE Jun-rui;CHEN Zhen-hua(College of Computer Science and Technology,Xi’an University of Science and Technology,Xi’an 710600,China;Guangxi Key Laboratory of Trusted Software,Guilin University of Electronic Technology,Guilin 541004,China)

机构地区：[1]西安科技大学计算机科学与技术学院,陕西西安710600 [2]桂林电子科技大学广西可信软件重点实验室,广西桂林541004

出　　处：《计算机工程与设计》2025年第3期741-748,共8页Computer Engineering and Design

基　　金：国家自然科学基金项目(61872289);广西可信软件重点实验室基金项目(KX202308)。

摘　　要：现有网络服务器利用函数加密在进行最小相似度计算时,会获得一些中间结果,导致用户的隐私泄露,并且需要在系统建立阶段提前固定用户数据个数,导致其非常不灵活。为此提出一种无界的最小相似度安全计算的函数加密方案。通过新的0-1编码和隐私保护映射变体,在最小相似度计算过程中不会向网络服务器泄露任何中间结果,利用伪随机函数实现无界的函数加密,不需要在系统建立阶段提前固定用户数据个数,可以在计算过程中按需确定,利用对偶对运算向量空间在保护用户数据隐私的同时保护函数隐私。When the web server uses the existing function encryption to calculate the minimum similarity,it obtains some intermediate results,which will lead to the privacy disclosure of users.In addition,the number of user data need to be fixed in advance during the setup stage,which makes it very inflexible.A function encryption scheme for unbounded minimum similarity secure computation was proposed to address these issues.Through the new 0-1 encoding and privacy-preserving mapping variants,no intermediate results were disclosed to the web server in the process of minimum similarity calculation,and pseudo-random functions were used to achieve unbounded function encryption,which did not need to fix the number of user data in advance in the setup stage,and could be determined according to demand in the calculation process.The user data privacy as well as function privacy was protected using the dual pairing vector spaces.

关 键 词：函数加密 最小相似度 无界 隐私保护 网络服务 函数隐私 伪随机函数 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]