基于信誉的多方非平衡隐私集合求交OA
Reputation-Based Multi-Party Unbalanced Private Set Intersection
现有的非平衡隐私集合求交协议存在局限性.大多数协议仅适用于两方场景,当应用于多方场景时,随着参与方数量的增加,计算和通信开销会显著上升.另外,大多数协议基于半诚实模型假设,而在实际应用中,若参与方存在恶意行为,可能会导致其他参与方的隐私信息泄露.本文设计了一个基于信誉的多方非平衡隐私集合求交方案.首先,结合不经意键值对存储、混淆布隆过滤器等技术,构建了一个轻量级多方非平衡隐私集合求交协议,使所有参与方都可以公平地得到集合交集.随后,为激励参与方诚实参与隐私集合求交计算,在多方非平衡隐私集合求交协议的基础上,引入了信誉管理机制,为每一参与方设置信誉值,并结合两方不完全信息博弈,激励参与方诚实参与计算过程.最后,通过理想-实际范例证明了协议的安全性,并进行了实验性能评估.实验结果显示,在面对更大规模的数据集和更多参与方的情况下,该协议的性能更好,其总体运行效率提升了约 35%.
Existing unbalanced private set intersection(PSI)protocols have some limitations.On one hand,most PSI protocols are only applicable to two-party scenarios,and when applied to multi-party scenarios,the computational and communication overheads will increase significantly as the number of participants increases.On the other hand,these protocols mostly assume a semi-honest model,and in practical applications,if participants engage in malicious behavior,it may lead to the leakage of other participants'privacy information.To address these issues,this study designs a multi-party unbalanced PSI scheme based on reputation.First,by combining technologies such as oblivious key-value storage and permuted Bloom filters,a lightweight multi-party unbalanced PSI protocol is constructed,enabling all participants to fairly obtain the intersection of sets.Subsequently,to encourage participants to honestly engage in the PSI computation,a reputation management mechanism is introduced on the basis of the multi-party unbalanced PSI protocol.A reputation value is set for each participant,and combined with the incomplete information game of two parties,participants are motivated to honestly participate in the computation process.Finally,performance tests were conducted on the multi-party unbalanced PSI computation stage,and the security of the protocol was demonstrated through ideal-real world examples.Specifically,the runtime efficicency is improved by 35%in sceanario of larger dataset and more participants.
左祥建;李楠;乔震宇;周由胜
重庆邮电大学 网络空间安全与信息法学院,重庆 400065重庆邮电大学 网络空间安全与信息法学院,重庆 400065重庆邮电大学 网络空间安全与信息法学院,重庆 400065南京理工大学 网络空间安全学院,南京 210094
信息技术与安全科学
多方非平衡隐私集合求交不经意键值对存储混淆布隆过滤器信誉管理
multi-party unbalanced private set intersectionoblivious key-value storegarbled Bloom filterreputation management
《密码学报(中英文)》 2026 (4)
728-747,20
国家重点研发计划(2023YFF0905300)National Key Research and Development Program of China(2023YFF0905300)
评论