Abstract: Privacy preserving set intersection was analyzed in the secure multi-party computation. Based on semi-honest model and scalar product protocol, a protocol of the bipartite set intersection was constructed, which complexity is O(ntp); another protocol of the tripartite set intersection needs O(2ntp). Then the correctness of these protocols was illustrated and the complexity and security were analyzed precisely. The analysis shows that the new protocols are more efficient than current protocols. Finally, the application and deficiency of these protocols were also discussed.

Key words: secure multi-party computation, privacy-preserving, set intersection, scalar product protocol

摘要： 研究了安全多方计算中的保护私有信息的集合交集问题。在半诚实模型下，基于点积协议设计的两方集合交集协议，复杂度为O(ntp);设计的三方集合交集协议，复杂度为O(2ntp)。给出了协议的正确性理论证明，并对其安全性和复杂度进行了理论分析，性能优于现有协议。最后，给出了协议的推广应用以及不足。

关键词: 安全多方计算, 私有信息, 集合交集, 点积协议