研究成果

国際会議

  • An Improvement of Multi-Party Private Set Intersection Based on Oblivious Programmable PRFs
    著者
    S. Shimizu, T. Nakai, Y. Watanabe, and M. Iwamoto
    会議名
    ISITA 2022
    出版社
    IEEE
    発行年
    2022
    (To appear)
    Abstract

    Multi-party private set intersection (PSI) allows parties to compute the set intersection of their private data sets without revealing outside of the intersection. Kolesnikov et al. (ACM CCS 2017) introduced Oblivious Programmable Pseudorandom Function (OPPRF) and showed a practical multi-party PSI protocol secure for arbitrary collusion of parties under the semi-honest model. We point out that their protocol contains some overkill OPPRFs for the required functionality. On the basis of this finding, we improve their PSI protocol by replacing these OPPRFs with more lightweight procedures. More precisely, we introduce a new functionality called Extended Programmable Pseudorandom Function (EPPRF). It provides functionality that excludes an expensive public-key operation from the OPPRF. We show that a multi-party PSI protocol can be realized even if the OPPRFs are replaced with EPPRFs. As a result of the replacement, we reduce the number of public-key operations n-1 times from Kolesnikov et al.'s protocol, where n is the number of parties.