Performing L pairs of multiplication with known (k, n)-linear secret sharing-based n party multiplication protocols using an (off-line) preprocessing by the Beaver multiplication triple requires 2Lk elements of on-line communications where k is the threshold. This paper proposes a simultaneous multiplication protocol for L pairs based on (k, L, n)-packed linear secret sharing. By embedding L secret information into each share using (k, L, n)-packed linear secret sharing, though partial information of secret information is leaked from \(k - l + 1\) shares for \(1 \le l \le L\) , the proposed protocol can perform L multiplications with 2k elements of online communications. As applications of the proposed simultaneous multiplication protocol, we also introduce a simultaneous comparison protocol and an improvement of the communication cost of Araki et al.’s n party group-wise maximum/minimum value aggregation protocol. For example, when computing the maximum/minimum value of eight 64-bit data, our proposed aggregation protocol can reduce \(59.6\%\) of the communication cost.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Simultaneous Multiplication Protocol Based on Packed Linear Secret Sharing and Its Applications

  • Shohei Mita,
  • Kazuki Yoneyama

摘要

Performing L pairs of multiplication with known (k, n)-linear secret sharing-based n party multiplication protocols using an (off-line) preprocessing by the Beaver multiplication triple requires 2Lk elements of on-line communications where k is the threshold. This paper proposes a simultaneous multiplication protocol for L pairs based on (k, L, n)-packed linear secret sharing. By embedding L secret information into each share using (k, L, n)-packed linear secret sharing, though partial information of secret information is leaked from \(k - l + 1\) shares for \(1 \le l \le L\) , the proposed protocol can perform L multiplications with 2k elements of online communications. As applications of the proposed simultaneous multiplication protocol, we also introduce a simultaneous comparison protocol and an improvement of the communication cost of Araki et al.’s n party group-wise maximum/minimum value aggregation protocol. For example, when computing the maximum/minimum value of eight 64-bit data, our proposed aggregation protocol can reduce \(59.6\%\) of the communication cost.