This project is a Golang implementation of BGW Secure Linear MultiParty Computation, which supports int and big.Int of golang.

The paper we mainly refer to when implementing this package is "Asharov, Gilad , and Y. Lindell . "A Full Proof of the BGW Protocol for Perfectly Secure Multiparty Computation." Journal of Cryptology 30(2015):1-94.".

LinearMultiPartyComputation generalizes the linear mpc scheme in "Ben-Or M, Goldwasser S, Wigderson A. Completeness theorems for non-cryptographic fault-tolerant distributed computation. In Proceedings of the twentieth annual ACM symposium on Theory of computing 1988 Jan 1 (pp. 1-10). ACM."

In Linear MPC, a given number of participants p₁, p₂, ..., p_n, each has private data, respectively x₁, x₂, ..., x_n. Participants want to compute the value of a public function on the private data: f(x₁, x₂, ..., x_n) while keeping their own inputs secret, if there are no more than t<n/2 semi-honest adversaries.

A linear function is in the form f(x₁, x₂, ..., x_n) = c₁x₁ + c₂x₂ + ... + c_nx_n, while c₁, c₂, ..., c_n are constants.

• Note 1: In the scheme, all element should be in some Zp, i.e. should be non-negative integers.
• Note 2: Each participant has a unique ID, starting from 0 to n-1.

This package is implemented in Golang (version 1.9+), without any external dependencies.

You can simply import our linear mpc module as a normal Golang package.

git clone https://github.com/zhengjilai/BGW-SecureLinearMultiPartyComputation.git
mkdir -p $GOPATH/src
cp -r BGW-SecureLinearMultiPartyComputation/loccs.sjtu.edu.cn $GOPATH/src

• /loccs.sjtu.edu.cn/acrypto/poly implements the calculation of polynomial over Zp with single variable and a system of solving linear equations over Zp.

• /loccs.sjtu.edu.cn/acrypto/secretshare implements Shamir's secret sharing scheme over Zp.

• /loccs.sjtu.edu.cn/acrypto/mpc implements BGW Linear MultiParty Computation, where everyone has an secret x_i, and they want to know the output of an linear function f(x₁, x₂, ..., x_n) while not exposing their own secret (on condition that there are only t<n/2 semi-honest adversaries).

• /doc: Basic documents of this project, including the original paper and our project docs(interfaces, principles and communication analysis). We also provide an easy explanation of BGW-mpc Multiplication gate, although we have not implemented it.

This mpc package is only written for study, and should never be leveraged for production.

All contributors of this repository come from Lab of Cryptology and Computer Security, SJTU.

• Haining Lu
• Jilai Zheng