commitment在密码学协议中是非常重要的组成部分。基本定义为:A commitment scheme allows a committer to publish a value, called the commitment, which binds her to a message (binding) without revealing it (hiding). Later, she may open the commitment and reveal the committed message to a verifier, who can check that the message is consistent with the commitment.
1.2 commitment特性commitment一般分为commit和reveal两个阶段,commit整个过程可形象描述为a committer P将某个信息m放进了一个密码箱中,箱子上锁后归a verifier V所有,锁钥匙归P所有;reveal过程为P将钥匙给V,V打开箱子可查看里面的信息m。commitment具有如下特性:
- hiding特性:即V拥有了上锁的箱子,由于没有钥匙无法获取里面的信息m。
- binding特性:即尽管P拥有了钥匙,但是箱子归V所有,P无法在上锁后再次开锁修改其中的信息为m’。即某个值commit后,将不可修改。
当使用公私钥对来满足hiding特性进行上锁过程时,可称为Trapdoor commitment或者chameleon commitment。
根据binding属性,reveal过程可分为hard和soft两种。在commit阶段,可选择是创建hard还是soft commitment。hard commitment是指标准的commitment,即针对消息m进行commit,reveal后的结果也仅可能为m。soft commitment是指初始时,对空消息(no message)进行commit,然后后续可reveal为任意的消息m。
可参加博客Pedersen Commitment扫盲及sage和python脚本,commitment具有加法同态性。
1.3 commitment的实现方式通常有3个著名的方式来实现a committer can use to commit to a message: 假设 g 和 h g和h g和h为group G G G(具有prime order p p p)的两个随机generator,对于消息 m ∈ R Z p m\in_RZ_p m∈RZp:
- 1)commit计算方式为: C < g > ( m ) = g m C_{}(m)=g^m C(m)=gm。此种方式,具有unconditionally binding特性,以及基于离散对数在 G G G中很难的假设的computationnally hiding特性。
- 2)Pedersen commitment,计算方式为: C < g , h > ( m , r ) = g m h r , r ∈ R Z p C_{}(m,r)=g^mh^r,r\in_R Z_p C(m,r)=gmhr,r∈RZp,由于 r r r为随机数,Pedersen commitment具有unconditionally hiding特性,以及基于离散对数假设的computationally binding特性。
- 3)计算方式为: H ( m ) o r H ( m ∣ ∣ r ) H(m)\ or\ H(m||r) H(m) or H(m∣∣r),其中 H H H为one-way function单向函数。实际常使用抗撞击的hash函数。
vector commitment是对一系列有序vector ( m 1 , m 2 , . . . , m q ) (m_1,m_2,...,m_q) (m1,m2,...,mq) 进行commit和reveal操作。 对第一节的binding特性进行了加强,即具有position binding特性。postion binding特性是指对于同一位置i,不可能reveal出两个不同的值 m i , m i ′ m_i,m_i' mi,mi′。
同时vector commitment具有updatable特性,即对位置i的消息 m i m_i mi更新为 m i ′ m_i' mi′,可以同时将相应的commitment C C C更新为 C ′ C' C′,满足proof update特性。
详细的内容可参看论文《Vector Commitments and their Applications》第三章。
论文《Constant-Size Commitments to Polynomials and Their Applications》中指出,polynomial commitment可用于:verifiable secret sharing, zero-knowledge sets, credentials and selective disclosure of signed data。
3.1 polynomial commitment定义基于verifiable secret sharing应用场景引出: 多项式 ϕ ( x ) ∈ R Z p [ x ] \phi (x)\in_RZ_p[x] ϕ(x)∈RZp[x],假设 ϕ ( x ) \phi (x) ϕ(x)多项式的阶为 t t t,系数为 ϕ 0 , . . . ϕ t \phi_0,...\phi_t ϕ0,...ϕt,polynomial commitment可表示为a commit to the string ( ϕ 0 ∣ ϕ 1 ∣ . . . ∣ ϕ t ) (\phi_0|\phi_1|...|\phi_t) (ϕ0∣ϕ1∣...∣ϕt),或者其它能明确代表 ϕ ( x ) \phi (x) ϕ(x)的string。
论文《Constant-Size Commitments to Polynomials and Their Applications》中主要基于以下代数属性: 对于多项式 ϕ ( x ) ∈ R Z p [ x ] \phi (x)\in_RZ_p[x] ϕ(x)∈RZp[x],对于任意的 i ∈ Z p i\in Z_p i∈Zp,多项式 ϕ ( x ) − ϕ ( i ) \phi(x)-\phi(i) ϕ(x)−ϕ(i)都可以整除 ( x − i ) (x-i) (x−i)。
-
论文《Constant-Size Commitments to Polynomials and Their Applications》: provided protocols to commit to polynomials and then evaluate them at a given point in a verifiable way. Their protocols only require a constant number of commitments but security relies on pairing assumptions。 基于的数学原理为:(论文《Sonic: Zero-Knowledge SNARKs from Linear-Size Universal and Updatable Structured Reference Strings》中的polynomial commitment也是基于此原理。)
-
论文《Efficient Zero-Knowledge Arguments for Arithmetic Circuits in the Discrete Log Setting》: Our polynomial commitment protocol has square root communication complexity but relies solely on the discrete logarithm assumption. 基于的原理是:(将系数拆分为矩阵,每行作为一个vector,将polynomial commitment拆分由vector commitment组成。)
参考资料: [1] 博客密码学上的commitment [2] 博客Pedersen Commitment扫盲及sage和python脚本 [3] 论文《Vector Commitments and their Applications》 [4] 论文《Constant-Size Commitments to Polynomials and Their Applications》 [5] 论文《Functional Commitment Schemes: From Polynomial Commitments to Pairing-Based Accumulators from Simple Assumptions》 [6] 论文《Zero-knowledge Argument for Polynomial Evaluation with Application to Blacklists》 [7] ppt分享《Efficient Batch Zero-Knowledge Arguments for Low Degree Polynomials》