We suggest practical sub-linear size zero-knowledge arguments for statements involving linear algebra. Given commitments to matrices over a finite field, we give a sub-linear size zero-knowledge argument that one committed matrix is the product of two other committed matrices. We also offer a sub-linear size zero-knowledge argument for a committed matrix being equal to the Hadamard product of two other committed matrices. Armed with these tools we can give many other sub-linear size zero-knowledge arguments, for instance for a committed matrix being upper or lower triangular, a committed matrix being the inverse of another committed matrix, or a committed matrix being a permutation of another committed matrix. A special case of what can be proved using our techniques is the satisfiability of an arithmetic circuit with N gates. Our arithmetic circuit zero-knowledge argument has a communication complexity of group elements. We give both a constant round variant and an O(√N) round variant of our zero-knowledge argument; the latter has a computation complexity of O(N/logN) exponentiations for the prover and O(N) multiplications for the verifier making it efficient for the prover and very efficient for the verifier. In the case of a binary circuit consisting of NAND-gates we give a zero-knowledge argument of circuit satisfiability with a communication complexity of O(√N) group elements and a computation complexity of O(N) multiplications for both the prover and the verifier. © 2009 Springer.
CITATION STYLE
Groth, J. (2009). Linear algebra with sub-linear zero-knowledge arguments. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5677 LNCS, pp. 192–208). https://doi.org/10.1007/978-3-642-03356-8_12
Mendeley helps you to discover research relevant for your work.