Interactive proofs with provable security against honest verifiers

0Citations
Citations of this article
37Readers
Mendeley users who have this article in their library.

This article is free to access.

Abstract

Nearly all of the work on constructing zero-knowledge proof systems relies on very strong complexity theoretic assumptions. We consider a form of “no use” zeroknowledge, and show that every language in PSPACE has an interactive proof system that provably achieves “no-use” zero-knowledge against honest verifiers.

Cite

CITATION STYLE

APA

Kilian, J. (1991). Interactive proofs with provable security against honest verifiers. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 537 LNCS, pp. 378–392). Springer Verlag. https://doi.org/10.1007/3-540-38424-3_28

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free