Adaptive security for threshold cryptosystems

Ran Canetti, Rosario Gennaro, Stanisław Jarecki, Hugo Krawczyk, Tal Rabin

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


We present adaptively-secure efficient solutions to several central problems in the area of threshold cryptography. We prove these solutions to withstand adaptive attackers that choose parties for corruption at any time during the run of the protocol. In contrast, all pre­viously known efficient protocols for these problems were proven secure only against less realistic static adversaries that choose and fix the subset of corrupted parties before the start of the protocol run. Specifically, we provide adaptively-secure solutions for distributed key generation in discrete-log based cryptosystems, and for the problem of distributed generation of DSS signatures (threshold DSS). We also show how to transform existent static solutions for threshold RSA and proac­tive schemes to withstand the stronger adaptive attackers. In doing so, we introduce several techniques for the design and analysis of adaptively-secure protocols that may well find further applications.

Original languageEnglish
Title of host publicationAdvances in Cryptology – CRYPTO 1999 - 19th Annual International Cryptology Conference, Proceedings
EditorsMichael Wiener
PublisherSpringer Verlag
Number of pages19
ISBN (Print)3540663479, 9783540663478
StatePublished - 1999
Externally publishedYes
Event19th Annual International Cryptology Conference, CRYPTO 1999 - Santa Barbara, United States
Duration: 15 Aug 199919 Aug 1999

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference19th Annual International Cryptology Conference, CRYPTO 1999
Country/TerritoryUnited States
CitySanta Barbara


Dive into the research topics of 'Adaptive security for threshold cryptosystems'. Together they form a unique fingerprint.

Cite this