Ad hoc PSM protocols: Secure computation without coordination

Amos Beimel, Yuval Ishai, Eyal Kushilevitz

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

5 Scopus citations

Abstract

We study the notion of ad hoc secure computation, recently introduced by Beimel et al. (ITCS 2016), in the context of the Private Simultaneous Messages (PSM) model of Feige et al. (STOC 2004). In ad hoc secure computation we have n parties that may potentially participate in a protocol but, at the actual time of execution, only k of them, whose identity is not known in advance, actually participate. This situation is particularly challenging in the PSM setting, where protocols are non-interactive (a single message from each participating party to a special output party) and where the parties rely on pre-distributed, correlated randomness (that in the ad-hoc setting will have to take into account all possible sets of participants). We present several different constructions of ad hoc PSM protocols from standard PSM protocols. These constructions imply, in particular, that efficient information-theoretic ad hoc PSM protocols exist for NC1 and different classes of log-space computation, and efficient computationally-secure ad hoc PSM protocols for polynomial-time computable functions can be based on a one-way function. As an application, we obtain an information-theoretic implementation of order-revealing encryption whose security holds for two messages. We also consider the case where the actual number of participating parties t may be larger than the minimal k for which the protocol is designed to work. In this case, it is unavoidable that the output party learns the output corresponding to each subset of k out of the t participants. Therefore, a “best possible security” notion, requiring that this will be the only information that the output party learns, is needed. We present connections between this notion and the previously studied notion of t-robust PSM (also known as “non-interactive MPC”). We show that constructions in this setting for even simple functions (like AND or threshold) can be translated into non-trivial instances of program obfuscation (such as point function obfuscation and fuzzy point function obfuscation, respectively). We view these results as a negative indication that protocols with “best possible security” are impossible to realize efficiently in the information-theoretic setting or require strong assumptions in the computational setting.

Original languageEnglish
Title of host publicationAdvances in Cryptology – EUROCRYPT 2017 - 36th Annual International Conference on the Theory and Applications of Cryptographic Techniques, Proceedings
EditorsJean-Sebastien Coron, Jesper Buus Nielsen
PublisherSpringer Verlag
Pages580-608
Number of pages29
ISBN (Print)9783319566160
DOIs
StatePublished - 1 Jan 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10212 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Fingerprint

Dive into the research topics of 'Ad hoc PSM protocols: Secure computation without coordination'. Together they form a unique fingerprint.

Cite this