SLAMP-FSS: Two-Party Multi-Point Function Secret Sharing from Simple Linear Algebra

Erki Külaots; Toomas Krips; Hendrik Eerikson; Pille Pullonen-Raudvere

Paper 2024/1394

SLAMP-FSS: Two-Party Multi-Point Function Secret Sharing from Simple Linear Algebra

Erki Külaots, University of Tarty

Toomas Krips

, University of Tartu

Hendrik Eerikson

, Cybernetica AS, University of Tartu

Pille Pullonen-Raudvere

, Cybernetica AS

Abstract

Multiparty computation (MPC) is an important field of cryptography that deals with protecting the privacy of data, while allowing to do computation on that data. A key part of MPC is the parties involved having correlated randomness that they can use to make the computation or the communication between themselves more efficient, while still preserving the privacy of the data. Examples of these correlations include random oblivious transfer (OT) correlations, oblivious linear-function evaluation (OLE) correlations, multiplication triples (also known as Beaver triples) and one-time truth tables. Multi-point function secret sharing (FSS) has been shown to be a great building block for pseudo-random correlation generators. The main question is how to construct fast and efficient multi-point FSS schemes. Here we propose a natural generalization of the scheme of Boyle et al 2016 using a tree structure, a pseudorandom generator and systems of linear equations. Our schemes SLAMP-FSS and SLAMPR-FSS are more efficient in the evaluation phase than other previously proposed multi-point FSS schemes while being also more flexible and being similar in other efficiency parameters.

Note: Fixing LaTex compilation mistakes. No change in content.

Metadata

Available format(s): PDF
Category: Cryptographic protocols
Publication info: Preprint.
Keywords: function secret sharing secret sharing
Contact author(s): erki kulaots @ ut ee
toomas krips @ ut ee
henrik eerikson @ cyber ee
pille pullonen-raudvere @ cyber ee
History: 2024-09-13: revised; 2024-09-05: received; See all versions
Short URL: https://ia.cr/2024/1394
License: CC BY

BibTeX

@misc{cryptoeprint:2024/1394,
      author = {Erki Külaots and Toomas Krips and Hendrik Eerikson and Pille Pullonen-Raudvere},
      title = {{SLAMP}-{FSS}: Two-Party Multi-Point Function Secret Sharing from Simple Linear Algebra},
      howpublished = {Cryptology {ePrint} Archive, Paper 2024/1394},
      year = {2024},
      url = {https://eprint.iacr.org/2024/1394}
}