Paper 2023/984 Original: eprint.iacr.org/2023/984 ↗

Generating Supersingular Elliptic Curves over $\mathbb{F}_p$ with Unknown Endomorphism Ring

Youcef Mokrani
David Jao
Abstract

A number of supersingular isogeny based cryptographic protocols require the endomorphism ring of the initial elliptic curve to be either unknown or random in order to be secure. To instantiate these protocols, Basso et al. recently proposed a secure multiparty protocol that generates supersingular elliptic curves defined over $\mathbb{F}_{p^2}$ of unknown endomorphism ring as long as at least one party acts honestly. However, there are many protocols that specifically require curves defined over $\mathbb{F}_p$, for which the Basso et al. protocol cannot be used. Also, the simple solution of using a signature scheme such as CSI-FiSh or SeaSign for proof of knowledge either requires extensive precomputation of large ideal class groups or is too slow for everyday applications. In this paper, we present CSIDH-SCG, a new multiparty protocol that generates curves of unknown endomorphism ring defined over $\mathbb{F}_p$. This protocol relies on CSIDH-IP, a new CSIDH based proof of knowledge. We also present CSIDH-CR, a multiparty algorithm that be used in conjunction with CSIDH-SCG to generate a random curve over $\mathbb{F}_p$ while still keeping the endomorphism ring unknown.

Metadata
Available format(s)
PDF
Category
Public-key cryptography
Publication info
Preprint.
History
2024-05-21: last updated
2023-06-23: published
Short URL
https://ia.cr/2023/984
License
https://creativecommons.org/licenses/by-sa/4.0/

BibTeX

@misc{cryptoeprint:2023/984,
      author = {Youcef Mokrani and David Jao},
      title = {Generating Supersingular Elliptic Curves over $\mathbb{F}_p$ with Unknown Endomorphism Ring},
      howpublished = {Cryptology {ePrint} Archive, Paper 2023/984},
      year = {2023},
      url = {https://eprint.iacr.org/2023/984}
}
Note: In order to protect the privacy of readers, this archive does not use cookies or embedded third party content.