Paper 2023/540

MAYO: Optimized Implementation with Revised Parameters for ARMv7-M

Arianna Gringiani, University of Trento
Alessio Meneghetti, University of Trento
Edoardo Signorini, Telsy
Ruggero Susella, STMicroelectronics (Italy)
Abstract

We present an optimized constant-time implementation of the MAYO signature scheme on ARMv7-M. MAYO is a novel multivariate proposal based on the trapdoor function of the Unbalanced Oil and Vinegar scheme. Our implementation builds on existing techniques for UOV-based schemes and introduces a new approach for evaluating the polar forms of quadratic maps. We modify MAYO's original parameters to achieve greater benefits from the proposed optimizations, resulting in slightly larger keys and shorter signatures for the same level of security. We evaluate the optimized implementation with the new parameters on the STM32H753ZIT6 microcontroller and measure its performance for the signing and verification procedures. At NIST security level I, signing requires approximately 43M cycles, and verification requires approximately 6M cycles. Both are 2.6 times faster than the results obtained from the original parameters.

Note: 05/07/2023: Minor revision. Added link to GitHub implementation built into pqm4 framework.

Metadata
Available format(s)
PDF
Category
Implementation
Publication info
Preprint.
Keywords
MAYOpost-quantum cryptographymultivariate quadratic cryptographyARMv7-M
Contact author(s)
a gringiani @ libero it
alessio meneghetti @ unitn it
edoardo signorini @ telsy it
ruggero susella @ st com
History
2023-07-05: revised
2023-04-15: received
See all versions
Short URL
https://ia.cr/2023/540
License
Creative Commons Attribution
CC BY

BibTeX

@misc{cryptoeprint:2023/540,
      author = {Arianna Gringiani and Alessio Meneghetti and Edoardo Signorini and Ruggero Susella},
      title = {MAYO: Optimized Implementation with Revised Parameters for ARMv7-M},
      howpublished = {Cryptology ePrint Archive, Paper 2023/540},
      year = {2023},
      note = {\url{https://eprint.iacr.org/2023/540}},
      url = {https://eprint.iacr.org/2023/540}
}
Note: In order to protect the privacy of readers, eprint.iacr.org does not use cookies or embedded third party content.