Complete Addition Law for Montgomery Curves

3Citations
Citations of this article
7Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Montgomery curves allow efficient and side-channel resistant computation of ECDH using the Montgomery ladder. But the addition law of a Montgomery curve derived from the chord-tangent method is less efficient than other curve models such as a short Weierstrass curve and an Edwards curve. So, the usage of a Montgomery curve is strictly limited to ECDH only, such as X25519 and X448 functions in IETF RFC 7748. For other operations including fixed-base and multiple scalar multiplications, their birationally-equivalent (twisted) Edwards curves are recommended for use since the conversions between Montgomery curves and their Edwards equivalents are simple. This conversion enables the use of the efficient complete addition law of the Edwards curve that works for all pairs of input points with no exceptional cases. As a result, the combination allows secure and exception-free implementations, but at the expense of additional storage for the two curve parameters and for the conversion between them. However, smart devices in IoT environments that mainly operate ECDH (for example, RawPublicKey mode of IETF RFC 7250) do not need to implement such a conversion if a complete addition law does exist for the Montgomery curves. To make such implementations possible, we provide a complete addition law on Montgomery curves. The explicit formulas for the complete addition law are not as efficient as those of Edwards curves, but they can make the Montgomery curve addition operation more efficient compared to using the conversion to the (twisted) Edwards equivalent. We also confirmed the validity of the comparison by implementing such two methods of realizing the addition operation on Curve25519.

Cite

CITATION STYLE

APA

Kim, J., Park, J. H., Kim, D. C., & Kim, W. H. (2020). Complete Addition Law for Montgomery Curves. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11975 LNCS, pp. 260–277). Springer. https://doi.org/10.1007/978-3-030-40921-0_16

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free