Pairing the Volcano

TL;DR

This paper introduces an efficient method to identify the direction of steps on isogeny volcanoes, improving the speed of navigating these graphs and computing endomorphism rings of elliptic curves.

Contribution

It presents a novel approach to predict isogeny directions on volcanoes and derives a new invariant for endomorphism class determination, enhancing existing algorithms.

Findings

Faster identification of horizontal and ascending isogenies.

Improved algorithm for endomorphism ring computation.

Effective for small prime degrees .

Abstract

Isogeny volcanoes are graphs whose vertices are elliptic curves and whose edges are $\ell$-isogenies. Algorithms allowing to travel on these graphs were developed by Kohel in his thesis (1996) and later on, by Fouquet and Morain (2001). However, up to now, no method was known, to predict, before taking a step on the volcano, the direction of this step. Hence, in Kohel's and Fouquet-Morain algorithms, many steps are taken before choosing the right direction. In particular, ascending or horizontal isogenies are usually found using a trial-and-error approach. In this paper, we propose an alternative method that efficiently finds all points $P$ of order $\ell$ such that the subgroup generated by $P$ is the kernel of an horizontal or an ascending isogeny. In many cases, our method is faster than previous methods. This is an extended version of a paper published in the proceedings of ANTS 2010. In addition, we treat the case of 2-isogeny volcanoes and we derive from the group structure of the curve and the pairing a new invariant of the endomorphism class of an elliptic curve. Our benchmarks show that the resulting algorithm for endomorphism ring computation is faster than Kohel's method for computing the $\ell$-adic valuation of the conductor of the endomorphism ring for small $\ell$.