Isogeny-based Primitives, Optimisations, and Dynamical Systems

Objective

The advent of large-scale quantum computers threatens the security of classical public-key cryptography, since the underlying factorization and discrete logarithm problems can be efficiently solved using quantum algorithms. Isogeny-based cryptography has emerged as a promising post-quantum alternative, relying on the hardness of computing isogenies, i.e. rational maps between elliptic curves. Despite remarkable progress, current isogeny-based schemes face significant challenges concerning flexibility, efficiency, and parameter generation. Moreover, the hard problems on which their security relies are relatively recent, making their analysis an essential and ongoing endeavour. IsoPODS (Isogeny-based Primitives, Optimisations, and Dynamical Systems) addresses these issues through three complementary directions. First, it investigates the design and optimisation of cryptographic primitives based on group actions, instantiating recent blind signature schemes (Tanuki) with state-of-the-art isogeny-based group actions (PEGASIS), and exploring the possibility of constructing advanced schemes including blind, aggregate, and multi-signatures. Second, it targets the optimisation of zero-knowledge proof frameworks for isogenies, leveraging different families of modular polynomials to obtain algebraically compact representations suited for zk-SNARK/IOP-compatible arithmetisations. These can be used, in particular, to generate supersingular elliptic curves with unknown endomorphism ring via a multi-party ceremony. Third, it addresses the latter task – but without the need for the multi-party setting – by studying discrete dynamical systems on the set of elliptic curves, with the further purpose of finding structural weaknesses in isogeny graphs. Combining tools from algebraic geometry, number theory, and cryptography, the project intends to deliver new theoretical insights and frameworks that can inform the design of secure and efficient isogeny-based cryptosystems.