Published by the Department of Computer Science, The University of Chicago.
Antoine Joux
CISPA Helmhotz Center for Information Security
Saarbrücken, Germany
antoine DOT joux AT mdx DOT org
and
Miklos Santha
Centre for Quantum Technologies and MajuLab
National University of Singapore
Singapore
miklos DOT santha AT gmail DOT com
We investigate the computational complexity of the discrete logarithm, the computational Diffie-Hellman and the decisional Diffie-Hellman problems in some identity black-box groups $G_{p,t}$, where $p$ is a prime number and $t$ is a positive integer. These are defined as quotient groups of vector space $Z_p^{t+1}$ by a hyperplane $H$ given through an identity oracle. While in general black-box groups that have unique encoding of their elements these computational problems are classically all hard and quantumly all easy, we find that in the groups $G_{p,t}$ the situation is more contrasted. We prove that while there is a polynomial time probabilistic algorithm to solve the decisional Diffie-Hellman problem in $G_{p,1}$, the probabilistic query complexity of all the other problems is $\Omega(p)$, and their quantum query complexity is $\Omega(\sqrt{p})$. Our results therefore provide a new example of a group where the computational and the decisional Diffie-Hellman problems have widely different complexity.
Submitted May 19 2021, revised January 5, 2023, published June 15, 2023.