Pub. online:4 Jun 2025Type:Research ArticleOpen Access
Journal:Informatica
Volume 36, Issue 2 (2025), pp. 315–335
Abstract
Establishing secure keys over untrusted networks is one of the most fundamental cryptographic tasks. While two-party key establishment protocols are available for many scenarios, even offering resistance to potential adversaries equipped with quantum computing resources, the multi-party scenario is not as well understood. In particular, there is a need to find designs that can make the most of the technologies available to each party involved in a cooperative n-party key establishment.
We propose an authenticated key establishment protocol involving $n\geqslant 2$ parties, assuming that some—possibly all—network nodes have the potential to implement quantum key distribution (in pairs), while others only have access to standard technology. The protocol allows for the cooperative construction of a shared secret key from partial keys established by quantum and post-quantum solutions, which in turn can be implemented by different building blocks. We give a formal security analysis of our proposal using a hybrid security model simultaneously capturing quantum and classical actions and capabilities.
Journal:Informatica
Volume 31, Issue 4 (2020), pp. 751–768
Abstract
In cryptography, key establishment protocols are often the starting point paving the way towards secure execution of different tasks. Namely, the parties seeking to achieve some cryptographic task, often start by establishing a common high-entropy secret that will eventually be used to secure their communication. In this paper, we put forward a security model for group key establishment ($\mathsf{GAKE}$) with an adversary that may execute efficient quantum algorithms, yet only once the execution of the protocol has concluded. This captures a situation in which keys are to be established in the present, while security guarantees must still be provided in the future when quantum resources may be accessible to a potential adversary.
Further, we propose a protocol design that can be proven secure in this model. Our proposal uses password authentication and builds upon efficient and reasonably well understood primitives: a message authentication code and a post-quantum key encapsulation mechanism. The hybrid structure dodges potential efficiency downsides, like large signatures, of some “true” post-quantum authentication techniques, making our protocol a potentially interesting fit for current applications with long-term security needs.