Many a Mickle Makes a Muckle: A Framework for Provably Quantum-Secure Hybrid Key Exchange

10Citations
Citations of this article
17Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Hybrid Authenticated Key Exchange (AKE) protocols combine keying material from different sources (post-quantum, classical, and quantum key distribution (QKD)) to build protocols that are resilient to catastrophic failures of the different components. These failures may be due to advances in quantum computing, implementation vulnerabilities, or our evolving understanding of the quantum (and even classical) security of supposedly quantum-secure primitives. This hybrid approach is a prime candidate for initial deployment of post-quantum-secure cryptographic primitives because it hedges against undiscovered weaknesses. We propose a general framework HAKE for analysing the security of such hybrid AKE protocols. HAKE extends the classical Bellare-Rogaway model for AKE security to encompass forward security, post-compromise security, fine-grained compromise of different cryptographic components, and more. We use the framework to provide a security analysis of a new hybrid AKE protocol named Muckle. This protocol operates in one round trip and leverages the pre-established symmetric keys that are inherent to current QKD designs to provide message authentication, avoiding the need to use expensive post-quantum signature schemes. We provide an implementation of our Muckle protocol, instantiating our generic construction with classical and post-quantum Diffie-Hellman-based algorithmic choices. Finally, we report on benchmarking exercises against our implementation, examining its performance in terms of clock cycles, elapsed wall-time, and additional latency in both LAN and WAN settings.

Cite

CITATION STYLE

APA

Dowling, B., Hansen, T. B., & Paterson, K. G. (2020). Many a Mickle Makes a Muckle: A Framework for Provably Quantum-Secure Hybrid Key Exchange. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 12100 LNCS, pp. 483–502). Springer. https://doi.org/10.1007/978-3-030-44223-1_26

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