Projet de recherche doctoral numero :7507

Description

Date depot: 1 janvier 1900
Titre: Continuous-variable quantum key distribution with a discrete modulation
Directeur de thèse: Anthony LEVERRIER (Inria-Paris (ED-130))
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Non defini

Resumé: Quantum key distribution (QKD), which allows two honest parties to distill a secret key despite the presence of arbitrarily powerful adversaries, is by now a well-understood technology and attempts at its large-scale deployment are underway in China or Europe for instance. Continuous-variable QKD protocols encode information in the quantized quadratures of the electromagnetic field, which can be measured with coherent detection with rather standard telecom equipment. This feature has attracted the interest of major industrial leaders in optical communication and the crucial question now is whether continuous-variable (CV) QKD is indeed the right approach to deploy QKD at large scale. A drawback of CV QKD compared to discrete-variable QKD is that its security is much less advanced. There a several reasons for that: first, CV quantum cryptography was invented 15 years later than BB84; second, security proofs for CV QKD are objectively more involved than their counterpart for discrete-variable protocols. The reason for this second point is that the relevant Hilbert space isn’t a finite-dimensional space anymore but rather an infinite-dimensional Fock space and therefore most of the techniques invented to prove the security of protocols like BB84 do not apply in the CV case. New techniques were recently developed, mostly mostly by the PI of this project, to establish the security of the simplest CV QKD protocols. Here, by simplest protocols, we mean the ones of where one party prepares coherent states centered around points in phase-space drawn from a Gaussian distribution. While very appealing from a theoretic point of view, these protocols are less practical than ones relying on a finite constellation of coherent states. establishing the full security of such protocols with a discrete modulation has remained elusive. The past year has seen the first significant progress on this question of the past decade with the development of techniques allowing to compute the secret key rate of protocols with the simplest discrete modulation consisting of 4 coherent states. The goals of this thesis are to generalize these first steps to more complex constellations of coherent states (which should lead to better performance) and to improve the security analysis by including finite-size effects which are crucial to compute the key size in any given experiment.

Doctorant.e: Denys Aurélie