Filtrer par type de contenu

Sélectionnez un ou plusieurs filtres. Ce choix permettra de recharger la page pour afficher les résultats filtrés.

Supprimer tous les filtres

634 résultats

    • Séminaire

    • Cryptographie

    Evaluation embarquée de l'entropie d'un générateur d'aléa pour la cryptographie

    • 12 janvier 2024

    • IRMAR - Université de Rennes - Campus Beaulieu Bat. 22, RDC, Rennes - Amphi Lebesgue

    Orateur : David Lubicz - DGA et IRMAR

    Dans cet exposé, nous présentons une méthode de mesure embarquée de la composante thermique du jitter de phase d'un couple d'anneaux d'oscillateurs : cette mesure est nécessaire pour calculer le taux d'entropie d'un générateur d'aléa physique basé sur des anneaux oscillants. La précision de cette mesure dont le principe repose sur un algorithme de type Monte-Carlo dépend de l' équidistribution de[…]
    • Séminaire

    • Cryptographie

    Recent advances in post-quantum signatures based on the MPC-in-the-Head paradigm

    • 22 mars 2024

    • IRMAR - Université de Rennes - Campus Beaulieu Bat. 22, RDC, Rennes - Amphi Lebesgue

    Orateur : Thibauld Feneuil - CryptoExperts

    Zero-knowledge proofs of knowledge are useful tools for designing signature schemes. Among the existing techniques, the MPC-in-Head (MPCitH) paradigm provides a generic framework to build quantum-resilient proofs using techniques from secure multiparty computation. This paradigm has recently been improved in a series of works which makes it an effective and versatile tool. In this talk, I will[…]
    • Séminaire

    • Cryptographie

    Correction of dual lattice attacks

    • 19 avril 2024

    • IRMAR - Université de Rennes - Campus Beaulieu Bat. 22, RDC, Rennes - Amphi Lebesgue

    Orateur : Kevin Carrier - CY Cergy

    The security of cryptographic systems such as Kyber and Dilithium (currently undergoing standardization by NIST) fundamentally hinges on the Learning With Errors (LWE) problem. However, the state of the art for attacking this problem is not yet entirely clear. Specifically, the validity of dual attacks on LWE is currently under scrutiny due to their analyses relying on heuristics that have proven[…]
    • Séminaire

    • Cryptographie

    ECDSA White-Box Implementations, Feedback on CHES 2021 WhibOx Contest

    • 29 septembre 2023

    • IRMAR - Université de Rennes - Campus Beaulieu Bat. 22, RDC, Rennes - Amphi Lebesgue

    Orateur : Agathe Houzelot - Idemia

    Cryptographic algorithms are primarily designed to be secure in the black-box model, where an attacker can only observe their input/output behavior. However in practice, algorithms are rarely executed in a completely isolated environment and additional information is often leaked. In the context of mobile applications or connected objects, devices often lack secure storage to protect secret keys,[…]
    • Séminaire

    • Cryptographie

    Password-Authenticated Key Exchange (PAKE) from Isogenies

    • 08 décembre 2023

    • IRMAR - Université de Rennes - Campus Beaulieu Bat. 22, RDC, Rennes - Amphi Lebesgue

    Orateur : Sabrina Kunzweiler - Université Bordeaux

    The passwords that we use in our everyday life are often chosen to be easily memorable which makes them vulnerable to attacks. This problem is addressed by password-authenticated key exchange (PAKE). The general idea is to enable two parties who share the same (potentially weak) password to establish a strong session key. Most PAKE protocols used today are based on Diffie-Hellman key exchange in[…]
    • Séminaire

    • Cryptographie

    Generic Attack on Duplex-Bases AEAD Modes using Random Function Statistics

    • 22 septembre 2023

    • IRMAR - Université de Rennes - Campus Beaulieu Bat. 22, RDC, Rennes - Amphi Lebesgue

    Orateur : Rachelle Heim Boissier - Inria

    Duplex-based authenticated encryption modes with a sufficiently large key length are proven to be secure up to the birthday bound 2^(c/2), where c is the capacity. However this bound is not known to be tight and the complexity of the best known generic attack, which is based on multicollisions, is much larger: it reaches 2^c/α where α represents a small security loss factor. There is thus an[…]