Víctor Zapatero

The performance of quantum key distribution (QKD) heavily depends on statistical inference. For a broad class of protocols, the central statistical task is a random sampling problem, customarily addressed using exponential tail bounds on the hypergeometric distribution. Here, we provide an alternative solution for this task of unprecedented tightness among QKD security analyses. As a by-product, confidence intervals for the average of non-identical Bernoulli parameters follow too. These naturally fit in statistical analyses of decoy-state QKD and also outperform standard tools. Lastly, we show that, in a vast parameter regime, the use of tail bounds is not enforced because the cumulative mass function of the hypergeometric distribution is accurately computable. This sharply decreases the minimum block sizes necessary for QKD, and reveals the tightness of our simple analytical bounds when moderate-to-large blocks are considered. Mannalath, V., Zapatero, V., & Curty, M. (2024). Sharp finite statistics for quantum key distribution. arXiv:2410.04095 (2024). Currently under consideration in Phys. Rev. Lett. (second round of revision).

Recently, different modulator-free decoy-state quantum key distribution transmitters have been proposed. Among their advantages, they are essentially immune to information leakage, including that potentially induced by an adversary via e.g. a Trojan-horse attack. However, practical implementations of these transmitters emit, in addition to the desired signals, some extra pulses that are not used as quantum carriers, but still may contain sensitive information about the intensity and bit/basis encoding of the signals. This unwanted pulses can be actively blocked with an intensity modulator (or an optical switch), but the extinction ratio of these devices is always finite, and thus it is still crucial to account for the residual amount of information leakage at the security-proof level. In this work, we analyze the security of these transmitters and evaluate their performance in the presence of this kind of inherent information leakage. We find that the secret-key rate of the protocol is severely affected when the information leakage is not sufficiently attenuated, which highlights the importance of accounting for such type of imperfections.

Quantum Key Distribution (QKD) is a crucial technology for secure communication, relying on the principles of quantum mechanics. The security of QKD protocols is often analyzed by bounding the probability of a "failure" during the parameter estimation step. This failure probability is typically addressed using tail bounds on the hypergeometric distribution. However, existing methods can sometimes be conservative, leading to inefficiencies. In this work, we present an alternative approach that provides a more refined bound by exploiting a simple yet effective link between hypergeometric and binomial random variables.

The decoy-state method is a prominent approach to enhance the performance of quantum key distribution (QKD) systems that operate with weak coherent laser sources. Current experimental decoy-state QKD setups increase their secret key rate by raising the repetition rate of the transmitter, which can lead to correlations between subsequently emitted optical pulses. This phenomenon leaks information about the encoding settings, including the intensities of the generated signals, thus invalidating a basic premise of decoy-state QKD. Here, we experimentally characterize intensity correlations between the nearest-neigbouring optical pulses in two commercial prototypes of decoy-state BB84 QKD systems and show that they significantly reduce the asymptotic key rate. In addition, we study intensity correlations between pulses spaced further apart (higher-order correlations) and find that, in contrast to what has been conjectured, their impact on the intensity of the generated signals can be much higher than that of the nearest-neighbour (first-order) correlations.