Tight finite-key security for twin-field quantum key distribution

Abstract

Quantum key distribution (QKD) offers a reliable solution to communication problems that require long-term data security. For its widespread use, however, the rate and reach of QKD systems must be improved. Twin-field (TF) QKD is a step forward toward this direction, with early demonstrations suggesting it can beat the current rate-versus-distance records. A recently introduced variant of TF-QKD is particularly suited for experimental implementation, and has been shown to offer a higher key rate than other variants in the asymptotic regime, where users exchange an infinite number of signals. Here, we extend the security of this protocol to the finite-key regime, showing that it can overcome the fundamental bounds on point-to-point QKD with ~1010 transmitted signals. In many practical regimes of interest, our analysis offers higher key rates than those of alternative variants. Moreover, some of the techniques we develop are applicable to the finite-key analysis of other QKD protocols.

Coherent phase transfer for real-world twin-field quantum key distribution

“…This idea, sketched in Fig. 1 a, was proved secure against general attacks 22 – 26 also in the finite-size scenario 27 – 29 and with the aid of two-way communication 30 , but it is based on the critical assumptions that the optical pulses are phase-coherent in Alice and Bob and preserve coherence throughout the path to Charlie. While the first requirement can be fulfilled by phase-locking the two QKD lasers in Alice and Bob to a common reference laser transmitted through a service channel, the uncorrelated fluctuations of the length and refractive index of the fibers caused by environmental acoustic noise and temperature changes introduce phase noise to the system and reduce the visibility of the interference measurement.…”

Finite-key analysis of loss-tolerant quantum key distribution based on random sampling theory

“…Namely, it requires users to attempt to predict the results that they expect to obtain in the experiment, before they actually run the experiment. This is an important step, since the inequality is only tight when the actual experimental data was reasonable close to their predictions [25].…”

“…Clearly, E[K 2 ] = (1 − p)n, and therefore n = E[K 2 ]/(1 − p). Using the inverse multiplicative Chernoff bound [25,31,32], we have that…”

Finite-key analysis for quantum key distribution with discrete phase randomization

“…Following the symbols in [11], let us consider the evolution of the gigantic quantum state |Φ = |φ ⊗Ntot = (|ψ AsAcAa ⊗ |ψ BsBcBb ) ⊗Ntot sent to Eve. After Step 2, where Eve performs her measurement on the subsystem ab, the initial quantum state is transformed to Meve |Φ where Meve denotes the measurement operator of Eve.…”