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

Rui Qiang Wang, Zhen Qiang yin, Rong Wang, Shuang Wang, Wei Chen, Guang can Guo, Zhen fu HanarXiv 2022 · DOI · arXiv

TL;DR

Researchers have developed the first security proof for Twin-Field Quantum Key Distribution (TF-QKD) using discrete phase randomization in the finite-key regime. By proving that just 8 discrete phases are sufficient to maintain high security and performance, this work moves TF-QKD out of the realm of theoretical "continuous" assumptions and closer to practical, high-speed experimental reality.

What Problem Does This Paper Address?

Quantum Key Distribution (QKD) is the gold standard for secure communication, but it faces a major physical hurdle: the "rate-loss limit." In standard point-to-point setups, the key rate drops linearly with channel loss. Twin-Field (TF) QKD changed the game by allowing the key rate to scale with the square root of the loss ($\sqrt{\eta}$), effectively beating the fundamental limits of distance.

However, most TF-QKD security proofs rely on a shaky assumption: that the phase of the light pulses can be randomized perfectly across a continuous range from 0 to $2\pi$. In a real lab, equipment (like phase modulators) can only handle a finite, discrete set of values. Furthermore, most proofs assume an infinite amount of data, whereas real-world sessions use a finite number of pulses. Until now, there was no rigorous proof that TF-QKD remains secure when you combine discrete phases with finite-key analysis.

Methods

The authors developed a new analytical framework to bound the "phase error rate"—the metric that tells us how much information a potential eavesdropper (Eve) might have intercepted.

  1. Equivalent Virtual Protocol: Like many QKD proofs, they start by imagining a "virtual" version of the protocol involving entangled states. This allows them to use the property of conjugate measurements to estimate errors.
  2. Lemma 1 (The "Gap" Bounder): This is the paper's core mathematical contribution. They developed a lemma to bound the difference in "yields" (the probability of a detection event) between two very similar quantum states in a non-asymptotic (finite-data) situation.
  3. Linear Programming: They fed these constraints into a linear programming model to find the worst-case scenario for the phase error rate. If the protocol is secure even in this worst-case scenario, it’s secure for the user.

Datasets

As a theoretical and computational study, this paper does not use external datasets. Instead, it uses numerical simulations based on realistic experimental parameters:

  • Fiber loss: 0.2 dB/km.
  • Detector dark counts: $1 \times 10^{-8}$ per pulse.
  • Total pulses ($N_{tot}$): Simulated at $10^{12}$, $10^{13}$, and $10^{14}$ to show how the key rate improves with longer sessions.

Key Insights and Ideas

  • 8 is Enough: One of the most practical findings is that using just 8 discrete phases (e.g., $\{0, \pi/4, \pi/2, \dots, 7\pi/4\}$) achieves performance nearly identical to the idealized continuous case. This is a huge relief for experimentalists.
  • Beating the Bound: The analysis confirms that even with discrete phases and finite data, TF-QKD can still "break" the linear rate-loss bound at distances beyond ~250 km.
  • Versatility: While the paper focuses on TF-QKD, the authors note that their "Lemma 1" technique is a general tool. It can be adapted to prove the security of other protocols, like the classic BB84, when they use discrete phase randomization.

Key Results

The authors’ simulations (visualized in Figure 1 of the paper) show that:

  • Distance: Secret keys can be generated at distances exceeding 400 km.
  • The "Break" Point: The protocol successfully surpasses the linear rate-loss limit at approximately 250 km.
  • Finite-Key Effects: As expected, larger total pulse numbers ($N_{tot} = 10^{14}$) yield significantly better key rates and longer distances than smaller sets ($10^{12}$), but even the smaller sets remain functional and secure.

Limitations and Future Work

While this paper provides a vital proof for discrete phase randomization, it assumes that the discrete phases are chosen uniformly. In real systems, there might be small "phase fluctuations" or inaccuracies in the modulator that aren't perfectly uniform.

The authors suggest that future work could extend this analysis to account for even more experimental imperfections, such as intensity fluctuations in the "decoy states" used to thwart certain types of attacks. Additionally, applying this framework to other emerging QKD variants (like phase-matching QKD) remains an open and promising direction.

Summary updated

Abstract

Quantum key distribution(QKD) allows two remote parties to share information-theoretic secret keys. Many QKD protocols assume the phase of encoding state can be continuous randomized from 0 to 2 pi, which, however, may be questionable in experiment. This is particularly the case in the recently proposed twin-field(TF) QKD, which has received a lot of attention, since it can increase key rate significantly and even beat some theoretical rate-loss limits. As an intuitive solution, one may introduce discrete phase-randomization instead of continuous one. However, a security proof for a QKD protocol with discrete phase-randomization in finite-key region is still missing. Here we develop a technique based on conjugate measurement and quantum state distinguishment to ana-lyze the security in this case. Our result shows that TF-QKD with reasonable number of discrete random phases, e.g. 8 phases from {0, pi/4, pi/2, ..., 7pi/4}, can achieve satisfactory performance. More importantly, as a the first proof for TF-QKD with discrete phase-randomization in finite-key region, our method is also applicable in other QKD protocols.
BibTeX
@misc{wang2022finite,
  author = {Rui Qiang Wang and Zhen Qiang yin and Rong Wang and Shuang Wang and Wei Chen and Guang can Guo and Zhen fu Han},
  title = {{Finite-key analysis for quantum key distribution with discrete phase randomization}},
  year = {2022},
  eprint = {2201.03039},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2201.03039},
}

Loading related content...

Figure