Abstract
Twin-field quantum key distribution (TF-QKD) and its variant protocols are highly attractive due to the advantage of overcoming the rate-loss limit for secret key rates of point-to-point QKD protocols. For variations of TF-QKD, the key point to ensure security is switching randomly between a code mode and a test mode. Among all TF-QKD protocols, their code modes are very different, e.g. modulating continuous phases, modulating only two opposite phases, and sending or not sending signal pulses. Here we show that, by discretizing the number of global phases in the code mode, we can give a unified view on the first two types of TF-QKD protocols, and demonstrate that increasing the number of discrete phases extends the achievable distance, and as a trade-off, lowers the secret key rate at short distances due to the phase post-selection.
Introduction
Quantum Key Distribution (QKD) provides a method for two parties, Alice and Bob, to share a secret key with information-theoretic security. However, point-to-point QKD protocols are fundamentally limited by the PLOB bound, which defines the maximum secret key rate achievable over a lossy channel. Twin-Field QKD (TF-QKD) was proposed to overcome this limit by utilizing a central untrusted node for interference measurements, effectively scaling the key rate with the square root of the transmittance rather than the transmittance itself.
While various TF-QKD implementations exist—ranging from continuous phase modulation to simple binary phase modulation—a unified security framework for discrete phase randomization has been lacking. This work addresses this gap by providing a rigorous security proof for TF-QKD with $2M$ discrete phases, offering a bridge between complex continuous systems and simpler discrete ones.
Methodology
The proposed protocol utilizes a discrete phase randomization scheme where Alice and Bob modulate their coherent states using one of $2M$ possible phases. The protocol alternates between a "code mode" for key generation and a "test mode" for parameter estimation.
Security is analyzed against collective attacks using a mathematical framework based on von Neumann entropy and Holevo information. By applying Jensen’s inequality and exploiting the concavity of Shannon entropy, the researchers derived a lower bound for the secret key rate that applies to any integer $M$. This approach effectively unifies previous TF-QKD variants into a single theoretical model, proving that as $M$ tends toward infinity, the discrete protocol becomes mathematically identical to Phase-Matching (PM) QKD.
Results and Performance
Numerical simulations were conducted using realistic experimental parameters, including a dark count rate of $8 \times 10^{-8}$, an error correction efficiency of $1.15$, and a detector efficiency of $14.5\%$. The results demonstrate that the discrete phase protocol successfully surpasses the PLOB bound at high channel losses.
As shown in Figure 1, increasing the number of discrete phases ($M$) extends the maximum achievable distance. However, this comes at the cost of a lower secret key rate at shorter distances due to the "sifting cost" associated with matching a larger set of phase choices. The simulation highlights a critical "sweet spot" at $M=2$.
The performance limits summarized in Figure 3 reveal a significant performance jump when moving from $M=1$ (72.3 dB) to $M=2$ (80.8 dB). Further increasing $M$ yields diminishing returns; $M=4$ reaches 81.3 dB, while the infinite phase limit ($M \to \infty$) caps at 81.5 dB.
Conclusion
This study provides a unified security proof for TF-QKD with discrete phase randomization. The findings demonstrate that $M=2$ (four discrete phases) represents an optimal balance for practical implementation, offering a substantial distance increase over binary modulation while maintaining higher key rates than more complex schemes at moderate distances. This framework simplifies the theoretical landscape of TF-QKD and provides clear guidance for future experimental developments in long-distance quantum communication.
Summary updated