WorksPhysics and Astronomy
Digital Coherent Twin-Field Quantum Key Distribution with Gaussian-Modulated Coherent States
Abstract
Quantum key distribution (QKD) enables information-theoretically secure key sharing between two distant parties. Recently, one promising QKD protocol which can distribute a secret key over long distances called twin-field QKD (TF-QKD) has been developed. However, it requires active compensation for experimental imperfections such as phase, polarization, and frequency fluctuations with expensive devices and this makes its experimental implementation challenging. Here, we propose a long-distance continuous-variable QKD (CV-QKD) protocol which is inspired by TF-QKD. Our protocol can remove these requirements by using Gaussian-modulated coherent states and digital signal processing which is widely used in recent digital coherent optical communication and experimental implementations of CV-QKD. For security analysis, we use a numerical optimization approach with a photon-number cutoff assumption. We propose a new set of constraints for numerical optimization which can efficiently constrain a highly non-Gaussian quantum state shared by two legitimate parties in our protocol. We show that the key rate of our protocol exceeds the key rate of a direct-transmission CV-QKD protocol at long distances and it can overcome the repeaterless bound.
INTRODUCTION
Quantum key distribution (QKD) enables two legitimate parties, Alice and Bob, to share an information-theoretically secure key even in the presence of an eavesdropper, Eve, whose capability is limited only by quantum mechanics. After the first proposal of a QKD protocol in 1984 [1], substantial theoretical and experimental progress on QKD has been made. See, e.g., Refs. [2, 3] for reviews.
Most QKD protocols are classified into one of the two categories: discrete-variable QKD (DV-QKD) and continuous-variable QKD (CV-QKD). In DV-QKD protocols such as BB84 [1] and E91 [4], a quantum state is measured with single-photon detectors. On the other hand, in CV-QKD protocols such as GG02 [5], a quantum state is measured with homodyne or heterodyne detectors. Although the homodyne and heterodyne detectors are compatible with current optical communication technology, CV-QKD is vulnerable to experimental noise, particularly excess noise, and a communication distance between Alice and Bob of CV-QKD protocols is shorter than that of DV-QKD protocols.
Twin-field QKD (TF-QKD), which is one of DV-QKD protocols, was proposed in Ref. [6]. The advantage of TF-QKD is that its key rate overcomes the fundamental rate-distance limit of QKD [7, 8, 9] over a lossy repeaterless channel. By utilizing single-photon interference, TF-
QKD can overcome this repeaterless bound and its key rate scales as the single-repeater bound [10]. Extensive theoretical [11, 12, 13, 14, 15, 16, 17, 18, 19] and experimental [20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35] research on TF-QKD has been conducted.
Implementation of TF-QKD, however, is experimentally challenging since it requires accurate stabilization of phase and polarization which fluctuate during channel transmission for the single-photon interference. In addition, frequencies of Alice’s and Bob’s light sources may deviate during the protocol. Then, frequency locking of two independent light sources is required, which complicates the implementation further. Usually, these requirements have been fulfilled by active compensation or locking with expensive devices and make TF-QKD less practical in QKD networks.
In this paper, we propose a novel TF-QKD protocol which we call the digital coherent TF-QKD (DC-TF-QKD). This is a CV counterpart of TF-QKD and we show that one can employ digital signal processing (DSP) techniques, commonly used in CV-QKD [36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49], to make the system free from active phase/polarization/frequency stabilization. More precisely, the phase, polarization, and frequency differences of the two fields are passively monitored and then compensated by classical postprocessing. In a standard (non-TF) CV-QKD, these are processed by DSP techniques originally developed for coherent optical communication [50]. We show that a similar approach is applicable in the TF-QKD setting.
Our protocol uses two weak coherent fields with Gaussian modulation (or its entanglement analogue) at two users, Alice and Bob, and these fields are sent to the central station, Charlie, with a 50:50 beamsplitter and two on-off-type photon detectors (see Figure 1). When one of the detectors at Charlie clicks, Alice and Bob share (virtual) entanglement which can be used to extract a secret key.

Figure 1. Schematic of DC-TF-QKD. (a) PM version. Alice and Bob randomly prepare coherent states from Gaussian distributions. They transmit them to a central node, Charlie, through pure-loss channels. Charlie interferes these two modes with a 50:50 beamsplitter and performs on-off measurement. Alice and Bob distil a secret key from successful events where only one of the two on-off detectors clicks. (b) EB version. Both Alice and Bob prepare TMSV and perform heterodyne measurement. The rest of the protocol is the same as the PM version.
Whereas the intuition and implementation are similar to the TF-QKD and the standard CV-QKD, its security analysis is not trivial. Security of CV-QKD is usually guaranteed by Gaussian optimality [51, 52]. In the DC-TF-QKD, however, Alice and Bob share a highly non-Gaussian state and the first and second moments do not fully characterize the heralded quantum state. Therefore, we utilize the numerical optimization approach [53, 54] with the photon-number cutoff assumption [55] with a new set of constraints which can effectively constrain the highly non-Gaussian state. By numerical optimization, we show that our protocol can overcome the repeaterless bound as in conventional TF-QKD.
Finally, we briefly compare our protocol with other existing TF-QKD variants without active phase stabilization. One such protocol is asynchronous measurement-device-independent QKD (AMDI-QKD) proposed in [56, 57] (also referred to as mode-pairing QKD [58]). This protocol removes the need for phase locking by pairing time slots within a block of the pulse train. One of the constraints of the protocol is that the coherence within the block should be maintained while the block length gets long in a highly lossy channel since the probability of mode pairing gets smaller. On the other hand, in Ref. [34], TF-QKD without phase locking has been demonstrated by utilizing high-rate single-photon detection and a fast-Fourier-transform-based algorithm in classical postprocessing. Nevertheless, although these protocols eliminate the need for active phase stabilization, our protocol can also compensate the polarization and frequency fluctuations in postprocessing.
PROTOCOL
The basic DC-TF-QKD protocol is illustrated in Figure 1. We propose two variants: (a) prepare-and-measure (PM) version and (b) entanglement-based (EB) version.
In the PM version, shown in Figure 1(a), Alice randomly prepares a coherent state from a Gaussian distribution of variance , and sends it to a (untrusted) central node, Charlie, through a pure-loss channel with transmissivity . Bob also randomly prepares a coherent state from a Gaussian distribution of variance , and sends it to Charlie through a pure-loss channel with transmissivity . Charlie interferes these two modes with a 50:50 beamsplitter. Then, he performs on-off detection and declares its click pattern. We describe the events where only one of the two detectors clicks as successful events. Here, the on-off detectors have detection efficiency of and dark count probability . For parameter estimation, Alice and Bob perform the numerical optimization, which is described later, by using a small subset of the successful events. If a result of this estimation shows a
non-zero secret key rate, they proceed with the protocol. Otherwise, they abort the protocol. By performing error correction and privacy amplification, they obtain a secret key.
In the EB version, shown in Figure 1(b), Alice and Bob each prepare a two-mode squeezed vacuum (TMSV) with mean photon number and per mode, respectively, and each transmit one mode to Charlie. can be expressed by as . is analogous to . Alice and Bob perform heterodyne measurement on their remaining modes and retain the results in the cases where only one of Charlie’s two detectors clicks. After they perform parameter estimation, error correction and privacy amplification, they share a secret key. The EB version is equivalent to the PM version if there are no imperfections on the heterodyne measurements since a Gaussian-modulated coherent state is equivalent to a TMSV state conditioned on an ideal heterodyne measurement of one mode.
PHASE/POLARIZATION/FREQUENCY NOISE COMPENSATION VIA POSTPROCESSING
In the transmission channel, such as optical fibers, optical phase and polarization are fluctuating in time. In addition, frequency of the laser sources at Alice and Bob may also fluctuate in time. In the experiments so far, some or all of them are usually actively stabilized [26, 30, 32, 33, 35]. In this section, we describe how to compensate phase, polarization, and frequency, without active stabilizations, via postprocessing in our DC-TF-QKD protocol. These compensation operations are implemented by pilot-tone signals and DSP, that are widely used in standard CV-QKD systems. The compensation scheme exploits the invariance of Gaussian modulation under phase rotation and, for two equally modulated polarization modes, under unitary polarization mixing. Therefore, these transformations can be incorporated into a classical relabeling of the preparation data.
Figure 2 shows the PM version of our protocol with these compensation functions. In Figure 2, Alice and Bob prepare Gaussian-modulated coherent states in horizontal and vertical polarizations independently. In other words, they implement the protocol of Figure 1(a) in both polarizations. To monitor the fluctuations, Alice and Bob also send pilot pulses, which are stronger coherent signals generated from the same laser sources as the quantum signals, in dedicated time slots. At Charlie’s station, the lower detection paths in Figure 2 are used to monitor the polarization of the pilot pulses from Alice and Bob separately. At the upper part, the signals are interfered by a 50:50 beamsplitter, and the four polarization-resolved single-photon detectors are used to detect both quantum signals and pilot pulses in their respective time slots. The relative phase can be estimated from the interference statistics of the pilot pulses, while the frequency difference can be monitored at Alice’s and Bob’s stations or estimated from the beat signal at Charlie. Alternatively, the pilot pulses and quantum signals can be frequency-multiplexed and separated by optical filters, allowing the pilot pulses to be measured using separate homodyne or heterodyne receivers. In the following, we describe how to compensate each fluctuation by postprocessing.

Figure 2. Schematic of phase, polarization, and frequency compensation in DC-TF-QKD. Alice and Bob send pilot tones to Charlie, and Charlie monitors its interference output to obtain information about phase difference between Alice and Bob. By using this information, Alice and Bob can achieve phase compensation with digital signal processing. Alice prepares two Gaussian-modulated coherent states with orthogonal polarization and , and Bob prepares and . By measuring polarization states after the channel transmission, they compensate their polarization states with postprocessing. In addition, by measuring frequencies of Alice’s and Bob’s signals and , respectively, and detection time of Charlie’s single-photon detectors, Alice and Bob can compensate phase fluctuations caused by frequency difference between Alice’s and Bob’s light sources. BS, beamsplitter; PBS, polarizing beamsplitter; SA, spectrum analyzer; TT, time tagger.
Relative phase fluctuation. Suppose Alice sends a coherent state , and Bob sends a coherent state . After the channel transmission, these coherent states are changed to and , where and are the phase fluctuations of Alice’s coherent state and Bob’s coherent state through the channel transmissions, respectively, and its difference, , is added as the phase noise. Here, we omit the effect of channel losses. By transmitting pilot tones produced by the same light sources as the quantum signals and monitoring their interference at Charlie’s site, we can obtain information about . Then, by using this information, Alice (Bob) performs postprocessing on her (his) data and achieves phase compensation. This is, for example, done by regarding Alice’s signal as instead of , which is possible since we use continuous Gaussian modulation.
Polarization rotation. We describe how to compensate the polarization of Alice’s signal while the same argument holds for Bob. Alice sends two Gaussian-modulated coherent states with orthogonal polarization and . After the transmission, its polarization state may be rotated. Then, a quantum state which corresponds to horizontal polarization after the channel transmission can be expressed by and as , where and are coefficients. The same holds for vertical polarization, that is, a quantum state which corresponds to vertical polarization after the channel transmission is , where the polarization mode of this state is orthogonal to that of . By sending pilot pulses with known polarization states and performing suitable polarization-resolved measurements after channel transmission, we can estimate and . Then, Alice supposes that she sent Gaussian-modulated coherent states and , and performs key rate analyses. As with Alice, Bob prepares two Gaussian-modulated coherent states and , and supposes that he sent or by using a pilot tone and postprocessing.
Frequency fluctuation. Finally, we explain how we eliminate the requirement of frequency locking of Alice’s and Bob’s light sources. To do so, we assume that the time resolution of Charlie’s photon detectors is much shorter than the pulse width of Alice and Bob’s quantum signals. This happens, for example, by using nanosecond pulses for the coherent states, which is typical in CV-QKD [59, 60, 61, 62], and superconducting single-photon detectors at Charlie, which have a time resolution of a few tens of picoseconds [63, 64].
Let and be the frequency of Alice’s and Bob’s sources, respectively. If their spectra do not overlap, they may not interfere. However, by utilizing time-resolved measurements, it is known that they can interfere [65]. An intuition behind this is as follows. When a measurement device has sufficiently fine time resolution, which is equivalent to a broad bandwidth in frequency domain, and both and are within this broad spectrum of the detector, the detector cannot distinguish which frequency component gave rise to the detection event. This intrinsic indistinguishability causes the quantum interference. In our protocol, we can realize single-photon interference between Alice’s and Bob’s signals by using a time-resolved measurement even if Alice’s frequency and Bob’s frequency are not the same. However, the frequency difference introduces an undesirable phase shift to their data. Alice and Bob can estimate the amount of the phase shift by measuring the frequencies and and detection time of the single-photon detectors at the Charlie’s site, and they can compensate it by applying phase rotation to their data. This is achieved by postprocessing and thus it can be free from active frequency locking. A simple but more formal justification is given in Appendix § 1.7.
Note that all these postprocessing operations have some compatibility with the techniques used in standard CV-QKD, which means that one can directly apply DSP techniques of CV-QKD to our protocol. In the following sections, we analyze the security for the elementary protocols in Figure 1. Detailed simulation of the implementation in Figure 2 is left for future work.
KEY RATE
The key rate of our protocol is expressed by the Devetak-Winter formula [66] as
where is the classical mutual information between Alice’s string and Bob’s string , is the Holevo information between Bob’s string and Eve’s information , is the probability that the Charlie’s on-off detection succeeds, is the reconciliation efficiency and is a feasible set that contains all density operators which are compatible with experimental constraints. Here, we must upper bound . For Gaussian CV-QKD protocols, we can use Gaussian optimality [51, 52] to upper bound this quantity. In our protocol, however, a quantum state conditioned on the successful central detection, which Alice and Bob share, is highly non-Gaussian. Therefore, we have to use another technique instead of Gaussian optimality. Then, we use the numerical optimization approach developed in [53, 54] with the photon-number cutoff assumption [55].
To apply this approach to our protocol, we rewrite the key rate formula (Eq. (1)) as follows:
where and are the Shannon entropy and the conditional von Neumann entropy, respectively. By applying the result of Ref. [67], we obtain the following expression:
where . is a completely positive and trace-nonincreasing map describing the key-map operation, where
is a Kraus operator. Here, is a register system and are region operators defined as
where and is the number of region operators. is a pinching quantum channel that dephases the register to read out the result of a key map:
where
is a projector. We can calculate the classical entropic functions by using the probability distribution .
In Eq. (3), we perform the optimization over all in the feasible set for estimating key rate. However, because acts on an infinite-dimensional Hilbert space, we cannot perform this optimization numerically. Therefore, we make finite dimensional. In our protocol, Alice’s Hilbert space and Bob’s Hilbert space which are spanned by the photon-number basis are infinite dimensional. Then, we make and finite dimensional by using the photon-number cutoff assumption [55], that is, we assume that and are spanned by a basis , where is a cutoff photon number. If we define , a quantum state under this assumption is . We investigate the dependence of the key rate on in Appendix B. In this paper, we set at 5.
Next, we describe the feasible set in the optimization problem Eq. (3). is expressed as
where is a set of operators and is the corresponding expectation values. must be chosen to constrain the highly non-Gaussian state efficiently. We choose the following constraints:
where , and we use all combinations of and as the constraints. is the identity operator, and are the quadrature operators, and and can be expressed by the creation operator and the annihilation operator as follows:
The quadrature operators are expressed by and in the following way
We can calculate these expectation values by using the Husimi -function. Details can be found in Appendix § 2.
RESULTS AND DISCUSSION
Here, we show results of the numerical optimization. The numerical results below consider the elementary protocol in Figure 1, and the effect of the DSP operations are not incorporated. We consider a symmetric scenario, that is, , where and is the distance between Alice and Bob in kilometers, and . In addition to channel loss, we also consider other imperfections, that is, dark count probability of the on-off detectors , efficiency of the on-off detectors and the reconciliation efficiency . We assume that the number of region operators throughout this paper unless otherwise stated. Unless otherwise stated, we numerically optimize the mean photon number to maximize the key rate.
In Figure 3, we show the key rate of DC-TF-QKD versus the distance . We take , and . The purple circles show key rates of our protocol which are obtained by numerical optimization. The blue solid line shows analytical key rates of our protocol where Alice and Bob can completely characterize their joint quantum state. We also plot the repeaterless bound (black dashed line) and the single-repeater bound (black dotted line). The scaling of the key rate of our protocol is the same as the single-repeater bound, and it overcomes the repeaterless bound at 300 km. Therefore, DC-TF-QKD has the same key rate scaling as the TF-QKD. In addition, the key rate obtained by the numerical optimization closely matches the analytical key rate. This means that the set of constraints we propose can yield a tight numerical key-rate bound for the quantum states in our protocol. In Appendix § 2.1, we investigate the effect of dark count probability and efficiency of the on-off detectors on the key rate.
Figure 3. Key rates of DC-TF-QKD versus the distance . Purple circles show key rates of our protocol which are obtained by numerical optimization. Blue solid line shows analytical key rates of our protocol. We use , , and . Black dashed line is the repeaterless bound and black dotted line is the single-repeater bound.
Finally, we compare performance of DC-TF-QKD with other QKD protocols. In Figure 4, we show the key rate of our protocol for , and . Also, we plot key rates of a direct transmission protocol and AMDI-QKD. In the direct transmission protocol, Alice prepares Gaussian-modulated coherent states and transmits them to Bob. Bob directly performs homodyne measurement on the coherent states. The orange dotted line is an analytical key rate of the direct transmission protocol [68], where we do not consider any imperfections. Although the key rate of the direct transmission protocol is high at short distances, we cannot overcome the repeaterless bound with the direct transmission protocol as opposed to our protocol. The light-blue solid line represents key rates of AMDI-QKD. Although the key rate of AMDI-QKD is higher than that of our protocol, the advantage of our protocol is that it eliminates not only active phase compensation but also active stabilization of polarization and frequency.
Figure 4. Key rate versus distance . Purple circles represent key rates of DC-TF-QKD where , and . Orange dotted line is key rate of the direct transmission protocol [68] where mean photon number of Alice’s coherent state is 5. Light blue solid line shows key rates of AMDI-QKD [57]. Black dashed line is the repeaterless bound.
CONCLUSION
In this paper, we propose DC-TF-QKD which has the same key-rate scaling as the single-repeater bound and can overcome the repeaterless bound, as in TF-QKD. Unlike TF-QKD, however, our protocol does not require active phase, polarization, and frequency stabilization thanks to digital signal processing. This fact makes experimental implementation of our protocol significantly simpler. For security analysis of our protocol, we use the recently developed numerical optimization approach with the photon-number cutoff assumption. We propose the new set of constraints for the numerical optimization which can efficiently constrain a highly non-Gaussian state and yield a tight numerical key-rate bound. Our idea of combining digital signal processing and TF-QKD may pave the way for future development of practical and long-distance QKD protocols.
Finally, we discuss directions for future work. First, we use the photon-number cutoff assumption to truncate the dimension of quantum states for security analysis. Removing this assumption is important to give more rigorous security. In addition, considering finite-size security [40, 69, 70] is a possible direction for future work. Another possibility of future work is finding a systematic way to make a set of constraints to constrain given quantum states, in particular, non-Gaussian states. Non-Gaussian states play an important role in CV quantum information processing [71]. However, estimating non-Gaussian states from limited information is difficult since its Hilbert space is infinite dimensional and we do not have specific parameters to characterize such state whereas Gaussian states are fully characterized by their first and second moments. Therefore, it is important to find a systematic way to constrain non-Gaussian state by using limited information.
Acknowledgments
This work was supported by JST Moonshot R&D, Grant No. JPMJMS226C and Grant No. JPMJMS256E, JST CRONOS, Grant No. JPMJCS24N6, JST COI-NEXT, Grant No. JPMJPF2221, and JST ASPIRE, Grant No. JPMJAP2427. We acknowledge support from the European Union’s Horizon Europe research and innovation programme under the project “Quantum Secure Networks Partnership” (QSNP, Grant Agreement No. 101114043), the Danish National Research Foundation through the Center for Macroscopic Quantum States (bigQ, DNRF142), the Innovation Fund Denmark through the CyberQ project (Grant Agreement No. 3200-00035B) and the AccessQKD project (Grant Agreement No. 43560-00004B). M.I. and K.T. were supported by JST SPRING, Grant No. JPMJSP2123.
TIME-RESOLVED INTERFERENCE OF TWO PULSES
In this appendix, we consider the interference of two coherent pulses with different central frequencies and show that even without spectrum overlap, they can effectively interfere by postprocessing compensation if the time resolution of the detector is much shorter than the pulse width.
We consider a simple situation where two coherent states are interfered by a 50:50 beamsplitter and then the output is detected by on-off photon detectors. We start with the ideal case, i.e. the two coherent pulses, and , are exactly in the same mode. Since our DC-TF-QKD protocol uses the single-photon interference, let us see the event that only one of the detectors is clicked. The state after the beamsplitter turns out to be and then the probability of detecting photons in the first mode and not in the second mode is given by
The question asked here is the following. Consider the time and frequency distribution of these coherent pulses. Suppose the carrier frequencies of two pulses are not identical, e.g. due to the laser source fluctuation, but the users can monitor these frequencies. Meanwhile, we can use on-off detectors with very good time resolution, can tag the detection time, and separately monitor the frequency difference of two pulses. Then is it possible to cancel the frequency difference by postprocessing to get the desired interference? To consider this, define a creation operator in spectral mode of as
Then the coherent state distributed in the spectrum of is described as
Similarly, distributed in is . Note that . Since we assume two laser sources have same spectrum with different frequency, holds. By using this, we can treat the single-photon interference in time and frequency domain. The beamsplitter mixes these coherent states as
where
and
To simplify the analysis, assume that the photon detectors have instantaneous time resolution. Then, the POVM element for detecting photons at time can be described by
Then the probability of having one click at time in one detector and no click in the other one is calculated to be
where we use . This exactly corresponds to the time resolved version of Eq. (15). The first term inside the curly bracket is the interference term and the difference from Eq. (15) is that it contains the time slice and the phase rotation by . It implies that, in the DC-TF-QKD protocol, by monitoring and the detection time , one can regard the quantum signal sent from Bob as instead of . This replacement can be done by classical postprocessing. More detailed feasibility analysis is left for future work.
DETAILS OF THE KEY RATE CALCULATION
We describe details about key rate calculation of our protocol. First, we reproduce Eq. (3) in the main text,
In Eq. (23), we perform the optimization over all in the feasible set for estimating key rate. However, we cannot perform this optimization numerically since is infinite dimensional. Thus, we use the photon-number cutoff assumption [55], that is, we assume that Alice’s Hilbert space and Bob’s Hilbert space are spanned by the photon-number basis up to a cutoff photon number , . Under this assumption, a quantum state can be expressed as , where .
The annihilation and creation operators in the photon-number basis up to cutoff photon number are, respectively,
We can write the region operators in this basis, shown in Figure 5. The overlap between a coherent state and a photon-number state is as follows:
Figure 5. Region Operator where .
Therefore, the matrix elements of in the photon-number basis are, for ,
where is the gamma function. In particular, the diagonal elements are
Constraints
We describe the feasible set in the optimization problem in Eq. (23). is expressed as
We choose the following constraints:
where , and we use all combinations of and as the constraints.
Here, we explain why we choose these operators. Let denote the element of a density matrix . In addition, let denote Bob’s quantum state conditioned on Alice’s measurement outcome in the photon-number basis in an honest implementation. Then, for the elements of , the following holds.
That is, the density matrix has the form in the photon-number basis as follows
where are diagonal elements and are nearest off-diagonal elements. From the symmetry of the protocol, Alice’s quantum state conditioned on Bob’s measurement outcome has the same structure. Therefore, we should choose operators which can efficiently constrain these diagonal and nearest off-diagonal elements. We consider the operators and expressed by the photon-number basis. Let and denote elements of and , respectively. Then, the following holds.
This means that these operators have only elements next to diagonal elements. Therefore, by using these operators, we can constrain the nearest off-diagonal elements of the quantum states and . Next, let and denote elements of the operators and in the photon-number basis, respectively. Then,
Thus, these operators are useful for estimating the diagonal elements . Therefore, the set of the operators can constrain a quantum state shared between Alice and Bob in our protocol.
In this paper, we use the cutoff photon number . To justify this choice, we plot analytically calculated key rates of our protocol for different cutoff in Figure 6. While the key rate decreases as the cutoff photon number increases, it saturates when is larger than 3. Note that this convergence check does not remove the photon-number cutoff assumption used in the security analysis.

Figure 6. Key rate versus cutoff photon number . We fix km and . Other parameters are , , and .
Calculation of expectation values
The operators in the set can be expressed by using the annihilation and creation operators in the following way
where are coefficients. For example, the quadrature operators and are expressed as
Also, the operators and are expressed as
By using the commutation relation between the annihilation and the creation operators, we can change the order of these operators in and .
Here, if Alice and Bob perform heterodyne measurements on a density operator , they obtain the Husimi -function
where and are Alice’s and Bob’s measurement outcomes, respectively. Then, we can calculate expectation values of the operator used for estimating a quantum state between Alice and Bob by using the following equation [72]:
where and represent the annihilation and the creation operators of Alice’s and Bob’s systems, respectively. is obtained from by replacing by , and by , and is also obtained in the analogous way.
Since the quantum state between Alice and Bob under the condition that the successful events occur can be expressed as a weighted difference of two Gaussian states, it is possible to analytically calculate the expectation values.
ADDITIONAL NOISE ANALYSES
In Figure 7 and Figure 8, we plot key rates versus the distance for different dark count probability and efficiency of the on-off detectors , respectively. We observe that the dark count probability has a stronger effect on the key rates than the efficiency of the on-off detectors. This is because we use single-photon interference at the central node. When the dark-count probability is comparable to the probability that Alice’s or Bob’s single photon arrives at the central station, we cannot distinguish whether the successful events are caused by the dark count or the single photon. On the other hand, decrease of efficiency of the on-off detectors corresponds to increase of channel losses. Therefore, the key rate is not changed significantly by a decrease in detection efficiency.
Figure 7. Key rate versus distance for different dark count probability . We take and .
Figure 8. Key rate versus distance for different efficiency of the on-off detectors . We take and .
References
References
- [1]C. H. Bennett and G. Brassard, Quantum cryptography: Public key distribution and coin tossing, Theor. Comput. Sci. 56, 7 (2014).DOI
- [2]F. Xu, X. Ma, Q. Zhang, H.-K. Lo, and J.-W. Pan, Secure quantum key distribution with realistic devices, Rev. Mod. Phys. 92, 025002 (2020).DOI
- [3]S. Pirandola, U. L. Andersen, L. Banchi, M. Berta, D. Bunandar, R. Colbeck, D. Englund, T. Gehring, C. Lupo, C. Ottaviani, J. L. Pereira, M. Razavi, J. S. Shaari, M. Tomamichel, V. C. Usenko, G. Vallone, P. Villoresi, and P. Wallden, Advances in quantum cryptography, Adv. Opt. Photonics 12, 1012 (2020).
- [4]A. K. Ekert, Quantum cryptography based on Bell’s theorem, Phys. Rev. Lett. 67, 661 (1991).pubmed.ncbi.nlm.nih.gov/10044956
- [5]F. Grosshans and P. Grangier, Continuous variable quantum cryptography using coherent states, Phys. Rev. Lett. 88, 057902 (2002).pubmed.ncbi.nlm.nih.gov/11863782
- [6]M. Lucamarini, Z. L. Yuan, J. F. Dynes, and A. J. Shields, Overcoming the rate-distance limit of quantum key distribution without quantum repeaters, Nature 557, 400 (2018).DOI
- [7]M. Takeoka, S. Guha, and M. M. Wilde, Fundamental rate-loss tradeoff for optical quantum key distribution, Nat. Commun. 5, 5235 (2014).DOI
- [8]S. Pirandola, R. Laurenza, C. Ottaviani, and L. Banchi, Fundamental limits of repeaterless quantum communications, Nat. Commun. 8, 15043 (2017).DOI
- [9]M. M. Wilde, M. Tomamichel, and M. Berta, Converse bounds for private communication over quantum channels, IEEE Trans. Inf. Theory 63, 1792 (2017).DOI
- [10]S. Pirandola, End-to-end capacities of a quantum communication network, Comm. Phys. 2, 51 (2019).
- [11]X. Ma, P. Zeng, and H. Zhou, Phase-matching quantum key distribution, Phys. Rev. X 8, 031043 (2018).DOI
- [12]K. Tamaki, H.-K. Lo, W. Wang, and M. Lucamarini, Information theoretic security of quantum key distribution overcoming the repeatersless secret key capacity bound, arXiv:1805.05511.arxiv.org/abs/1805.05511
- [13]X.-B. Wang, Z.-W. Yu, and X.-L. Hu, Twin-field quantum key distribution with large misalignment error, Phys. Rev. A 98, 062323 (2018).DOI
- [14]J. Lin and N. Lütkenhaus, Simple security analysis of phase-matching measurement-device-independent quan- tum key distribution, Phys. Rev. A 98, 042332 (2018).DOI
- [15]C. Cui, Z.-Q. Yin, R. Wang, W. Chen, S. Wang, G.-C. Guo, and Z.-F. Han, Twin-field quantum key distribution without phase postselection, Phys. Rev. Appl. 11, 034053 (2019).DOI
- [16]M. Curty, K. Azuma, and H.-K. Lo, Simple security proof of twin-field type quantum key distribution protocol, npj Quantum Inf. 5, 64 (2019).arxiv.org/abs/1807.07667
- [17]K. Maeda, T. Sasaki, and M. Koashi, Repeaterless quantum key distribution with efficient finite-key analysis overcoming the rate-distance limit, Nat. Commun. 10, 3140 (2019).DOI
- [18]F.-Y. Lu, Z.-Q. Yin, C.-H. Cui, G.-J. Fan-Yuan, R. Wang, S. Wang, W. Chen, D.-Y. He, W. Huang, B.-J. Xu, G.-C. Guo, and Z.-F. Han, Improving the performance of twin-field quantum key distribution, Phys. Rev. A 100, 022306 (2019).DOI
- [19]G. Currás-Lorenzo, Á. Navarrete, K. Azuma, G. Kato, M. Curty, and M. Razavi, Tight finite-key security for twin-field quantum key distribution, npj Quantum Inf. 7, 22 (2021).
- [20]M. Minder, M. Pittaluga, G. L. Roberts, M. Lucamarini, J. F. Dynes, Z. L. Yuan, and A. J. Shields, Experimental quantum key distribution beyond the repeaterless secret key capacity, Nat. Photon. 13, 334 (2019).DOI
- [21]S. Wang, D.-Y. He, Z.-Q. Yin, F.-Y. Lu, C.-H. Cui, W. Chen, Z. Zhou, G.-C. Guo, and Z.-F. Han, Beating the fundamental rate-distance limit in a proof-of-principle quantum key distribution system, Phys. Rev. X 9, 021046 (2019).DOI
- [22]Y. Liu, Z.-W. Yu, W. Zhang, J.-Y. Guan, J.-P. Chen, C. Zhang, X.-L. Hu, H. Li, C. Jiang, J. Lin, T.-Y. Chen, L. You, Z. Wang, X.-B. Wang, Q. Zhang, and J.-W. Pan, Experimental twin-field quantum key distribution through sending or not sending, Phys. Rev. Lett. 123, 100505 (2019).DOI
- [23]X. Zhong, J. Hu, M. Curty, L. Qian, and H.-K. Lo, Proof-of-principle experimental demonstration of twin-field type quantum key distribution, Phys. Rev. Lett. 123, 100506 (2019).DOI
- [24]X.-T. Fang, P. Zeng, H. Liu, M. Zou, W. Wu, Y.-L. Tang, Y.-J. Sheng, Y. Xiang, W. Zhang, H. Li, Z. Wang, L. You, M.-J. Li, H. Chen, Y.-A. Chen, Q. Zhang, C.-Z. Peng, X. Ma, T.-Y. Chen, and J.-W. Pan, Implementation of quantum key distribution surpassing the linear rate-transmittance bound, Nat. Photon. 14, 422 (2020).DOI
- [25]J.-P. Chen, C. Zhang, Y. Liu, C. Jiang, W. Zhang, X.-L. Hu, J.-Y. Guan, Z.-W. Yu, H. Xu, J. Lin, M.-J. Li, H. Chen, H. Li, L. You, Z. Wang, X.-B. Wang, Q. Zhang, and J.-W. Pan, Sending-or-not-sending with independent lasers: Secure twin-field quantum key distribution over 509 km, Phys. Rev. Lett. 124, 070501 (2020).DOI
- [26]M. Pittaluga, M. Minder, M. Lucamarini, M. Sanzaro, R. I. Woodward, M.-J. Li, Z. Yuan, and A. J. Shields, 600-km repeater-like quantum communications with dual-band stabilization, Nat. Photon. 15, 530 (2021).DOI
- [27]H. Liu, C. Jiang, H.-T. Zhu, M. Zou, Z.-W. Yu, X.-L. Hu, H. Xu, S. Ma, Z. Han, J.-P. Chen, Y. Dai, S.-B. Tang, W. Zhang, H. Li, L. You, Z. Wang, Y. Hua, H. Hu, H. Zhang, F. Zhou, Q. Zhang, X.-B. Wang, T.-Y. Chen, and J.-W. Pan, Field test of twin-field quantum key distribution through sending-or-not-sending over 428 km, Phys. Rev. Lett. 126, 250502 (2021).DOI
- [28]J.-P. Chen, C. Zhang, Y. Liu, C. Jiang, W.-J. Zhang, Z.-Y. Han, S.-Z. Ma, X.-L. Hu, Y.-H. Li, H. Liu, F. Zhou, H.-F. Jiang, T.-Y. Chen, H. Li, L.-X. You, Z. Wang, X.-B. Wang, Q. Zhang, and J.-W. Pan, Twin-field quantum key distribution over a 511 km optical fibre linking two distant metropolitan areas, Nat. Photon. 15, 570 (2021).DOI
- [29]C. Clivati, A. Meda, S. Donadello, S. Virzì, M. Genovese, F. Levi, A. Mura, M. Pittaluga, Z. Yuan, A. J. Shields, M. Lucamarini, I. P. Degiovanni, and D. Calonico, Coherent phase transfer for real-world twin-field quantum key distribution, Nat. Commun. 13, 157 (2022).DOI
- [30]S. Wang, Z.-Q. Yin, D.-Y. He, W. Chen, R.-Q. Wang, P. Ye, Y. Zhou, G.-J. Fan-Yuan, F.-X. Wang, W. Chen, Y.-G. Zhu, P. V. Morozov, A. V. Divochiy, Z. Zhou, G.-C. Guo, and Z.-F. Han, Twin-field quantum key distribution over 830-km fibre, Nat. Photon. 16, 154 (2022).DOI
- [31]J.-P. Chen, C. Zhang, Y. Liu, C. Jiang, D.-F. Zhao, W.-J. Zhang, F.-X. Chen, H. Li, L.-X. You, Z. Wang, Y. Chen, X.-B. Wang, Q. Zhang, and J.-W. Pan, Quantum key distribution over 658 km fiber with distributed vibration sensing, Phys. Rev. Lett. 128, 180502 (2022).DOI
- [32]L. Zhou, J. Lin, Y. Jing, and Z. Yuan, Twin-field quantum key distribution without optical frequency dissemination, Nat. Commun. 14, 928 (2023).
- [33]Y. Liu, W.-J. Zhang, C. Jiang, J.-P. Chen, C. Zhang, W.-X. Pan, D. Ma, H. Dong, J.-M. Xiong, C.-J. Zhang, H. Li, R.-C. Wang, J. Wu, T.-Y. Chen, L. You, X.-B. Wang, Q. Zhang, and J.-W. Pan, Experimental twin-field quantum key distribution over 1000 km fiber distance, Phys. Rev. Lett. 130, 210801 (2023).DOI
- [34]W. Li, L. Zhang, Y. Lu, Z.-P. Li, C. Jiang, Y. Liu, J. Huang, H. Li, Z. Wang, X.-B. Wang, Q. Zhang, L. You, F. Xu, and J.-W. Pan, Twin-field quantum key distribution without phase locking, Phys. Rev. Lett. 130, 250802 (2023).DOI
- [35]J.-P. Chen, F. Zhou, C. Zhang, C. Jiang, F.-X. Chen, J. Huang, H. Li, L.-X. You, X.-B. Wang, Y. Liu, Q. Zhang, and J.-W. Pan, Twin-field quantum key distribution with local frequency reference, Phys. Rev. Lett. 132, 260802 (2024).DOI
- [36]B. Qi, P. Lougovski, R. Pooser, W. Grice, and M. Bobrek, Generating the local oscillator “locally” in continuous-variable quantum key distribution based on coherent detection, Phys. Rev. X 5, 041009 (2015).DOI
- [37]D. B. S. Soh, C. Brif, P. J. Coles, N. Lütkenhaus, R. M. Camacho, J. Urayama, and M. Sarovar, Self-referenced continuous-variable quantum key distribution protocol, Phys. Rev. X 5, 041010 (2015).
- [38]T. A. Eriksson, T. Hirano, B. J. Puttnam, G. Rademacher, R. S. Luís, M. Fujiwara, R. Namiki, Y. Awaji, M. Takeoka, N. Wada, and M. Sasaki, Wavelength division multiplexing of continuous variable quantum key distribution and 18.3 tbit/s data channels, Commun. Phys. 2, 9 (2019).DOI
- [39]T. A. Eriksson, R. S. Luís, B. J. Puttnam, G. Rademacher, M. Fujiwara, Y. Awaji, H. Furukawa, N. Wada, M. Takeoka, and M. Sasaki, Wavelength division multiplexing of 194 continuous variable quantum key distribution channels, J. Lightwave Technol. 38, 2214 (2020).DOI
- [40]T. Matsuura, K. Maeda, T. Sasaki, and M. Koashi, Finite-size security of continuous-variable quantum key distribution with digital signal processing, Nat. Commun. 12, 252 (2021).
- [41]N. Jain, H.-M. Chin, H. Mani, C. Lupo, D. S. Nikolic, A. Kordts, S. Pirandola, T. B. Pedersen, M. Kolb, B. Ömer, C. Pacher, T. Gehring, and U. L. Andersen, Machine learning aided carrier recovery in continuous-variable quantum key distribution, Nat. Commun. 13, 4740 (2022).DOI
- [42]Z. Chen, X. Wang, S. Yu, Z. Li, and H. Guo, Continuous-mode quantum key distribution with digital signal processing, npj Quantum Inf. 9, 28 (2023).DOI
- [43]A. A. E. Hajomer, I. Derkach, H.-M. Chin, U. L. Andersen, and T. Gehring, Long-distance continuous-variable quantum key distribution over 100-km fiber with local local oscillator, Sci. Adv. 10, eadi9474 (2024).DOI
- [44]A. A. E. Hajomer, C. Bruynsteen, I. Derkach, N. Jain, A. Bomhals, S. Bastiaans, U. L. Andersen, X. Yin, and T. Gehring, Continuous-variable quantum key distribution at 10 GBaud using an integrated photonic-electronic receiver, Optica 11, 1197 (2024).
- [45]N. Jain, H.-M. Chin, A. A. E. Hajomer, D. Null, H. Larfort, N. Nysom, E. Bidstrup, U. L. Andersen, and T. Gehring, Future proofing network encryption technology with continuous-variable quantum key distribution, Opt. Express 32, 43607 (2024).
- [46]A. A. E. Hajomer, U. L. Andersen, and T. Gehring, High-rate continuous-variable measurement device-independent quantum key distribution with finite-size security, Quantum Sci. Technol. 10, 025032 (2025).arxiv.org/abs/2303.01611
- [47]A. A. E. Hajomer, F. Kanitschar, N. Jain, M. Hentschel, R. Zhang, N. Lütkenhaus, U. L. Andersen, C. Pacher, and T. Gehring, Experimental composable key distribution using discrete-modulated continuous variable quantum cryptography, Light Sci. Appl. 14, 255 (2025).DOI
- [48]A. A. E. Hajomer, I. Derkach, V. C. Usenko, U. L. Andersen, and T. Gehring, Coexistence of Continuous-Variable Quantum Key Distribution and Classical Data over 120 km Fiber, Phys. Rev. Lett. 135, 170804 (2025).DOI
- [49]A. A. E. Hajomer, A. Bomhals, C. Bruynsteen, A. Sidhique, O. Kovalenko, I. Derkach, V. C. Usenko, U. L. Andersen, X. Yin, and T. Gehring, Chip-based 16 GBaud continuous-variable quantum key distribution, Optica 13, 1035 (2026).arxiv.org/abs/2504.09308
- [50]K. Kikuchi, Fundamentals of coherent optical fiber communications, J. Lightwave Technol. 34, 157 (2016).DOI
- [51]M. Navascués, F. Grosshans, and A. Acín, Optimality of Gaussian attacks in continuous-variable quantum cryptography, Phys. Rev. Lett. 97, 190502 (2006).DOI
- [52]R. García-Patrón and N. J. Cerf, Unconditional optimality of Gaussian attacks against continuous-variable quantum key distribution, Phys. Rev. Lett. 97, 190503 (2006).
- [53]P. J. Coles, E. M. Metodiev, and N. Lütkenhaus, Numerical approach for unstructured quantum key distribution, Nat. Commun. 7, 11712 (2016).DOI
- [54]A. Winick, N. Lütkenhaus, and P. J. Coles, Reliable numerical key rates for quantum key distribution, Quantum 2, 77 (2018).DOI
- [55]J. Lin, T. Upadhyaya, and N. Lütkenhaus, Asymptotic security analysis of discrete-modulated continuous-variable quantum key distribution, Phys. Rev. X 9, 041064 (2019).DOI
- [56]Y.-M. Xie, Y.-S. Lu, C.-X. Weng, X.-Y. Cao, Z.-Y. Jia, Y. Bao, Y. Wang, Y. Fu, H.-L. Yin, and Z.-B. Chen, Breaking the rate-loss bound of quantum key distribution with asynchronous two-photon interference, PRX Quantum 3, 020315 (2022).DOI
- [57]L. Zhou, J. Lin, Y.-M. Xie, Y.-S. Lu, Y. Jing, H.-L. Ying, and Z. Yuan, Experimental quantum communication overcomes the rate-loss limit without global phase tracking, Phys. Rev. Lett. 130, 250801 (2023).DOI
- [58]P. Zeng, H. Zhou, W. Wu, and X. Ma, Mode-pairing quantum key distribution, Nat. Commun. 13, 3903 (2022).DOI
- [59]D. Huang, D. Lin, C. Wang, W. Liu, S. Fang, J. Peng, P. Huang, and G. Zeng, Continuous-variable quantum key distribution with 1 Mbps secure key rate, Opt. Express 23, 17511 (2015).DOI
- [60]S. Ren, S. Yang, A. Wonfor, I. White, and R. Penty, Demonstration of high-speed and low-complexity continuous variable quantum key distribution system with local local oscillator, Sci. Rep. 11, 9454 (2021).DOI
- [61]T. Wang, Z. Zuo, L. Li, P. Huang, Y. Guo, and G. Zeng, Continuous-variable quantum key distribution without synchronized clocks, Phys. Rev. Applied 18, 014064 (2022).DOI
- [62]A. Weerasinghe, M. Alhussein, A. Alderton, A. Wonfor, and R. Penty, Practical, high-speed Gaussian coher- ent state continuous variable quantum key distribution with real-time parameter monitoring, optimised slicing, and post-processed key distillation, Sci. Rep. 13, 21543 (2023).
- [63]B. Korzh, Q.-Y. Zhao, J. P. Allmaras, S. Frasca, T. M. Autry, E. A. Bersin, A. D. Beyer, R. M. Briggs, B. Bumble, M. Colangelo, G. M. Crouch, A. E. Dane, T. Gerrits, A. E. Lita, F. Marsili, G. Moody, C. Peña, E. Ramirez, J. D. Rezac, N. Sinclair, M. J. Stevens, A. E. Velasco, V. B. Verma, E. E. Wollman, S. Xie, D. Zhu, P. D. Hale, M. Spiropulu, K. L. Silverman, R. P. Mirin, S. W. Nam, A. G. Kozorezov, M. D. Shaw, and K. K. Berggren, Demonstration of sub-3 ps temporal resolution with a superconducting nanowire single-photon detector, Nat. Photon. 14, 250 (2020).DOI
- [64]J. Chang, J. W. N. Los, J. O. Tenorio-Pearl, N. Noordzij, R. Gourgues, A. Guardiani, J. R. Zichi, S. F. Pereira, H. P. Urbach, V. Zwiller, S. N. Dorenbos, and I. E. Zadeh, Detecting telecom single photons with (99.5^{+0.5}_{-0.7})% system detection efficiency and high time resolution, APL Photon. 6, 036114 (2021).
- [65]T.-M. Zhao, H. Zhang, J. Yang, Z.-R. Sang, X. Jiang, X.-H. Bao, and J.-W. Pan, Entangling different-color photons via time-resolved measurement and active feed-forward, Phys. Rev. Lett. 112, 103602 (2014).
- [66]I. Devetak and A. Winter, Distillation of secret key and entanglement from quantum states, Proc. R. Soc. A 461, 207 (2005).
- [67]P. J. Coles, Unification of different views of decoherence and discord, Phys. Rev. A 85, 042103 (2012).DOI
- [68]J. Lodewyck, M. Bloch, R. Garcia-Patrón, S. Fossier, E. Karpov, E. Diamanti, T. Debuisschert, N. J. Cerf, R. Tualle-Brouri, S. W. McLaughlin, and P. Grangier, Quantum key distribution over 25 km with an all-fiber continuous-variable system, Phys. Rev. A 76, 042305 (2007).DOI
- [69]A. Leverrier, Security of continuous-variable quantum key distribution via a Gaussian de Finetti reduction, Phys. Rev. Lett. 118, 200501 (2017).
- [70]F. Kanitschar, I. George, J. Lin, T. Upadhyaya, and N. Lütkenhaus, Finite-size security for discrete-modulated continuous-variable quantum key distribution protocols, PRX Quantum 4, 040306 (2023).DOI
- [71]M. Walschaers, Non-Gaussian quantum states and where to find them, PRX Quantum 2, 030204 (2021).DOI
- [72]M. O. Scully and M. S. Zubairy, Quantum Optics (Cambridge University Press, 1997).




