Abstract
Free-space ground-to-ground links will be an integral part of future quantum communication networks. The implementation of free-space and fiber links in daylight inter-modal configurations is however still hard to achieve, due to the impact of atmospheric turbulence, which strongly decreases the coupling efficiency into the fiber. In this work, we present a comprehensive model of the performance of a free-space ground-to-ground quantum key distribution (QKD) system based on the efficient-BB84 protocol with active decoy states. Our model takes into account the atmospheric channel contribution, the transmitter and receiver telescope design constraints, the parameters of the quantum source and detectors, and the finite-key analysis to produce a set of requirements and optimal design choices for a QKD system operating under specific channel conditions. The channel attenuation is calculated considering all effects deriving from the atmospheric propagation (absorption, beam broadening, beam wandering, scintillation, and wavefront distortions), as well as the effect of fiber-coupling in the presence of a partial adaptive optics correction with finite control bandwidth. We find that the channel fluctuation statistics must be considered to correctly estimate the effect of the saturation rate of the single-photon detectors, which may otherwise lead to an overestimation of the secret key rate. We further present strategies to minimize the impact of diffuse atmospheric background in daylight operation by means of spectral and temporal filtering.
Export citation and abstract BibTeX RIS
1. Introduction
Quantum key distribution (QKD) [1–4] has the potential to allow secure communication between any two points on Earth. In a future continental-scale quantum network (or quantum internet) [5–9] satellite, fiber, and free-space links will be required to operate jointly, in an inter-modal configuration. While satellite-to-ground QKD has been demonstrated [10–14] and the development of fiber-based QKD is technologically mature [15–22], the inter-modal operation of free-space and fiber links has only recently started to be investigated [23–25]. Free-space ground-to-ground links, although more lossy than a fiber equivalent over the same distance, require a lighter infrastructure investment, may exploit mobile stations and offer connectivity in remote locations.
An inter-modal QKD network must guarantee the compatibility of the free-space links with the fiber-based infrastructure, which is based on the achievement of stable coupling of the free-space signal into a single-mode fiber (SMF) and on the use of a shared signal wavelength, typically in the telecom band. Coupling the received signal into a SMF brings in of itself several advantages, since the narrow field-of-view (FOV) of the fiber limits the amount of background solar radiance that can reach the detector and the small mode-field-diameter (MFD) of a standard SMF (typically 10 μm) allows the use of detectors with a small active area, which are typically faster than larger detectors. This opens the way to daylight free-space QKD, thus enabling for continuous-time operation [25].
However, the SMF coupling efficiency is strongly affected by the wavefront perturbations introduced by atmospheric turbulence, requiring the introduction of mitigation techniques such as adaptive optics (AOs) [26].
The performance of fiber-based QKD systems was studied in detail by reference [27], where the authors calculated the secret key rate (SKR), optimal decoy-state parameters, and key block length for the finite-key analysis, considering the channel loss as a fixed parameter. This approach is not appropriate for the case of free-space channels, since the statistics of atmospheric turbulence induces a random fading of the transmitted signal.
The statistics of the free-space channel transmission was derived in [28, 29] to calculate the SKR of decoy-state QKD including the effect of collection losses due to beam-wander and scintillation. This treatment is however limited to the case of a QKD receiver with free-space detectors, and thus excludes SMF-coupled receivers. A similar approach was recently adopted in [30, 31], for the specific case of continuous-variable (CV) QKD.
The effect of wavefront perturbations and atmospheric scintillation was calculated in [32], where the SMF-coupling probability distribution was derived to extract the fading statistics of a satellite-to-ground link, considering the effect of a partial AO correction of the perturbed wavefront received. This approach was found by [25] to be applicable also to ground-to-ground links.
In this article, we develop a comprehensive model of the performance of a free-space ground-to-ground QKD system. Differently from reference [30], we focus on the commonly used efficient-BB84 protocol with active decoy states, in the one-decoy variant of reference [27], in which Alice randomly chooses between two intensity levels to counter the photon-number splitting attack. We generalize the approach of [29, 32] to include both the collection losses at the receiver aperture, and the losses due to SMF coupling. Moreover the model of [32] is further extended to include the effect of a finite control bandwidth of the AO system.
The model considers the effect of atmospheric absorption, receiver collection efficiency as a function of beam broadening, beam wandering and atmospheric scintillation, and SMF-coupling in the presence of atmospheric turbulence with partial AO correction of the wavefront deformations and finite AO control bandwidth to calculate the overall channel loss. The finite efficiency and saturation of the single-photon detectors are also included.
The expected error rate is calculated considering the intrinsic coding error caused by imperfect preparation and measurement of quantum states, the noise introduced by the detectors (dark counts and afterpulses), and the amount of diffuse atmospheric background coupled into the receiver in daylight operation.
The present model gives as output the obtainable SKR, which takes into account the atmospheric channel contribution, the transmitter and receiver design constraints, the parameters of the quantum source and detectors, and the finite-key analysis to produce a set of requirements and optimal design choices for a QKD system operating under specific free-space channel conditions. The workflow of our QKD model is sketched in figure 1.
In section 2, we study the several contributions to the channel efficiency of ground-to-ground links, which reduce the signal detection rate. In section 3, we consider the effects which introduce errors in the exchange of qubits and reduce the SKR. Finally, in section 4, we combine the channel analysis with the decoy-state and finite-key analysis to estimate the final SKR. To ease the readability of the paper we report in appendix
2. Channel efficiency and detection rate
In the case of an horizontal free-space link where the optical signal is coupled into a SMF at the receiver, the overall channel efficiency ηCH is given by the product of three terms and can be written as:
where ηA denotes the atmospheric channel absorption, the receiver collection efficiency, and ηSMF the SMF coupling efficiency.
As a first order analysis, we will consider the effect that the atmospheric turbulence has on the average value of the different terms composing the channel efficiency. However, when the detection rate at the receiver approaches the saturation limit of the single-photon detectors, the statistics of the collection efficiency and single-mode coupling efficiency can no longer be ignored, and the whole probability distribution pCH(ηCH) has to be considered for the expected SKR to be estimated correctly (see section 2.5).
In our model, the probability distributions are numerically calculated and normalized as weight functions over a discretized array {η1, ..., ηN }, so that:
where the probability p(ηi ) is the probability that the efficiency lies within the interval [ηi−1, ηi ] with δηi = ηi − ηi−1 the spacing. Starting from the analytic probability density function (pdf), for sufficiently fine binning we have
2.1. Atmospheric absorption
The first term entering the channel efficiency in equation (1) takes into account the attenuation due to atmospheric absorption. The channel absorption efficiency ηA for a link distance z depends on the absorption coefficient A(λ) for a specific wavelength λ and can be modeled as
where the absorption coefficient is assumed constant for a horizontal link. An established tool for calculating the spectral properties of the atmosphere is the LOWTRAN software package [33], which can be used to predict atmospheric absorption and scattering over a wide wavelength range, on horizontal or slanted paths, taking into account both geographical and seasonal atmospheric variations. In figure 2, we show the atmospheric absorption coefficient computed by LOWTRAN as a function of wavelength for a horizontal link, considering a sub-arctic winter atmosphere. Clearly, signal wavelengths for which the atmospheric absorption coefficient is high should be avoided in free-space links.
Download figure:
Standard image High-resolution image2.2. Collection efficiency
The second term entering the overall channel efficiency in equation (1) is the collection efficiency and takes into account the finite size of the receiver aperture DRx that is used to collect the incoming beam. We now detail the impact of turbulence-induced beam broadening, and beam wander and scintillation on the collection efficiency .
2.2.1. Turbulence-induced beam broadening
In vacuum, the beam size W(z) of a collimated Gaussian beam of waist W0 and wavelength λ propagating for a distance z is given by the formula for diffraction-limited propagation:
According to the Kolmogorov's theory of turbulence [34], when the propagation happens through the turbulent air, the index of refraction is treated as a fluctuating random field around a mean value which induces a perturbation of the wavefront, resulting in an overall loss of coherence of the optical wave. The atmospheric perturbation is captured by the so-called power spectral density Φn(κ), which is the Fourier transform of the refractive-index covariance function in terms of the spatial frequency κ. In our model, we use the power spectral density Φn(κ) for refractive-index fluctuations given by the well-known Kolmogorov spectrum of atmospheric turbulence
which is widely used in theoretical calculations. The strength of the turbulence is parametrized by the refractive-index structure constant , which may be considered constant along a horizontal link.
The effect of turbulence-induced coherence loss on Gaussian beam propagation has been studied by [35, 36], who found that the following formula holds:
where
is the spherical-wave atmospheric spatial coherence radius (with k = 2π/λ the wave-number). A parameter typically used to quantify the strength of the turbulence on a channel is the atmospheric coherence width r0, also known as the Fried parameter [34]. It describes the maximum size of a telescope aperture whose resolution is comparable to the diffraction limited one under that specific turbulence, and is related to ρ0 by:
While for the diffraction-limited case a larger beam waist W0 implies a smaller intrinsic divergence θ0 = λ/πW0, in the turbulence-affected case the larger the ratio W0/ρ0, the stronger is the loss of coherence. Indeed, for z ≫ 1 we have that the two competing effects, diffraction-broadening and turbulence broadening, compensate each other and the beam size tends to a value that is independent of the initial waist W0:
as shown in figure 3 for different values of W0 and .
Download figure:
Standard image High-resolution imageAssuming for simplicity that the pointing error is negligible, the average contribution to the collection efficiency caused by diffraction and beam broadening for a receiver of finite aperture diameter DRx is given by
which is the integral of a Gaussian distribution of standard deviation W(z)/2 over a concentric circular area of diameter DRx. Although the easiest way to optimize this term is to increase DRx, this strategy decreases the fiber coupling efficiency, as we will study in section 2.3.4, and it might not be a viable solution since it requires larger and more expensive optics.
2.2.2. Beam wander and scintillation
A consequence of the asymptotic trend of W(z) shown in figure 3 is that, for relatively long links, the average collection efficiency becomes independent of the choice of the waist at the transmitter. However, this does not necessarily imply that a smaller transmitter waist should be preferred when, faced with equivalent performance at the receiver, a smaller telescope is less complex and less costly.
This is because the beam-size W(z) appearing in equation (11) is the so-called long-term spot size, which represents the size of the beam averaged over a timescale much longer than the turbulence dynamics. The instantaneous short-term (ST) beam size WST(z) at a distance z is given by:
where is the variance of beam-wander fluctuations at the receiver aperture plane. Beam wandering is caused by the larger-sized turbulence eddies, and results in a shift of the ST spot on the receiver aperture. For a collimated beam transmitted along a horizontal channel one has [34]:
We can then see that, since the beam wandering depends on , a smaller transmitter waist always leads to a higher beam wandering variance at the receiver. The effect of beam wandering may be removed by automatically tracking and correcting the tip/tilt fluctuation at both the transmitter and receiver using of two counter-propagating beacons [37, 38]. This link configuration assumes a symmetry between the front- and back-propagating channels, and its treatment for QKD links is covered in [29]. In this work, we will restrict ourselves to the case of a single pointing feedback at the receiver, which does not enable the removal of beam wandering.
In addition to beam wander, one must also consider the effect of atmospheric scintillation, which introduces random fluctuations in the beam irradiance profile. The pdf of was derived analytically in [28, 29] exploiting the law of total probability and separating the contributions from turbulence-induced beam wandering and atmospheric scintillation. The distribution varies depending on the strength of turbulence, which is parametrized by the Rytov variance , which is defined as:
For weak turbulence resembles a log-negative Weibull distribution, while for stronger turbulence one finds a truncated log-normal distribution. The exact form and derivation of can be found in [29] and requires the knowledge of the following quantities: the average collection efficiency of (11), the collection efficiency calculated using the ST waist WST of (12), the beam-wander variance , and the mean-squared efficiency , that is:
where is the aperture-averaged scintillation index (flux variance) [34]:
with and . Some examples of probability distributions calculated in this way are provided in figure 4.
Download figure:
Standard image High-resolution image2.3. Fiber coupling efficiency
The third term entering the channel efficiency in equation (1) is due to the coupling of the free-space beam into a SMF at the receiver. The fiber coupling efficiency is given by the normalized overlap integral between the fiber mode, and the incident optical field ,
where χ is the log-amplitude perturbation term, and Ψ is the wavefront phase term. The different origins of χ and Ψ perturbations imply the statistical independence of scintillation and phase effects [32, 39], and the coupling efficiency ηSMF can be factorized into three terms:
where η0 is the optical efficiency of the receiver telescope, ηAO is the coupling efficiency due to wavefront perturbations that may be partially corrected by AO, and ηS is the coupling efficiency due to the spatial structure of atmospheric scintillation. We now discuss the three terms separately.
2.3.1. Optical coupling efficiency
The optical coupling efficiency η0 of an optical system measures the matching between an unperturbed received beam and the MFD of the SMF. η0 is determined by the design optics of the receiving telescope, and particularly by the ratio α = DObs/DRx between the diameters of the central obscuration and of the telescope aperture [40]. The ideal coupling efficiency can be parametrized by
where β is given by
with DRx the receiver diameter, f the effective focal length of the optical system, and MFD the MFD of the SMF. Given a particular α, the value of β can be optimized to achieve the optimal coupling efficiency . Knowing the value of βopt allows to choose the optimal design value of f, since typically the working wavelength λ, fiber MFD and receiver diameter are constrained. Figure 5 shows and βopt as a function of the obscuration ratio α. For α = 0, we have βopt = 1.12, that allows to achieve a maximum optical coupling efficiency of 81.5% ≈ −0.89 dB.
Download figure:
Standard image High-resolution image2.3.2. Effect of wavefront perturbations and AOs correction
As was first derived by [41, 42], the instantaneous wavefront aberration introduced by the turbulent channel at a point on the receiver aperture can be decomposed in a superposition of Zernike polynomials defined over the normalised pupil coordinates (r, φ), with . In the decomposition of , each polynomial term of radial degree n and azimuthal degree m is weighted by a time-dependent coefficient , yielding
The Zernike coefficient variances represent the statistical strength of a particular aberration order, and depend on the ratio of the receiver aperture DRx to the Fried parameter r0 with a modal term scaling with the radial order n [42, 43]:
An expression of the instantaneous coupling efficiency in the presence of wavefront perturbations was derived by [32, 44] directly in terms of the Zernike coefficients :
From (23), since the coefficients are independent, Gaussian-distributed random variables with zero mean and variance given by (22)—i.e. —we derive the average coupling efficiency:
Equation (24) makes the calculation of the average SMF coupling efficiency in the presence of a partial AO compensation of turbulence up to an order nmax straightforward. Assuming an ideal AO system with infinite control bandwidth, it is sufficient to completely suppress the coefficients in the productory corresponding to radial orders n ⩽ nmax. Figure 6 shows the average SMF coupling efficiency as a function of the ratio DRx/r0 for increasing order of AO correction, assuming infinite control bandwidth.
Download figure:
Standard image High-resolution imageFrom (22), we have that the strength of the aberration orders decreases for higher orders. It is interesting then to view the results of figure 6 in a different form: figure 7 shows the required correction order for different target AO efficiencies, as a function of turbulence strength—parametrized by the ratio DRx/r0. If the receiver diameter is fixed and r0 is known from a previous characterization of the channel value, then using equation (24) it is possible to derive the number of aberration orders that need to be corrected in order to achieve a given target coupling efficiency.
Download figure:
Standard image High-resolution imageSo far, we have assumed that the AO system is able to perfectly correct turbulence in real time. However, this is not the case, as various imperfections affect the efficiency, such as wavefront reconstruction errors and control bandwidth limitations. In this work, we focus on the latter term. As discussed in reference [45], the effect of control bandwidth limitations can be taken into account by introducing a mode attenuation factor for the nth order aberration coefficients:
where represents the power spectral density of the temporal spectrum of the nth order aberrations, is the frequency of the AO loop, and represents the transfer function between the residual phase and the turbulent wavefront fluctuations, and depends on the AO system's open-loop transfer function :
In this work, we consider a typical AO system with a pure-integrator control based on wavefront sensing with a Shack–Hartmann wavefront sensor, and correction with a deformable piezo-electric mirror. The open-loop transfer function is then given by
where Ki is the gain of the integrator, τ is the overall latency of the control-actuator stage, and T is the wavefront sensor integration time.
The temporal power spectra of the different aberration orders have been studied in reference [46], where the authors find that the power spectral density scales polynomially with a cut-off frequency depending on the radial order n, average wind velocity and receiver diameter DRx:
with
The SMF efficiency of equation (24) is thus modified in the case of finite control bandwidth as:
with nmax the maximum aberration order corrected.
To conclude, in figure 8 we show as a function of the corrected aberration order and wavefront sensor integration time, in a scenario with DRx/r0 = 8.5 and an average wind velocity corresponding to a light breeze, assuming in this instance a negligible latency of the control loop. We can see that to reach the desired coupling efficiency, represented by dashed blue lines in the figure, both the number of corrected orders and the bandwidth of the sensor must be increased, otherwise, if either one of the two is kept fixed, the SMF coupling has an upper bound that cannot be exceeded.
Download figure:
Standard image High-resolution image2.3.3. Effect of atmospheric scintillation
In addition to phase perturbations, we also take into account the irradiance fluctuations introduced by atmospheric scintillation, which result in a random apodization of the pupil transmittance function, which in turn affects the maximum SMF coupling efficiency. A rigorous calculation of the scintillation contribution ηS to the SMF efficiency, which considers the optical system modulation transfer function and log-amplitude spatial covariance function Cχ (r) can be found in reference [32] and is based on the result of references [39, 41]. Nonetheless, a good approximation for the average value of ηS is only dependent on the on-axis pupil-plane scintillation index , which is given by equation (16) in the limit of an infinitesimal aperture i.e., d → 0:
where we use the fact that the log-amplitude variance is related to the scintillation index through .
Figure 9 shows a plot of and as a function of the Rytov variance of equation (14), highlighting the behavior of , which increases for the weak fluctuation regime, reaches a maximum value in the focusing regime, and then tends to in the strong fluctuation regime. The coupling efficiency reduction due to scintillation is contained to ∼1 dB, consistent with the fact that in the wave structure function the phase contribution dominates over the log-amplitude one [39]. This term should anyway be taken into account, as it cannot be corrected by the AO system.
Download figure:
Standard image High-resolution image2.3.4. Optimum receiver diameter
We have seen that the average collection efficiency of equation (11) increases as the receiver diameter increases. Conversely, for a fixed AO correction order n, the SMF coupling efficiency of (24) decreases with increasing receiver diameter. This leads to a trade-off between the collection efficiency and the fiber-coupling efficiency (as represented in figure 10). Once the wavelength, link distance, transmitter waist, and turbulence strength are fixed, if the maximum order of the AO correction is also fixed—for example by the available number of deformable mirror actuators, or the size of the wavefront sensor—then it is possible to find the optimum receiver diameter Dopt that maximizes the overall channel efficiency ηCH.
Download figure:
Standard image High-resolution imageFigure 11 shows the optimum receiver diameter as a function of link distance, for different orders of aberration correction and a moderate turbulence strength of . For short link distances (shorter than the transmitter Rayleigh distance ) the beam size does not diverge much, and the fiber-coupling term dominates, causing the optimal diameter to decrease with increasing distance. For link distances z ∼ z0, the beam size starts increasing, the collection efficiency term dominates, leading to an increasing optimum beam diameter. For longer propagation distances (z ≫ z0), the decrease in spatial coherence of the beam is more severe and the turbulence term dominates again, leading to a smaller optimum diameter.
Download figure:
Standard image High-resolution image2.3.5. Probability distribution of coupling efficiency
As anticipated, when the receiver rate approaches the saturation limit of the single-photon detectors, the full statistics of channel efficiency can no longer be ignored, and the whole probability distribution pCH has to be considered for the expected SKR to be estimated correctly, as we will present in section 2.5.
The derivation of the probability distribution of the SMF coupling efficiency with partial AOs correction can be found in reference [32], which, however, does not include the effect of finite control bandwidth. Since the irradiance fluctuation statistical contribution is already taken into account for the collection efficiency, we restrict the calculation of the SMF coupling distribution to phase distortions only, and present some examples of probability distributions calculated as a function of the maximum corrected aberration order (figure 12), and varying the bandwidth of the AO control loop (figure 13).
Download figure:
Standard image High-resolution imageDownload figure:
Standard image High-resolution imageGiven a set of Zernike coefficients {bj } (where j indicates the OSA/ANSI index j = [n(n + 1) + m]/2) with variances , which may be corrected by a compensation factor —where for perfect compensation, and for uncorrected coefficients [32]—we define the quantity ξ(t) as the instantaneous sum of the squared Zernike coefficients:
The probability distribution of ξ, including the effect of finite AO control bandwidth is then, using the result of reference [47]:
The probability distribution of ηSMF is then
where ηmax = η0 ⋅ ηS is the maximum normalized coupled flux, given by the product of the optical coupling efficiency of the system η0, and the spatial scintillation term ηS.
2.4. Channel probability distribution
Based on the results of reference [32], which calculated the channel loss term due to SMF coupling, and reference [29], which calculated the channel loss due to the finite receiver aperture in the presence of beam broadening, beam wandering and scintillation, we calculate the overall channel transmittance probability distribution considering both contributions (pupil-plane and focal-plane losses) and exploiting the law of total probability to write pCH(ηCH) as:
where is the probability of obtaining a normalized flux ηCH in the SMF fiber, given a maximum input normalized flux , through equation (34).
In figure 14 we show some examples of channel probability distributions for eight case studies, with average link losses between −7 dB and −48 dB and input parameters summarized in table 1. In all case studies, we assume an unobstructed receiver aperture, and maximum optical efficiency of η0 = 81.5%. Cases 1 and 3 correspond to a scenario with a short urban link with strong turbulence, small aperture Tx/Rx telescopes, and mere tip/tilt correction. Cases 4, and 5 correspond to a scenario with moderate turbulence and longer link distance and highlight the effect of a smaller/larger receiver aperture, which leads to a trade-off between large collection efficiency and minimum aberration order corrected. Cases 2 and 6 show the effect of AO on longer, moderately turbulent links with large aperture receivers. Cases 7 and 8 show examples of highly lossy channels.
Download figure:
Standard image High-resolution imageTable 1. Input parameters for the simulation of the eight case studies.
Case 1 | Case 2 | Case 3 | Case 4 | Case 5 | Case 6 | Case 7 | Case 8 | ||
---|---|---|---|---|---|---|---|---|---|
⟨ηCH⟩ | −7 | −15 | −17 | −23 | −25 | −38 | −43 | −48 | (dB) |
10–13 | 10–14 | 10–13 | 10–14 | 10–14 | 10–14 | 10–14 | 10–14 | (m−2/3) | |
W0 | 25 | 60 | 25 | 60 | 60 | 60 | 60 | 25 | (mm) |
DRx | 50.8 | 200 | 50.8 | 200 | 50 | 200 | 400 | 200 | (mm) |
z | 1 | 10 | 2 | 10 | 10 | 20 | 20 | 30 | (km) |
nmax | 1 | 4 | 1 | 1 | 1 | 1 | 2 | 1 |
2.5. Effect of channel fluctuations on QKD performance
The channel efficiency ηCH that we have studied in the previous sections has a significant impact on the key performance indicators of QKD, such as the SKR and error rate. Clearly, a low efficiency decreases the signal rate and increases the relative weight of errors, which both have a negative effect on the SKR. These parameters are estimated using large samples of data acquired during a long experiment, averaging over the distributions pCH. This might suggest that the fluctuations of ηCH can be neglected and only its mean value is relevant. Yet, this would be equivalent to calculating the expected value of a function by applying it to the expected value of its parameters. Since the functions that model the generation of a key are not all affine, i.e., they are not all compositions of a translation and a linear map, neglecting the fluctuations and using only the mean values is in general incorrect.
A minor non-affine effect is due to the threshold behavior of single-photon detectors, which produce an electrical signal whenever they observe one or more photons. Consequently, the output rate is:
where R is the repetition rate of the source, μ is the mean number of photons per pulse, ηdet represents all the fixed efficiency terms at the receiver not accounted in ηCH, such as the detector quantum efficiency or insertion losses of fiber components, Rbkg is the background photon rate, Rdark is the dark rate, and pap is the probability that an afterpulse is triggered by another detection (higher-order afterpulses are neglected). This is not an affine function of ηCH, which enters the formula in the exponential term characteristic of the Poisson distribution of the number of photons in laser pulses. Yet, in typical conditions, in which μ ⋅ ηdet ⋅ ηCH ≪ 1, it is legitimate to approximate this function to an affine one because .
However, there is another, much more relevant, non-affine effect, caused by the saturation of the detectors. These devices are blinded just after an event and might be further kept off to combat the phenomenon of afterpulses, which increases noise. This so-called dead time Td implies that there is maximum rate of output signals Rsat = 1/Td that the detectors can produce.
This means that the actual (output) detection rate is not R0 but [48]
Because this is not an affine function of ηCH, the fluctuations of the latter cannot be neglected. We note that, although equation (37) is derived for a continuous source, it is approximately valid also for a pulsed one if the repetition rate is much greater than Rsat.
To quantify the importance of these fluctuations, we estimate the detection rate of a QKD system for the eight cases of table 1, first considering the entire distributions of ηCH shown in figure 14 and then only their mean values. In figure 15 we show the overestimation factor (converted to dB to visualize it better) caused by neglecting fluctuations. This can reach a value of almost 9 dB for the distribution of case 8. The effect is larger when ⟨R0⟩ ≈ Rsat and vanishes for ⟨R0⟩ ≫ Rsat or ⟨R0⟩ ≪ Rsat, when equation (37) is well approximated by its linearization. The distributions for which this error is greater are those which have stronger tails (i.e., a high kurtosis). Indeed, neglecting fluctuations means neglecting the suppression of the tails caused by the saturation of the detectors: the greater the tails are, the graver is the error caused by neglecting them.
Download figure:
Standard image High-resolution imageIn figure 16, we consider an arbitrary QKD scenario with a 1 GHz source, a dead time Td = 10 μs, and 15% detection efficiency, which are realistic features for a single-photon avalanche diode (SPAD) [49]. We show a simulation of the detection rate (as a function of the channel efficiency ηCH) which neglects fluctuations, and compare it with the more correct values which consider this effect. We can see a clear separation between the two methods of estimation, which grows larger when ⟨R0⟩ ≈ Rsat and for distributions of greater kurtosis. This shows that performance predictions that consider only the mean value of the channel efficiency can be severely inaccurate.
Download figure:
Standard image High-resolution image3. Contributions to the error rate
Mismatches between Alice and Bob's raw keys influence the performance of QKD in two ways. First, the more the errors, the more bits must be published to correct them. Second, they indicate the amount information leaked to an attacker. In the security scenario in which QKD operates, all errors are attributed to attacks and reduce the length of the final secret key. Therefore, when simulating a QKD system, several physical sources of error must be considered.
One is intrinsic to the signal: inaccurate quantum state preparation or measurement might cause a mismatch between Alice's encoded bit and Bob's decoded one, even if the carrier photon arrives at the detector. We quantify this with the coding error, which we define as the conditional probability of a mismatch given that a signal photon is detected. In principle, the channel can also increase it if it can change the state of the photons, but this does not happen in typical stationary free-space systems with polarization encoding, because the medium in which light travels is not birefringent. The only way to reduce the coding error is to build better quantum state encoders and decoders, and better systems to align them to each other [50, 51].
Then, there is random noise. A portion of it is caused by single-photon detectors, in the form of dark counts and afterpulses. The former are random events that happen even in the total absence of light, whereas the latter are additional spurious signals caused by true ones and are typical of avalanche diodes. Another portion is introduced by the channel background light, especially in the free-space case that we are studying.
In section 3.1 we quantify this background light and in section 3.2 we study a way to counter noise with temporal filtering.
3.1. Diffuse atmospheric background
A crucial requirement for the realization of daylight free-space QKD is the successful filtering of the background radiation. Figure 17 shows the spectrum of the diffuse atmospheric radiance Idiff, extracted with LOWTRAN. As we can see, the spectrum peaks at blue wavelengths, and decreases for wavelengths in the infra-red.
Download figure:
Standard image High-resolution imageThis datum considers only the sky brightness and neglects the fact that the transmitter partially blocks the FOV of the receiver, therefore it is overestimated in realistic scenarios, especially for short links. Nonetheless, we will use it in the following discussion, which focuses on the impact of background on the performance of QKD.
As a first-order approximation, we can consider the diffuse radiance to be uniform over the receiver FOV. We can estimate the detection rate of background photons per detection window at the quantum signal wavelength as a function of the receiver aperture DRx, solid-angle FOV Ω, and filtering bandwidth δλ (h is the Planck's constant and c the speed of light)
For the typically small FOV characteristic of free-space communication systems, Ω can be approximated by
where FOV is the one-dimensional field-of-view.
The FOV of the receiver optical system depends on the optical design: if the optical fiber, or free-space detector, is placed on an image-plane of the entrance pupil, which is the case for free-space detectors with large active diameter or large-core multi-mode fibers (MMF), the FOV is essentially a free parameter, limited only by the size of the optical elements (lenses, mirrors) used in the optical system, and may be as large as 400 μrad. In the case of SMFs or free-space detectors with small active diameter m, the optimal choice is to place them on the focal plane of the optical system. In this configuration the FOV is constrained by the size of the active area or fiber MFD. Since the receiver focal length f is chosen so that the optical system efficiency in equation (19) is maximized, we have
where βopt is shown in figure 5 and βopt = 1.12 for unobstructed apertures. If we define the FOV as the pupil incidence angle at which the spot on the focal plane is deflected to a distance equal to half the MFD, then we have that:
Another consequence of this constraint is that the detection rate of diffuse background photons coupled into the system becomes almost independent of the receiver optical system parameters. Indeed, combining equations (38) and (41), we find that the background light coupled into the SMF is given by
which does not depend on f nor DRx. Figure 18 shows the expected noise count rate for a SMF coupled receiver.
Download figure:
Standard image High-resolution image3.2. Temporal gating
Typical DV-QKD systems apply a temporal filter to all detected events, with the purpose of reducing the impact of noise. Indeed, the latter is uniformly distributed in time whereas signal photons, being emitted at regular intervals, have a predictable time of arrival. In post-processing, one can apply a Tgat-wide temporal window centered at this time and discard all events that fall outside of it, thus suppressing noise by a factor Tgat ⋅ R. However, there is a tradeoff, because the temporal distribution of the signal events is enlarged by the optical pulse width, by the jitter of the source, of the detectors, and of the time-digitizing hardware. A small value of Tgat, while strongly reducing noise, might discard too much of the signal, negatively impacting the final SKR. Assuming a normal distribution of standard deviation J for the signal time of arrival, the filter reduces the signal detection rate by a factor .
A numerical study of the tradeoff can guide the choice of Tgat. The figure of merit to maximize is the final SKR, which includes the contribution of the detection and error rates. We focus on the ratio Tgat/J between it and the standard deviation J of the temporal distribution of the signal (including all the aforementioned jitter contributions). We can expect the tradeoff to be influenced by (i) the signal-to-noise ratio (SNR) when the noise is gated to a window as wide as the signal (±3 times the standard deviation J) and (ii) the coding error.
The SNR obtained before any gating is not sufficient to describe the situation, because it does not consider the width of the signal. Intuitively, for the same ungated SNR, a temporally wider signal favors smaller values of Tgat/J to eliminate more noise. Our definition of the SNR, by considering only the portion of noise that falls under the signal, effectively combines the ungated SNR and the width of the signal in a single parameter. Other protocol parameters such as decoy intensity levels and probabilities also have an importance, but we focus only on the two quantities above for simplicity. All the parameters of the model except the SNR and coding error are arbitrarily fixed to a realistic QKD scenario.
In figure 19, we can see the results of an optimization with the Nelder–Mead algorithm [52] of the Tgat/J ratio to maximize the SKR, for a grid of values of the SNR (noise gated at ±3J) and coding error. Predictably, the smaller the SNR, the smaller the gating window should be, in order to discard more noise. High values of the coding error decrease the optimal Tgat/J. Indeed, although temporal gating alone cannot change the coding error, the higher the total error contribution, the more important it is to reduce it, even at the expense of discarding part of the signal. This shifts the balance of the tradeoff toward smaller windows, as can be seen in the top-left corner of the figure.
Download figure:
Standard image High-resolution imageThe small ridge in figure 19 indicates a sudden jump in the optimal value of Tgat/J. This is because there are several terms contributing to the SKR (see (44)), and each is computed with several methods [27], choosing the best for every configuration of the parameters. This leads to the presence of multiple local maxima: when one of them is promoted to global maximum, overcoming another, the optimal Tgat/J changes. The exact position and entity of the ridge is not universal and depends on the specific scenario that we chose in this simulation, but its presence is to be expected in general when the SKR is optimized.
4. Estimation of the SKR
The rate of production of the secret key in a QKD experiment is calculated from the detection and error rates through a security analysis which bounds the amount of information leaked to an adversary Eve. In a real implementation of the BB84 with decoy states protocol, detections are accumulated for a time tacc until their number reaches a predefined threshold, the block length. Clearly, tacc is inversely proportional to the detection rate Rdet: the slower the rate, the more time it takes to reach the block length. In simulations, we can predict that the number of detections corresponding to each state a sent by Alice, state b observed by Bob, and intensity level k chosen by Alice is
Rdet, equation (37), can be distinguished into a, b, k by calculating R0 of equation (36) for each intensity level, incorporating the probabilities that Alice chooses a and k, and computing the probability that the a state sent as a is measured as b, which is influenced by the contributions to the error rate of section 3.
During the post-processing of QKD, the detections corresponding to different bases are discarded. The remaining (sifted) ones are divided into erroneous (for which a ≠ b) and non-erroneous ones (for which a = b). Given this breakdown of the number of detections, the security analysis can start. Specifically, we follow reference [27] and find
Here, sZ,0 is the lower bound on the number of vacuum detections in the key-generating Z basis. The rate sZ,0/tacc is influenced by the choices of intensities used for the decoy-state method, but strongly depends on the background and dark rate. Although this term has a positive impact on the SKR, these rates should be kept low with physical means (e.g., filtering the background, cooling the detectors, etc) because they increase the error rate. Similarly, sZ,1 is the lower bound on the number of single-photon detections in the key-generating Z basis. Given the same choices of decoy-state parameters, sZ,1/tacc is strongly affected by channel losses and the signal part of the detection rate Rdet. ϕZ is the upper bound on the phase error rate corresponding to single photon pulses, and descends from the ratio between error rate and detection rate. Hence, it is affected by the background and dark rates, and by the coding error. Finally, h(⋅) is the binary entropy, ℓEC and ℓc are the number of bits published during the error correction and confirmation of correctness steps, and , where sec = 10−9 is the secrecy parameter associated to the key.
All these terms are directly accessible from the experimental properties of the raw key block through the formulas reported in reference [27]. Yet, in turn, these properties descend from the detection and error rates that we estimated in the previous sections, and from other protocol parameters. In what follows, we study how to optimize these parameters to maximize the SKR and finally estimate it in some exemplary scenarios.
4.1. Finite-key effects
Because actual experiments accumulate only a finite amount of data, the parameters mentioned above cannot be estimated with perfect precision, leaving an opening for potential attackers. To counter this, QKD uses a broad range of statistical analyses [53–55]. Following reference [27], we construct confidence intervals based on the Hoeffding inequality [56] around the parameters of interest, and then use the pessimistic extrema of the intervals as our estimates. This penalizes the performance of the system, but guarantees that the key is secure with a very high probability 1 − sec.
Fortunately, we can mitigate this cost by optimizing some parameters of the protocol. These are: the probability pZ that each of Alice and Bob choose the key basis Z for their preparations and measurements, the probability pμ that Alice chooses the stronger intensity level, and the intensities μ, ν of each level. Their optimal values that maximize the SKR change depending on the amount of data (block length) that is used in the statistical analysis. In our study, we consider two example QKD scenarios: a high-end system with superconducting nanowire single-photon detectors, a GHz source, and low coding error (scenario A), and a less expensive one with SPADs, a slower source, and a higher coding error. The detector parameters are arbitrary, but similar to those of commercial devices [49, 57]. The most relevant parameters of each scenario are listed in table 2.
Table 2. Relevant parameters for the two scenarios considered in the studies of sections 4.1 and 4.2. For the former, ⟨ηCH⟩ ≈ −7 dB, whereas for the latter, the distributions of figure 14 are used.
Quantity | Scenario A | Scenario B |
---|---|---|
Source repetition rate | 1 GHz | 100 MHz |
Detector efficiency | 80% | 15% |
Coding error | 0.5% | 1.5% |
Dark count rate | 10 Hz | 2 kHz |
Dead time | 10 ns | 20 μs |
Afterpulse probability | 0 | 10% |
Temporal jitter | 10 ps | 200 ps |
Additional receiver losses | 3 dB | 3 dB |
Wavelength | 1550 nm | 1550 nm |
Background photons rate | 5 kHz | 5 kHz |
In figure 20 we show the results of the optimization (with a simulated annealing procedure [58]) when the channel is characterized by the distribution of case 1 (figure 14), with ⟨ηCH⟩ ≈ −7 dB. We can see an upward trend for pZ and pμ for growing block length, indeed the larger the total sample size, the easier it is to accumulate the needed statistics in the check-basis basis X (mutually unbiased with respect to Z) and intensity level ν even if they are chosen rarely. The mean photon numbers μ and ν show more stability, with a slight downward trend for ν in the right-hand side of the plot. This is justified considering that its limit for infinite block length is zero, as this makes the bounds of the decoy-state method tighter. For shorter block lengths, ν must grow to increase the detection rate and accumulate the needed statistics. The optimal value of μ is the one that strikes the right balance between a low multi-photon emission probability and a high detection rate. However, it is also influenced by other factors like afterpulses and detector saturation, which is why it is smaller in scenario B. For low and decreasing block length, as pμ shrinks to increase the statistics accumulated in the low intensity level, μ grows to keep the signal rate high, and ν responds by slightly decreasing to tighten the bounds.
Download figure:
Standard image High-resolution imageThe jump which can be seen in figure 20(b) is similar to the ridge of figure 19: a local maximum is promoted to global and the optimal parameters suddenly move. The exact position and entity of the jump depends strongly on the scenario, but its presence is to be expected in optimizations like these.
In figure 21 we show the cost of finite-key effects and of using a wrong set of parameters. We express it as the ratio between the SKR and SKR∞, i.e., the SKR that would be obtained by optimizing the above parameters for infinite length and by accumulating an infinitely long key block. The dashed line corresponds to the choice of optimizing the parameters for a fixed block length of 107 bits. Generating a key is possible but the SKR quickly drops to zero if block lengths shorter than 107 are used. This happens more slowly in scenario B because the noisier detectors reduce the impact of tuning signal-related parameters.
Download figure:
Standard image High-resolution imageThe solid line shows what happens if the parameters are optimized for each block length. The results improve drastically, underlining the importance of optimizing the protocol parameters for the predicted size that will be used operatively. However, even with a large block length of 109 bits, the SKR is reduced by a sizeable portion with respect to SKR∞. While the precise value of this portion depends on the chosen scenarios, the fact that finite-key effects should not be neglected even for large block lengths is general.
4.2. SKR in typical scenarios
To conclude our analysis and give an example of the capabilities of our full model, we calculate the final SKR of a QKD system considering all the effects we have studied in this work. We do this for the two scenarios of table 2 and the eight channel efficiency distributions of figure 14. As explained in section 2.5, we calculate Rdet using equations (36) and (37) for multiple values of the fluctuating ηCH, and then find its weighted average over the corresponding distributions. For each configuration, we use the simulated annealing algorithm to optimize the protocol parameters (the same of section 4.1) and the temporal gating, for a fixed block length of 107 bits.
The results are shown in figure 22. We can observe the characteristic linear behavior of the SKR with transmittance, and glimpse a drop for strong losses caused by the prevalence of noise. However, thanks to the breadth of phenomena that our model includes, these values go beyond the simple verification of this typical trend, and are accurate estimates of the performance of the considered QKD systems. Scenario B is strongly penalized by its slower source, lower detection efficiency, afterpulses and dead time. Because of this, only the first four distributions yield a positive SKR.
Download figure:
Standard image High-resolution image5. Conclusion and outlook
In this work, we studied many of the relevant phenomena that influence the performance of a ground-to-ground QKD BB84 system. Particular focus was given to the channel model that estimates the efficiency of the link, considering atmospheric absorption, turbulence-induced beam broadening, wandering, scintillation, and the effect of SMF coupling in the presence of AO correction. Many parameters enter the design of a free-space QKD system, which makes it impossible to construct a single optimization function over the entire parameter space. Trade-offs between different design choices should be made considering target performance, system complexity, and cost. The first inputs of the design are the fixed constraints, typically the wavelength λ, the link distance z, the channel turbulence strength , and the target SKR. In our work, we showed that the choice of the transmitter waist W0 is less critical for longer links, but has an effect on the channel fluctuations as wandering increases for smaller beams.
We further showed how AOs can reduce losses and suggested ways to optimize the receiver diameter—when the AO correction order is fixed by constraints—or vice versa how to optimize the AO correction order—when the receiver diameter is fixed, and we factor in the bandwidth limitation of the AO control loop. We found that calculating only the mean efficiency can sometimes be insufficient, and the entire probability distribution is needed. This is because of the saturation of single-photon detectors, which may suppress the high tails of distribution, reducing the detection rate more than one would expect by considering only the mean.
We analyzed most of the sources of error in QKD and some mitigation techniques, showing how to find the best temporal gating to filter out noise. We included also the finite-key effects that reduce the performance because of imperfect parameter estimation. We highlighted how optimizing the probabilities of basis choice and the properties of the decoy states can alleviate this cost. Finally, we put everything together to estimate the final SKR in some example scenarios.
Our model can be expanded further, for instance to include tracking imprecision in moving links and imperfect quantum state preparation beyond the coding error. However, it is comprehensive enough to guide the design of QKD systems and underline what problems should be considered. This can help the implementation and deployment of free-space daylight links in future QKD networks.
Acknowledgments
This work was supported by Agenzia Spaziale Italiana, project Q-SecGroundSpace (Accordo n. 2018-14-HH.0, CUP: E16J16001490001). CloudVeneto is acknowledged for the computational resources.
Data availability statement
Data supporting the findings of this study are available within the article.
Appendix A.: Notations
We report in tables 3, 4 and 5 the notations and the acronyms used in the manuscript.
Table 3. Notation with Greek symbols used in the manuscript.
Symbol | Meaning |
---|---|
α | Obscuration ratio of the telescope aperture |
Attenuation factor for the nth order mode | |
Γ(⋅) | Gamma function |
δλ | Filtering bandwidth |
Transfer function | |
sec | Secrecy parameter |
η0 | Optical efficiency of the receiver telescope |
ηA | Atmospheric channel absorption efficiency |
ηAO | Coupling efficiency due to wavefront perturbations, that may be partially corrected by AO |
ηCH | Channel efficiency |
ηdet | Fixed efficiency terms at the detector, not accounted in ηCH |
Receiver collection efficiency | |
ηS | Coupling efficiency due to the spatial structure of atmospheric scintillation |
ηSMF | Single-mode-fiber coupling efficiency |
θ0 | Diffraction-limited beam divergence |
κ | Spatial frequency |
λ | QKD signal wavelength |
μ, ν | Mean photon numbers of photons per pulse used in the QKD protocol |
Frequency of the AO loop | |
Cut-off frequency for the nth Zernike mode | |
ξ(t) | Instantaneous sum of the squared Zernike coefficients |
ρ0 | Atmospheric spatial coherence radius |
Log-amplitude variance | |
Aperture-average scintillation index (flux variance) | |
Pupil-plane scintillation index | |
Rytov variance | |
τ | Overall latency of the control-actuator state |
ϕZ | Upper bound on the phase error rate |
Φn(κ) | Power spectral density of refractive index fluctuations |
Log-amplitude perturbation term | |
Wavefront phase term | |
Ω | Solid-angle field of view |
Table 4. Acronyms used in the manuscript.
Symbol | Meaning |
---|---|
AO | Adaptive optics |
CV | Continuous variable |
DV | Discrete variable |
FOV | One-dimensional field of view |
MFD | Mode field diameter of the SMF |
MMF | Multi-mode fiber |
QKD | Quantum key distribution |
Rx | Receiver |
SNR | Signal-to-noise ratio |
SMF | Single mode fiber |
SKR | Secret key rate |
SKR∞ | SKR for infinite key length |
SPAD | Single-photon avalanche diode |
Tx | Transmitter |
Table 5. Notation with Latin symbols used in the manuscript.
Symbol | Meaning |
---|---|
A, A(λ) | Absoprtion coefficient |
, | Coefficients of Zernike expansion |
Zernike coefficient variance | |
Cχ (r) | Log-amplitude spatial covariance function |
Refractive-index structure constant | |
DRx | Receiver aperture diameter |
erf(⋅) | (Gauss) error function |
f | Effective focal length of the receiver |
Open-loop transfer function | |
h(⋅) | Binary entropy h(p) := −p log2 p − (1 − p)log2(1 − p) |
Idiff | Daylight diffuse atmospheric radiance |
j | OSA/ANSI index |
J | Standard deviation of the signal time of arrival |
k | Wave-number |
Ki | Gain of the integrator of the AO loop |
ℓEC | Number of bits published during the error correction step |
ℓc | Number of bits published during the confirmation of correctness step |
m | Azimuthal degree of Zernike modes expansion |
n | Air refractive index |
n | Radial degree of Zernike modes expansion |
nmax | Maximum radial degree corrected by AO |
pμ , pν | Probabilities of choosing the intensity levels μ,ν |
pap | Probability of triggering an afterpulse event |
plabel, ηlabel | plabel is the probability distribution of ηlabel efficiency |
pZ , pX | Probabilities of choosing the key-generatin basis Z or the check-basis basis X |
r0 | Atmospheric coherence width, also known as the Fried parameter |
(r, φ) | Normalized pupil coordinates |
R | Repetition rate of the QKD source |
R0 | Raw output rate of a single photon detector (not corrected for saturation) |
Rbkg | Background photon rate |
Rdark | Dark count rate |
Rdet | Actual output detection rate of a single photon detector (corrected for saturation) |
Rsat | Saturation rate of the detector |
sZ,0 | Lower bound on vacuum detections in the key-generating basis |
sZ,1 | Lower bound on single-photon detections in the key-generating basis |
tacc | Key accumulation time |
T | Wavefront sensor integration time |
Td | Dead time of the detector |
Tgat | With of the temporal gating of the detection events |
Average wind velocity | |
Power spectral density of the temporal spectrum of the nth order aberration | |
W0 | Beam waist |
W(z) | Beam size (radius) at a distance z |
WST(z) | Short-term beam size at a distance z |
X | Check-basis basis |
z | Link distance |
z0 | Rayleigh range |
Z | Key-generating basis |
Zernike polynomial of radial degree n and azimuthal degree m |