Conference PaperPDF Available

On the Filter Issues in Multiplexing Classical and QKD Links through WSS-based Nodes

On the Filter Issues in Multiplexing Classical and
QKD Links through WSS-based Nodes
D. Zavitsanos1, G. Giannoulis1, A. Raptakis1, C. Kouloumentas1, and H. Avramopoulos1
1School of Electrical and Computer Engineering, National Technical University of Athens, Athens, Greece
e-mail address:
Abstract: We present an extensive study on the quantum link multiplexed with intense classical
signal in a WSS-based node. Theoretical results on QBER and SKR performance revealed the
dependence of BB84-QKD link on passband spectral response.
OCIS codes: (060.5565) Quantum communications, (230.7408) Wavelength filtering devices.
1. Introduction
Besides the deployment of wireless QKD links for global-scale quantum networks [1], the quantum communication
infrastructure will be integrated in the deployed fiber infrastructure where classical optical streams carrying large
amount of traffic are going to coexist with quantum links being responsible to guarantee the security layer operating
in single-photon level. This shared use of installed optical infrastructure layer makes a strong economic case
considering the high costs of new fiber deployment while it also leverages the significant progress in the field of elastic
optical networks and prevents synchronization issues over separate links [2]. Exploiting the legacy of classical
Wavelength Division Multiplexing (WDM) networking and boosted from the Software Defined Networking (SDN)
capabilities of elastic optical topologies [3], the classical-quantum coexistence schemes relying on the use of WDM-
enabled nodes with enhanced flexibility have been promoted in reconfigured optical networking segments [4]. So far,
several theoretical and experimental studies have been conducted to investigate the impact of different noise photon
sources on the QKD link performance under different coexistence scenarios enabled through Reconfigurable Optical
Add Drop Multiplexer (ROADM) nodes [4-6]. Emphasis has been placed on the study of noise sources associated
with the photons generated through the nonlinear interaction of intense classical channels with the telecom fibers, such
as the Four-Wave Mixing (FWM) and the spontaneous Raman scattering. In contrast, the role of the passband filtering
has been reported much more scarcely through the literature with limited contributions focusing on its impact through
channel crosstalk mechanism [7]. Moreover, even though the system requirements for Wavelength Selective Switch
(WSS) filter shapes have been identified as critical parameters for WDM-enabled classical reconfigured environments
[8], this perspective is still missing from the studies on the ROADM-based nodes for QKD networking.
Our research attempts to fill the above gap by presenting a numerical study on the performance evaluation of a Discrete
Variable-QKD (DV-QKD) link implemented through state-of-the-art WSS-based nodes. We present for the first time,
to the best of our knowledge, a thorough study focusing on the effect of the filtering parameters of the WSS-based
node on the quantum layer performance. The Quantum Bit Error Rate (QBER) performance and the Secure Key Rate
(SKR) were calculated for the phase coding BB84 protocol by considering bandwidth and shape parameters of channel
bands for Liquid Crystal on Silicon (LCoS)-based WSS nodes. Critical system requirements such as insertion loss and
center frequency drifts were also considered in our study.
2. Spectral modeling of WSS-based node supporting QKD multiplexing
Fig.1a illustrates the proposed WSS-based ROADM node suited for spectrally multiplexing a classical flow with the
QKD link. Our study aims to identify the crosstalk penalty induced by the leakage photons of adjacent classical
channel to the quantum passband. For this purpose, we assumed a coexistence scheme through this node while the
transmission of the quantum signal was considered for a lossless fiber segment without any noise sources associated
with fiber nonlinearities. The proposed ROADM node was implemented by considering the channel band shapes of
WSS, a kind of filtering nodes that has been widely used in flexible-λ classical networking scenarios and has also been
extensively investigated regarding its spectral modeling. The optical field spectrum of the bandpass is given by [9]:
 
where  is the standard deviation of the Gaussian Optical Transfer Function (OTF),  the
OTF Bandwidth and B the Bandwidth of the channel shapes. Fig. 1b illustrates the obtained channel shape of a 50
GHz bandwidth WSS for two different BWOTF values, revealing the strong effect of the BWOTF parameter to the
spectral components of lower amplitude which are responsible for the contamination to the quantum passband. Since
the quantum link is extremely sensitive to leakage noise photons, a 100 GHz grid WDM topology was selected as
depicted in the inset of Fig.1.a, where the filtering isolation varies from ~85 dB to ~105 dB, as shown in the Fig.1b.
Fig. 1. a) Proposed WSS-based node multiplexing classical signal and phase-coding BB84-QKD links. b) The bandpass filter response, 
of 50 GHz Bandwidth for different BWOTF parameters (13.4 GHz, 15 GHz). The inset illustrates an added leakage power as the BWOTF increases.
For the quantum layer, we considered a phase coding BB84-QKD system where the visibility was set at 99%. The
optical pulses were generated with a repetition rate of 5 MHz. The photons were assumed to be detected by a pair of
Infrared Single-Photon Detectors (, ) operated in gated mode with a time-window of 1 ns for the detection. The
Single Photon Avalanche Diode (SPAD) unit was assumed to exhibit a dead time of , an afterpulse probability
of 0.8%, a Dark Count Rate (DCR) of  and a quantum efficiency of . The insertion loss of the WSS
was initially set to be 2.7 dB and the optical loss of the internal components at Bob side was fixed at 2.65 dB.
To identify the amount of the leakage power generated from the classical Power Spectrum Density (PSD) with average
power of 0 dBm, we calculated the leakage ratio by integrating within the quantum passband of 100 GHz:
  [10]. In order to evaluate the QBER and the SKR values, we expressed the leakage power in terms
of noise photons per ns (gate duration time) and performed a numerical optimization of the average photon number
per pulse to achieve the maximum rate [6].
3. Results and Discussion
Fig.2 illustrates the QBER and the SKR performance as functions of the BWOTF and Bandwidth (B) parameters
respectively. By keeping constant the B parameter at 50 GHz, as the OTF value increases the power leaked in the
quantum passband becomes higher, as shown in the inset of Fig.1b. This higher leakage ratio naturally leads to higher
QBER and lower SKR values as depicted in Fig. 2a. By increasing the bandwidth parameter for a fixed OTF value
(13.4 GHz) the QBER penalty was estimated to be ~ 0.6% while the calculated SKR was found to be above the value
of 14.4 Kbps (Fig. 2b).
Fig. 2. a) Dependence of QBER and SKR on filter BWOTF parameter, for B = 50 GHz. b) Dependence of QBER and SKR on filter’s Bandwidth,
for BWOTF=13.4 GHz.
The QBER performance wasn’t drastically degraded since the change on B parameter affects mainly the spectrally
components located closer to the passband peak with significantly lower impact on the lower amplitude components.
On the contrary, the BWOTF affects mainly the spectral components at the tails of the channel shape leading thereby
to significant changes to the spectral isolation performance as indicated in Fig.1b. Consequently, in contrast with the
classical crosstalk penalties which are dominated from the spectral components of higher amplitude, the role of the
BWOTF parameter becomes essential for WSS-based ROADM nodes supporting the multiplexing of QKD links.
Degradation on the filtering isolation may also be caused by a center frequency shift (Δfc) of the channel passband
shape. For this case, we assumed that the optical PSD - suited for classical passband - shifts closer to the quantum
passband, contaminating thereby with severe leakage photons the QKD passband. Fig. 3a shows that, for fixed values
of Bandwidth, , and OTF Bandwidth,  , a QBER penalty of ~ 6% was calculated for
a Δfc value of 3 GHz. This obtained ultra-high sensitivity of QKD link performance on the Δfc parameter of the WSS-
based node can be useful for monitoring and control purposes, a great challenge for classical networks exploiting
cascaded ROADMs [8]. We also identified the dependence of QKD performance on the insertion loss of WSS-based
node (Fig. 3b). By increasing the optical loss of the WSS from 2.7 dB to 4.7 dB, the QBER is remained practically
constant (~ 0.1% penalty) since the number of leakage photons arrived at Bob side attenuated in the same way as the
encoded photons generated from Alice. On the contrary, the SKR decreases rapidly from 16 Kbps to 6 Kbps, since
the click rate (and so the sifted key rate) becomes apparently lower. It should be also mentioned that in all the above
cases, the QBER approaches the baseline error rate (~ 0.9%), which is determined by the interference visibility and
the false detection events associated with the SPAD units (e.g. DCR, afterpulse probability).
Fig. 3. a) Dependence of QBER and SKR on centre frequency shift, for B = 50 GHz and BWOTF=13.4 GHz. b) Dependence of QBER and SKR
on filter’s insertion loss increase, for B = 50 GHz and BWOTF=13.4 GHz.
4. Conclusions
We presented a thorough investigation through numerical simulations on the wavelength multiplexing of an intense
classical signal with a weak QKD link. The WSS-based node parameters assessing the channel passband shapes
strongly affect the QKD link performance since their variation leads to strong leaking ratios which consequently affect
the QBER and SKR values. The role of BWOTF parameter has a great impact on the BB84-QKD link performance
while the quantum layer performance is also extremely sensitive to the center frequency shifts of the channel
passbands. The reported results contribute towards developing a broader understanding of how the currently deployed
WSS-based nodes could be exploited for practical scenarios in shared quantum/classical networks.
This work has been supported by the European Commission through H2020 FET-Flagship project UNIQORN, Contract Number 820474.
[1] S.-K. Liao, et al., "Satellite-to-ground quantum key distribution.", Nature 549, 43-47 (07 September 2017).
[2] T.A. Eriksson, et al., "Crosstalk Impact on Continuous Variable Quantum Key Distribution in Multicore Fiber Transmission", IEEE Photonics
Technology Letters, Vol.31, No.6, pp. 467-470 (2019).
[3] Y. Ou, et al., "Field-Trial of Machine Learning-Assisted Quantum Key Distribution (QKD) Networking with SDN", in Proc. of ECOC 2018,
Rome Italy.
[4] N. A. Peters, et al. "Dense wavelength multiplexing of 1550 nm QKD with strong classical channels in reconfigurable networking
environments." New Journal of physics 11.4 (2009): 045012.
[5] P. Toliver, et al., "Demonstration of 1550 nm QKD with ROADM-based DWDM Networking and the Impact of Fiber FWM", in Proc. of CLEO
2007, paper No. CThBB1, Baltimore, MD, USA.
[6] D. Zavitsanos, et al., "Coexistence of Discrete-Variable QKD with WDM classical signals in the C-band for fiber access environments", in
Proc. of ICTON 2019, paper No. Th.A5.4, Angers, France.
[7] T.-J. Xia, et al., "In-Band Quantum Key Distribution (QKD) on Fiber Populated by High-Speed Classical Data Channels", in Proc. of OFC
2006, paper No. OTuJ7, Anaheim, CA, USA.
[8] F. Heismann, "System Requirements for WSS Filter Shape in Cascaded ROADM Networks", in Proc. of OFC 2010, paper No. OThR1, San
Diego, CA, USA.
[9] C. Pulikkaseril, et al., "Spectral modeling of channel band shapes in wavelength selective switches", Optics Express, Vol.19, No.9 (2011).
[10] T. Honjo, T. Inoue, K. Inoue, "Influence of light source linewidth in differential-phase-shift quantum key distribution systems. " Optics
Communications, 284(24), 5856-5859 (2011).
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
Differential-phase-shift (DPS) quantum key distribution (QKD) is one of the QKD protocols, featuring simplicity for practical implementation. It uses a coherent pulse train whose phase should be stable at least within the pulse interval. This paper quantitatively investigates the phase stability required for DPS-QKD systems. The phase stability is characterized by the spectral linewidth of the light source. A theoretical model and experiments are presented, the results of which indicate that the linewidth should be, for example, less than 0.35% of the free-spectral-range of an asymmetric Mach–Zehnder interferometer in a receiver to achieve quantum bit error rate of less than 0.5% due to linewidth broadening of the light source.
Full-text available
To move beyond dedicated links and networks, quantum communications signals must be integrated into networks carrying classical optical channels at power levels many orders of magnitude higher than the quantum signals themselves. We demonstrate the transmission of a 1550 nm quantum channel with up to two simultaneous 200 GHz spaced classical telecom channels, using reconfigurable optical add drop multiplexer (ROADM) technology for multiplexing and routing quantum and classical signals. The quantum channel is used to perform quantum key distribution (QKD) in the presence of noise generated as a by-product of the co-propagation of classical channels. We demonstrate that the dominant noise mechanism can arise from either four-wave mixing or spontaneous Raman scattering, depending on the optical path characteristics as well as the classical channel parameters. We quantify these impairments and discuss mitigation strategies.
Conference Paper
Full-text available
We used a commercial QKD system to conduct key exchange over 50-km fiber span that supported classical 10 Gbit/sec traffic. Both quantum and classical channels were carried by the standard ITU grid C-band wavelengths.
We investigate the excess noise induced by crosstalk in a trench-assisted homogeneous 19-core fiber for a continuous variable quantum key distribution (CV-QKD) receiver. Crosstalk is studied for both 30 WDM channels of 24.5 Gbaud PM-16QAM with 100 GHz channel spacing and for an ASE signal with 37.5 GHz bandwidth. We show that in-band crosstalk from neighboring cores prohibits secret key generation. However, it is possible to place CV-QKD channels at wavelengths not used by the classical channels even though wideband amplified spontaneous emission noise is present. Excess noise induced by crosstalk from a neighboring core is shown to be time dependent.
Quantum key distribution (QKD) uses individual light quanta in quantum superposition states to guarantee unconditional communication security between distant parties. In practice, the achievable distance for QKD has been limited to a few hundred kilometers, due to the channel loss of fibers or terrestrial free space that exponentially reduced the photon rate. Satellite-based QKD promises to establish a global-scale quantum network by exploiting the negligible photon loss and decoherence in the empty out space. Here, we develop and launch a low-Earth-orbit satellite to implement decoy-state QKD with over kHz key rate from the satellite to ground over a distance up to 1200 km, which is up to 20 orders of magnitudes more efficient than that expected using an optical fiber (with 0.2 dB/km loss) of the same length. The establishment of a reliable and efficient space-to-ground link for faithful quantum state transmission constitutes a key milestone for global-scale quantum networks.
We describe the impact of passband shape and random center frequency variations on the bandwidth narrowing caused by successive filtering in concatenated wavelength-selective switches. Passband flatness is extremely important for signals at 40-Gb/s and beyond.
We demonstrate compatibility of 1550 nm QKD with a MEMS-based ROADM and also show that four-wave mixing resulting from copropagating DWDM signals can become the dominant source of background noise within the QKD channel passband. ©2007 Optical Society of America OCIS Codes: (060.2330) Fiber optics communications; (190.4380) Nonlinear optics, four-wave mixing.
A model for characterizing the spectral response of the passband of Wavelength Selective Switches (WSS) is presented. We demonstrate that, in contrast to the commonly used supergaussian model, the presented model offers a more complete match to measured results, as it is based on the physical operation of the optical system. We also demonstrate that this model is better suited for calculation of WSS channel bandwidths, as well as predicting the final bandwidth of cascaded WSS modules. Finally, we show the utility of this model in predicting channel shapes in flexible bandwidth WSS channel plans.