ETSI GS QKD 003 V1.1.1 (2010-12)

Quantum Key Distribution (QKD); Components and Internal Interfaces

DGS/QKD-0003_CompInternInterf

General Information

• Status: Withdrawn
• Publication Date: 08-Dec-2010
• Withdrawal Date: 07-Nov-2018
• Technical Committee: QKD - Quantum Key Distribution
• Current Stage: 50 - Withdrawn
• Due Date: 15-Nov-2018
• Completion Date: 08-Nov-2018

Standards Content (Sample)

Group Specification
Quantum Key Distribution (QKD);
Components and Internal Interfaces
Disclaimer
This document has been produced and approved by the Quantum Key Distribution (QKD) ETSI Industry Specification Group
(ISG) and represents the views of those members who participated in this ISG. It does not necessarily represent the views of
the entire ETSI membership.
2 ETSI GS QKD 003 V1.1.1 (2010-12)

Reference
DGS/QKD-0003_CompInternInterf
Keywords
interface, Quantum Key Distribution
ETSI
650 Route des Lucioles
F-06921 Sophia Antipolis Cedex - FRANCE

Tel.: [phone]  Fax: [phone]

Siret N° 348 623 562 00017 - NAF 742 C
Association à but non lucratif enregistrée à la
Sous-Préfecture de Grasse (06) N° 7803/88

Important notice
Individual copies of the present document can be downloaded from:
http://www.etsi.org
The present document may be made available in more than one electronic version or in print. In any case of existing or
perceived difference in contents between such versions, the reference version is the Portable Document Format (PDF).
In case of dispute, the reference shall be the printing on ETSI printers of the PDF version kept on a specific network drive
within ETSI Secretariat.
Users of the present document should be aware that the document may be subject to revision or change of status.
Information on the current status of this and other ETSI documents is available at
http://portal.etsi.org/tb/status/status.asp
If you find errors in the present document, please send your comment to one of the following services:
http://portal.etsi.org/chaircor/ETSI_support.asp
Copyright Notification
No part may be reproduced except as authorized by written permission.
The copyright and the foregoing restriction extend to reproduction in all media.

© European Telecommunications Standards Institute 2010.
All rights reserved.
TM TM TM TM
DECT , PLUGTESTS , UMTS , TIPHON , the TIPHON logo and the ETSI logo are Trade Marks of ETSI registered
for the benefit of its Members.
TM
3GPP is a Trade Mark of ETSI registered for the benefit of its Members and of the 3GPP Organizational Partners.
LTE™ is a Trade Mark of ETSI currently being registered
for the benefit of its Members and of the 3GPP Organizational Partners.
GSM® and the GSM logo are Trade Marks registered and owned by the GSM Association.
ETSI
3 ETSI GS QKD 003 V1.1.1 (2010-12)
Contents
Intellectual Property Rights . 4
Foreword . 4
1 Scope . 5
2 References . 5
2.1 Normative references . 5
2.2 Informative references . 5
3 Definitions and abbreviations . 6
3.1 Definitions . 6
3.2 Abbreviations . 6
4 QKD System Components . . 7
4.1 Generic Description . 7
4.2 Weak Laser Pulse QKD . 7
4.2.1 One-Way Mach-Zehnder Implementation . 8
4.2.2 Send and Return Mach Zehnder Implementation . 9
4.2.3 Phase-Intensity Modulator Implementation . 10
4.3 Entanglement-based QKD . 10
4.4 Continuous-Variable QKD . 11
4.4.1 Principle of Continuous-Variable QKD Protocols . 11
4.4.2 Implementation Example of the CV QKD protocol. . 11
5 Photon Detector . 12
5.1 Single-Photon Detector . 12
5.1.1 Generic Description and Parameterisation . 12
5.1.2 Test Measurements . 15
5.1.3 InGaAs Avalanche Photodiodes . 17
5.2 Photon Detector for a CV-QKD Set-up . 19
5.2.1 Coherent Detection . 19
5.2.2 Multiplexing . 20
5.2.3 Homodyne Detection . 20
5.2.4 Heterodyne Detection . 20
6 QKD Source . 20
6.1 Generic Description and Parameterisation . 20
6.2 Test Measurements . 22
6.3 Single-Photon Sources . 24
6.4 Weak Pulses . 25
6.4.1 Weak Laser . 25
6.4.2 Intensity-Modulated Weak Laser . 26
6.4.3 Composite Weak Laser . 27
6.5 Continuous-Variable QKD Source . 27
7 Modulators . 28
History . 30

ETSI
4 ETSI GS QKD 003 V1.1.1 (2010-12)
Intellectual Property Rights
IPRs essential or potentially essential to the present document may have been declared to ETSI. The information
pertaining to these essential IPRs, if any, is publicly available for ETSI members and non-members, and can be found
in ETSI SR 000 314: "Intellectual Property Rights (IPRs); Essential, or potentially Essential, IPRs notified to ETSI in
respect of ETSI standards", which is available from the ETSI Secretariat. Latest updates are available on the ETSI Web
server (http://webapp.etsi.org/IPR/home.asp).
Pursuant to the ETSI IPR Policy, no investigation, including IPR searches, has been carried out by ETSI. No guarantee
can be given as to the existence of other IPRs not referenced in ETSI SR 000 314 (or the updates on the ETSI Web
server) which are, or may be, or may become, essential to the present document.
Foreword
This Group Specification (GS) has been produced by ETSI Industry Specification (ISG) Group Quantum Key
Distribution (QKD).
ETSI
5 ETSI GS QKD 003 V1.1.1 (2010-12)
1 Scope
The present document is a preparatory action for the definition of properties of components and internal interfaces of
QKD Systems. Irrespective of the underlying technologies, there are certain devices that appear in most QKD Systems.
These are e.g. quantum physical devices such as photon sources and detectors, or classical equipment such as protocol
processing computer hardware and operating systems. For these components, relevant properties must be identified that
will subsequently be subject to standardisation. Furthermore, a catalogue of relevant requirements for interfaces
between components must be established, to support the upcoming definition of internal interfaces.
2 References
References are either specific (identified by date of publication and/or edition number or version number) or
non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the
reference document (including any amendments) applies.
Referenced documents which are not found to be publicly available in the expected location might be found at
http://docbox.etsi.org/Reference.
NOTE: While any hyperlinks included in this clause were valid at the time of publication ETSI cannot guarantee
their long term validity.
2.1 Normative references
The following referenced documents are necessary for the application of the present document.
Not applicable.
2.2 Informative references
The following referenced documents are not necessary for the application of the present document but they assist the
user with regard to a particular subject area.
[i.1] J. F. Dynes et al, Opt. Express 15, 8465 (2007).
[i.2] N Gisin et al, Rev.Mod. Phys. 74, 145 (2002).
[i.3] L.Duraffourg et al, Opt. Lett 26, 18 (2001).
[i.4] A. Ekert, Phys. Rev. Lett. 67, 661 (1991).
[i.5] J. Clauser et al., Phys. Rev. Lett. 23, 880-884 (1969).
[i.6] C. H. Bennett, G. Brassard and N. D. Mermin Phys. Rev. Lett. 68, 557 (1992).
[i.7] Fossier et al., New J. Phys. 11 045023 (2009.
[i.8] Leverrier & Grangier, Phys. Rev. Lett. 102, 180504 (2009).
[i.9] Dixon et al, Applied Physics Letters 94, 231113 (2009).
[i.10] Appl. Phys. Lett. 91, 041114 (2007).
[i.11] Lodewyck & Grangier, Phys. Rev. A 76, 022332 (2007).
[i.12] Fossier et al., J. Phys.: Atomic, Molecular and Optical Physics 42, 114014 (2009).
[i.13] Intallura et al., J. Opt. : Pure Appl. Opt. 11, 054005 (2009).
ETSI
6 ETSI GS QKD 003 V1.1.1 (2010-12)
3 Definitions and abbreviations
3.1 Definitions
For the purposes of the present document, the following terms and definitions apply:
Alice: quantum information sender/transmitter in a QKD system
Bob: quantum information receiver in a QKD system
classical channel: communication channel that is used by two communicating parties for exchange of classical
information
Eve or eavesdropper: any adversary intending to intercept communication between Alice and Bob
intensity modulator: device that can actively set the intensity of an optical pulse that is passing through the modulator
phase modulator: device that can actively set the phase of a photon that is passing through the modulator
quantum channel: communication channel for transmitting quantum signals
quantum photon source: optical source for carrying quantum information
random number generator: physical device outputting unpredictable binary bit sequences
single-photon detector: device that transforms a single-photon into a detectable signal with finite probability
single-photon source: photon source that emits at most one photon at a time
weak laser pulse: optical pulse obtained through attenuating a laser emission
NOTE: A weak laser pulse typically contains less than one photon per pulse on average.
3.2 Abbreviations
For the purposes of the present document, the following abbreviations apply:
AMZI Asymmetric Mach-Zehnder Interferometer
APD Avalanche PhotoDiode
BB84 QKD protocol published by Bennett and Brassard in 1984
BNC Bayonet Neill-Concelman connector
CV Continuous Variable
DC direct current
ECL Emitter Coupled Logic
LDPC Low Density Parity Check codes
LO Local Oscillator
NIM Nuclear Instrumentation Module
PNS Photon Number Splitting attack
QKD Quantum Key Distribution
QND Quantum Non Demolition
SM Single Mode
SMA Sub-Miniature version A connector
SPC Single-Photon Counting
SPDC Spontaneous Parametric Down-Conversion process
TTL Transistor-Transistor Logic
ETSI
7 ETSI GS QKD 003 V1.1.1 (2010-12)
4 QKD System Components
4.1 Generic Description
A QKD system is comprised of a number of internal components. The purpose of the present document is to identify the
components which are common to many systems, to define the interfaces between these common components, to define
how these components shall be characterised in a relevant and controlled manner and to define the component
performance required for QKD.
A survey of the academic literature reveals that there have been many different types of QKD system proposed. Many
of these have been implemented physically with different levels of sophistication. At the most basic level these systems
utilise the laws of quantum theory to make claims about the security levels of the shared key. Most commonly they use
signal encoding upon quantum light states using several different bases which are non-orthogonal to one another.
Quantum theory dictates that it is impossible to gain full information of this encoding through measurement without
prior information about the encoding bases or post-selection of the bases used. In QKD this property is used to ensure
that the legitimate users of the system share more information, than an eavesdropper can determine.
One convenient method of categorising different types of QKD systems is according to the photon source that they use.
Examples include single-photon sources, entangled photon pair sources and weak laser pulses. Common methods for
encoding the qubit information include controlling the phase or the polarisation state of the transmitted photon.
A QKD system consists of two units which are physically separated at opposite ends of a communication channel as
illustrated by figure 4.1. The sending unit consists of a signal source and an encoder for the source. The sending and
receiving unit consist a source of randomness for use in the key generation protocol. The source of randomness can be
either an active random number generator or a passive random selection component, such as a non-polarising beam
splitter. The receiving unit consists of a component for signal demodulation, or in other words for selecting the
measurement basis, as well as one or more signal detectors. Control electronics, with access to an independent random
number generator, is necessary to generate the drive signals for these devices. The detected signals are used by the
control electronics to form the shared key.

Figure 4.1: Schematic of a generic QKD system showing internal interfaces and connections
4.2 Weak Laser Pulse QKD
In weak laser pulse QKD, the bit values are encoded upon attenuated laser pulses. The sender (Alice) in a weak laser
pulse QKD contains at least one weak laser pulse source that is used as quantum information carrier. In
implementations involving more than one weak laser source, the sources must be indistinguishable from one another in
every measurable attribute.
ETSI
8 ETSI GS QKD 003 V1.1.1 (2010-12)
Alice shall consist also a quantum encoder that encodes qubit information on each weak laser pulse. This encoder shall
have a source of randomness that determines an encoding basis and an encoding bit value for each weak pulse. The
source of randomness shall come from either a random number generator or a passive optical component that acts as a
source of randomness.
The photon number splitting attack must be appropriately included in the privacy amplification process in a QKD
session. To achieve this, the intensity and photon number statistics of each weak laser source shall be carefully
calibrated. The source stability shall also be calibrated. In the case that the source is instable, the worst case scenario
shall be considered in the privacy amplification process.
In the following, we give a few example realizations of weak laser pulse QKD systems.
4.2.1 One-Way Mach-Zehnder Implementation
Figure 4.2 shows an example of a QKD system using weak laser pulses as the signal carriers and Asymmetric
Mach-Zehnder Interferometers (AMZI) to encode the quantum states, based on the paper by J. F. Dynes et al, Opt.
Express 15, 8465 (2007) [i.1]. The system shown uses the decoy pulse protocol to obtain higher secure bit rates than are
otherwise possible using weak laser pulses. In one implementation, the transmitter source is a distributed feedback laser
and emits a fixed intensity train of pulses at a repetition rate of 7,143 MHz. An intensity modulator is used to produce
signal and decoy pulses of differing intensities. The vacuum decoy pulse is produced by omitting trigger pulses to the
signal laser. All signal, decoy and vacuum pulses are produced at random times and have pre-determined relative
occurrence probabilities assigned to them. The signal and decoy pulses are attenuated strongly to the single-photon level
after which a much stronger clock pulse is wavelength division multiplexed with them to provide synchronization
between Alice and Bob's electronics (not shown in figure 4.2).

Figure 4.2: Schematic of a one-way, weak-laser-pulse QKD system
Bob's detectors are two single-photon InGaAs avalanche photodiodes (APDs), operated in conventional gated Geiger
mode.
This system uses active stabilisation to lock the path difference in the sending and receiving AMZI. The pulsed
1 550 nm laser generates both weak signal pulses and a later, stronger reference pulses with a delay in one arm. Care
must be taken to ensure that the reference pulses are not modulated by the phase modulator in the sending AMZI.
Detection of the reference pulses in the reference detector is used to provide a feedback signal to vary the setting of the
fibre stretcher in the receiving AMZI. A similar active stabilisation technique is used to control the polarisation state
entering Bob's AMZI.
In this implementation, combination of the 1 550 nm laser diode, the intensity modulator and the attenuator forms the
photon source. Because only one laser diode is used for encoding all qubits, the indistinguishability of the source is
naturally guaranteed. Use of an intensity modulator is required to implement the decoy QKD protocol. Alice's AMZI is
the encoder. The standard Single Mode (SM) fibre is used as the quantum channel. In Bob, Combination of the active
polarisation recovery, active fibre stretcher and AMZI forms the decoder.
The source of randomness arises from a random number generator in each control electronics.
ETSI
9 ETSI GS QKD 003 V1.1.1 (2010-12)
4.2.2 Send and Return Mach Zehnder Implementation
Figure 4.3 depicts a typical send-and-return Mach-Zender architecture, described in detail in N Gisin et al, Rev.Mod.
Phys. 74, 145 (2002) [i.2]. Pulses emitted from the source S are split in two pulses. The first pulse propagates along the
short arm and launches in the quantum channel through a polarization splitter PS. The second pulse is delayed and its
polarization is rotated using the polarisation rotator (PR) by 90 degrees and launched into the quantum channel via the
same polarisation beam splitter. The phase shifter present in this long arm is left inactive. The polarization splitter PS
allows launching the second pulse into the quantum channel. At the emitting unit, a beam splitter BS2 reflects a weak
part of the incoming signals to a detector D3:
i) providing a timing signal; and
ii) preventing so-called Trojan horse attacks.
The transmitted pulses are then reflected by a Farady mirror (FM) to compensate any birefringence effects of the
quantum channel. An attenuator (AT) allows reducing the intensity of the pulses to a suitably weak intensity (depending
on the protocol used). A phase difference Φ1 is then introduced between the delayed pulses in order to encode a bit
value. At the receiving unit, the pulses are separated by the polarization splitter PS and a phase Φ2 is applied to one of
the two pulses to implement the measurement basis choice. Single-photon detectors D1 and D2 are then used to indicate
which output port was chosen by the photon. A circulator C ensures the isolation between the laser source and the
photon detectors.
Figure 4.3: Plug and Play phase-intensity modulator system
ETSI
10 ETSI GS QKD 003 V1.1.1 (2010-12)
4.2.3 Phase-Intensity Modulator Implementation
Figure 4.4 depicted a simplified Single Sideband (SSB) system, according to L.Duraffourg et al, Opt. Lett 26,
18 (2001) [i.3]. The source S1 is an attenuated pulsed laser diode operating at optical frequency ω (quantum signal).
An unbalanced integrated Mach-Zehnder modulator MZ1 modulates the intensity of the reference beam at Ω << ω
with a modulation depth m < 1. The modulating signal is produced by a local oscillator (OS) that drives simultaneously
a second integrated Mach-Zehnder MZ2. The light emitted by the source S2 (synchronisation signal), operating at
optical frequency ω , is then modulated at the same frequencyΩ. Both optical signals are launched in a standard fibre.
s
Their optical spectra are composed by a central peak and two sidebands ω ± Ω (ω ± Ω) with phase Φ (0) relative to
0 s 1
the central peak. At the receiver, a WDM demultiplexer allows to separate the transmitted signals. The synchronisation
signal is converted by a detector (DS) that generates an electrical signal at frequency Ω. The amplitude of the electrical
signal is matched to the modulation depth m and drives a phase modulator MZ2 with a 3λ / 4-optical path difference
bias. When a phase shift Φ is added to the electrical signal we can show that the probability P and P of detecting one
2 1 2
photon in the lower-sideband and the upper-sideband of the quantum signal is governed respectively by a sine-squared
and a cosine-squared function of the phase difference (Φ - Φ ). One of the sideband and the reference beam are
1 2
separated by optical filter F. Any protocol can in principle be implemented with this system, which features two outputs
with complementary probabilities of photon detection. The advantage of transmitting the synchronisation signal in the
same fibre link is to reduce drastically the sensitivity of the system to optical path fluctuations and thus allow long-
distance key distribution.
Figure 4.4: Schematic of a one-way, weak-laser-pulse Frequency domain QKD system
4.3 Entanglement-based QKD
Whereas many other QKD principles introduced here are asymmetric in the sense that one entity (station A, Alice)
prepares a quantum state and the second entity (station B, Bob) performs measurements to yield quantum correlations, it
is also possible to use entanglement to build up a QKD system. Thereby pairs of photons are generated in contrast to
single-photons in the other schemes. Each entangled photon-pair is distributed between Alice and Bob, who
independently measure the photon distributed and jointly form a secret key based on a series of measurements.
Two important families of entanglement-based protocols exist:
1) The first protocol was given in A. Ekert, Phys. Rev. Lett. 67, 661 (1991) [i.4]. The security could be
guaranteed by an ongoing test of Bell's inequality to root out Eve's attack. The commonly used version in
experiments as well as proposed here are the CHSH-inequalities by J. Clauser et al., Phys. Rev. Lett. 23,
880–884 (1969) [i.5].
ETSI
11 ETSI GS QKD 003 V1.1.1 (2010-12)
2) The second protocol, suggested by C. H. Bennett, G. Brassard and N. D. Mermin Phys. Rev. Lett. 68, 557
(1992) [i.6], adopts the conventional BB84 protocol. Instead of Alice preparing photons and Bob measuring, a
third party generated entangled photon pairs. For each pair, one photon is sent to Alice while the other is sent
to Bob. Alice and Bob perform their measurements, independent of each other, randomly using one of at least
two non-compatible bases. By comparing the measurement basis Alice and Bob can obtain correlated
measurement results, from which a secret key can be distilled.
4.4 Continuous-Variable QKD
Most of QKD implementations are based on single photons or weak pulses for encoding qubits on the degree of
polarization or phase. Alternativley, alternative protocols based on continuous variable (CV) can also be used for
encoding qubits. For example, the two quadratures of a coherent state can be used as conjugate variables.
4.4.1 Principle of Continuous-Variable QKD Protocols
Several CV protocols make use of light pulses to encode the key, and we will assume this condition in the following.
Their specificity is to use light pulses with a few photons per pulse instead of single-photon pulses as well as coherent
optical detection instead of photon counters. Coherent state CV protocols make use of a sequence of light pulses
described by coherent states x + ip . The two quadratures x and p are modulated according to a two dimensions
Gaussian distribution centred at ( x = 0 , p = 0 ) and with variance V N . N is the quantum noise variance that occurs
A 0 0
in the Heisenberg relation ΔxΔp ≥ N . Those coherent states are sent from the emitter, Alice, to the receiver, Bob,
through a quantum channel at the same time as an intense phase reference called local oscillator (LO). At the receiver,
signal and local oscillator interfere in a shot noise limited coherent detection. The simplest configuration makes use of
homodyne detection. Choosing the phase of the local oscillator, Bob can choose at random x or p quadrature. It is
particularly important to keep the symmetry between those to quadratures in order to prevent QND attacks (Quantum
Non Demolition) where the eavesdropper, Eve, would not introduce any noise on the quadrature to be measured and
would transfer all the noise on the quadrature that is not measured by Alice and Bob.
In CV QKD, even with a perfect detection and with no eavesdropper, Bob's measurements are always affected by the
intrinsic quantum noise that adds to each quadrature measurement. Consequently, after the quantum transmission, Alice
and Bob do not share identical quadrature measurements, but only correlated data. Thus, CV QKD requires an intensive
data treatment to extract the secret keys from those correlated data. This makes an important difference in comparison
with discrete variable QKD protocols for which Alice and Bob share identical data after reconciliation in the ideal case.
Part of Alice and Bob's data, chosen at random, is revealed publicly in order to evaluate the parameters of the
transmission channel. The remaining data is used to establish the secret key between Alice and Bob. Alice and Bob first
perform a classical error correction. For example, they can use a multilevel decoding based on efficient, one-way low
density parity check codes (LDPC). Then they proceed with privacy amplification to process a secret key common to
Alice and Bob on which Eve has no information.
4.4.2 Implementation Example of the CV QKD protocol.
As presented in detail in Fossier et al., New J. Phys. 11 045023 (2009) [i.7], Alice uses a pulsed 1 550 nm telecom laser
diode to generate coherent light pulses with a duration of 100 ns and a repetition rate of 500 kHz (see figure 4.5). The
pulses are separated into a weak signal and a strong local oscillator (LO) using a 99/1 asymmetric coupler. The signal is
then randomly modulated, using amplitude and phase modulators, following a centred Gaussian distribution in both
quadratures x and p , so that the variance of the Gaussian distribution reaches a target value of V N . It must be noted
A 0
that CV QKD is not limited to Gaussian modulation. Other modulation schemes can be considered including discrete
modulation protocols. In that case one must be very careful with the security proofs Leverrier & Grangier, Phys. Rev.
Lett. 102, 180504 (2009) [i.8].

ETSI
12 ETSI GS QKD 003 V1.1.1 (2010-12)
Figure 4.5: Scheme depicting the implementation of a coherent state CV QKD set-up
Time and polarization multiplexing are used so that the signal and LO are transmitted to Bob in the same optical fibre
without any cross-talk. First, the signal is delayed by 400 ns using a 2 × 40 m delay line, in which the pulse is reflected
by a Faraday mirror, as shown in the figure. This system imposes a π 2 polarization rotation to the pulse when it is
reflected, and thus compensates all the polarization drifts undergone by the signal. The LO is then coupled with the
signal in the transmission fibre, using a polarization beamsplitter (PBS). Thanks to this double multiplexing, the two
pulses can be separated at Bob's site very efficiently and with minimal losses, by using a simple PBS and delaying the
LO after the separation.
Finally, in Bob's system, the signal and LO interfere in a pulsed, shot-noise limited homodyne detector. This detection
system outputs an electric signal, whose intensity is proportional to the quadrature x of the signal, where ϕ is the
ϕ
phase difference between the signal and the LO. Following the implemented protocol, Bob measures randomly either
x or x to select one of the two quadratures. For this purpose, he imposes randomly a π 2 phase shift to the local
0 π 2
oscillator using a phase modulator placed in the LO path.
5 Photon Detector
5.1 Single-Photon Detector
5.1.1 Generic Description and Parameterisation
A single-photon detector is an optically-sensitive device that probabilistically transforms a single-photon into a
macroscopically detectable signal. Figure 5.1 shows a generic single photon detector with optical input, electrical input
and output.
ETSI
13 ETSI GS QKD 003 V1.1.1 (2010-12)

Figure 5.1: Schematic of a generic single-photon detector
showing electrical and optical connections
In operation the detector is used to determine the times at which the output voltage rises above the discrimination level
(detection times) and/or the number of detection events within certain time duration, from which the detection count
rate can be determined.
The performance of a single-photon detector shall be characterised by a number of parameters, as described below.
The photon detection efficiency shall be defined as the probability that a photon incident at the input to the device will
be detected. This parameter shall be defined for the external input to the device and shall not be adjusted for any losses
occurring after the optical input. The photon detection efficiency shall not be confused with the quantum efficiency of
the detection element, which describes the probability that a photon is absorbed in the active region of the detection
element.
More generally the photon detection efficiency shall be defined as a function of the wavelength of the incident photon.
The uncertainty in deteremining the photon detection time shall be referred to as the timing jitter. The timing jitter shall
be characterised as the full width half maximum (FWHM) in the distribution of detection times when the detector is
incident with a pulsed laser, the pulse duration of which is shorter than the jitter determined.
The detector may sometimes record a detection event when there is no photon incident on the device. This is commonly
referred to as a dark count. The dark count probability shall be defined as the probability that a detector registers a
detection event per gate or per unit time, when the detector is not illuminated.
Afterpulse are false counts which are secondary detection events triggered by previous photon detection events. The
afterpulse probability shall be defined as the probability that a detector registers an afterpulse event, conditional on a
true photon detection event.
Detection times of single photons shall be determined in QKD. The precision in the detection times shall be finer than
the clock period of the QKD.
The dead time of the detector shall be defined as the smallest time duration after which the detection efficiency is
independent of the previous photon detection history.
The recovery time shall be defined as the time duration after a photon detection event for the detection efficiency to
return to 90 % of its steady-state value.
The maximum count rate shall be defined as the maximum rate of photon detection events under strong illumination.
The maximum clock frequency shall be defined as the maximum clock frequency at or below which the detector can be
operated in a QKD system.
Table 5.1 lists the parameters that define the performance of a single-photon detector. These parameters shall be
specified for a defined set of operating conditions, given in table 5.2. Table 5.3 list additional attributed to be specified
for the detector.
In QKD systems that require multiple single-photon detectors for qubit detections, the detectors shall be set so as to
have balanced photon detection efficiencies. Ideally, the detection rates shall be maintained exactly the same for all the
qubit detectors. The parameters in tables 5.1, 5.2 and 5.3 shall be defined for each single photon detector in the system.
ETSI
14 ETSI GS QKD 003 V1.1.1 (2010-12)
Table 5.1: Parameters that shall be used to specify a single-photon detector
Parameter Symbol Units Definition
Photon detection probability η Unitless (probability/gate) The probability that a photon incident at the optical
input will be detected within a detection gate.
Dark count probability P Gated: Unitless The probability that a detector registers a
dark
(probability/gate) detection event per gate, despite the absence of
-1
optical illumination. For a free running detector this
Free Running: ns
may be defined as the probability that a detector
(probability/ns)
registers a detection event per ns, despite the
absence of optical illumination.
Afterpulse probability P Gated : Unitless The probability that a detector registers a false
afterpulse
(probability/gate) detection event in the absence of illumination,
-1
conditional on a true photon detection event in the
Free Running: ns
preceding detection gate.
(probability/ns)
Dead time T The smallest time duration after which the
ns/μs
dead
detection efficiency is independent of previous
photon detection history.
Recovery Time T The time duration after a photon detection event
ns/μs
rec
for the detection efficiency to return to 99 % of its
steady-state value.
Maximum count rate C MHz/GHz The maximum rate of photon detection events
max
under strong illumination condition in the
single/few photon/gate regime.
Timing jitter t ps/ns The uncertainty in determining the arrival time of a
jitter
photon at the optical input.
Photon number resolution N Unitless For detectors than can resolve the number of
photons in the incident pulse, this is the maximum
number of photons that can be distinguished.
Maximum clock frequency F MHz/GHz The maximum clock frequency at or below which a
max
detector can be operated in a QKD system without
giving rise to an intolerable bit error rate.
Spectral Responsivity R unitless The photon detection efficiency as a function of
s
wavelength of the incident photons.

Table 5.2: Operating conditions that shall be specified for a single-photon detector
Operating Condition Symbol Units Definition
Detector Temperature T °C or K Physical temperature of the detection element during
operation.
Environmental Requirement N/A N/A The environment conditions under which a detector module
operates. These conditions include environmental temperature,
humidity, pressure, and requirement for surrounding
electromagnetic radiation.
Mode of Operation N/A N/A Describes how the electrical bias is applied to the detector.
Three modes of operation are common: DC current mode, DC
voltage mode, and gated mode.
Operating Wavelength λ nm Wavelength of the photons to be detected.
Gating Frequency F MHz/GHz The frequency of the gating signal applied to the detector, if
operating in gated mode.
Gate Width W Volts For detectors operating in gated mode, this is the nominal
duration of the electrical signal applied to turn the detector on.
DC Bias Vdc Volts The dc voltage level applied to the detector.
AC Bias V Volts The peak-to-peak ac voltage level applied to the detector. The
ac
ac voltage is defined to vary between 0 and Vac. The total bias
applied to the device therefore varies between Vdc and
(V + V ).
dc ac
Discrimination level V Volts Voltage threshold above (or below) which the amplitude of an
disc
output pulse must overcome to be registered as a detection
event.
ETSI
15 ETSI GS QKD 003 V1.1.1 (2010-12)
Table 5.3: Additional attributes to be specified for a single-photon detector
Parameter Definition
Electrical input Defines electrical input signals to the device along with the type of connector used. Input
signals may be used for biasing the detector, providing a trigger signal or as a power supply.
Optical input Defines the format of the optical input to the device. Often this is through SM or MM optical
fibre. The fibre connector should also be specified, e.g. FC/PC. The device may also be
coupled through free space, in which case the active area and location within the unit should
be specified.
Electrical output Defines the format of electrical output signal from the device upon photon detection, such as
ECL, TTL, NIM, etc., as well as the type of connector, e.g. BNC, SMA.
Optical robustness The maximum illumination power that a detector can endure without altering its detection
parameters.
Physical dimensions The physical size of a detector module that is independently operational.
Power consumption Power consumption is the total power that is needed to continuously operate a detector.
Handling instructions Instructions for the safe handling of the detector, such as information regarding toxicity and
the presence of high voltages.

5.1.2 Test Measurements
This clause describes some of the measurements that shall be carried out in order to quantify the parameters defined
above. Some of the parameters are straightforward to determine or do not need to be known to high accuracy, for
example maximum count rate. Other parameters require specific measurement techniques. The most important
parameters to specify for operation in a QKD system are the photon detection probability, dark count probability,
afterpulse probability and the timing jitter. These parameters shall be determined using the test set up illustrated in
figure 5.2 for a gated detector.
A pulse generator (Pulse Gen) shall be used to generate the ac gating bias. This shall be combined with a dc bias from a
voltage source (DC) using a bias-T and is applied to the device under test. The pulse generator shall be used to trigger a
laser diode that produces light pulses with ps duration that are used to illuminate the device. The frequency of the laser
trigger shall be stepped down by a factor R (R ≥ 10) compared to the detector gating bias using a frequency divider
(Freq Div). An electrical delay line shall be used to ensure that the gate bias and optical pulse overlap temporally. The
laser input to the device under test shall be attenuated with a programmable attenuator (Att) to the single-photon level.
Time correlated single-photon counting (SPC) shall be used to record a histogram of time delays between the laser
pulse and output pulses from the device under test. The attenuated laser intensity shall be described as n photons per
illumination pulse on average.
DCDC
PulPulssee Ge Genn Bias-Bias-TT
1/R1/R F Frreqeq DDiivv
DeDelalay Liy Linene
FiFibbeerr
AtAttt
LaLaserser SPCSPC
NOTE: Pulse Gen: pulse generator that triggers the laser and gates the detector; Freq Div: frequency divider;
Att: optical attenuator; and SPC: time-correlated photon counters.

Figure 5.2: A measurement setup that shall be used to measure photon detection probability,
afterpulse probability, dark count probability and timing jitter
ETSI
16 ETSI GS QKD 003 V1.1.1 (2010-12)
Under pulsed optical excitation, the time-correlated photon counter produces the histogram shown in the top panel of
figure 5.3, in which the dominant peak at the zero-delay is due to detection of the optical excitation. The FWHM of this
peak shall be taken as a measure of the timing jitter of the detector, provided that the measured value is considerably
larger than the pulse duration of the exciting laser. The other peaks are due to dark and afterpulse counts in non-
illuminated gates. A histogram recorded without illumination shall also be measured, as shown in the bottom panel of
figure 5.3. In both cases the detected count rate shall be normalised to the total number of applied gates to form a
detection probability, and the following parameters shall be extracted:
• P : probability of a detection event for each illuminated gate.
i
• P : probability of a detection event for each non-illuminated gate under optical excitation.
n-i
• P : probability of a detection event for each gate under no optical excitation.
d
Based on these parameters, the calibrated photon flux (n), the frequency ratio (R) of the gating to the illumination or the
frequency division ratio, the following parameters can be calculated:
P − P
n−i d
• P = ⋅ R : the afterpulse probability.
afterpulse
P − P
i n−i
P − P
i d
• η = ⋅ : the photon detection probability.
n 1+ P
afterpulse
• P = P : the dark count probability.
dark d
These parameters will depend upon the clock frequency and the photon wavelength. They shall be measured for each
clock frequency and wavelength at which the detector will be used.
The precision of the measurements depends on the calibration of the intensity of the optical source. The intensity of the
optical
...