WO2021126011A1 - Educational apparatus for conducting experiments in quantum optics - Google Patents

Abstract

The invention relates to teaching aids intended for carrying out experiments, and more particularly to an educational apparatus for conducting experiments in quantum optics for the purposes of studying quantum cryptography protocols. The claimed apparatus comprises a transmitting unit and a receiving unit connected by a quantum data transfer channel, wherein the transmitting unit contains a functional programmable logic device (PLD) connected to a control PLD that is connected in turn to a first control computer. The functional PLD makes it possible to delay electrical pulses in order to ensure that they arrive at a phase modulator at the required time. The receiving unit contains a functional PLD connected to a control PLD that is connected in turn to a second control computer, wherein the functional PLD makes it possible to shorten to several nanoseconds an electrical pulse sent by a laser module; and the first and second computers make it possible to control the process of quantum key distribution in order to allow convenient and rapid setting of the desired parameters of an experiment using an established optical circuit.

Description

TRAINING UNIT FOR PERFORMING EXPERIMENTS ON

QUANTUM OPTICS

FIELD OF TECHNOLOGY

[0001] The present invention relates to teaching aids for conducting experiments, in particular to an educational setup for performing experiments in quantum optics for the purpose of learning quantum cryptography protocols.

LEVEL OF TECHNOLOGY

[0002] Currently, more and more development is underway for the development of quantum computing and related data processing processes. Quantum cryptography is one of the most promising areas in the art, but there are problems in the part that the fundamental principles of quantum mechanics are difficult to understand, and many attempts to popularize it have been either difficult to understand or incorrect. Over the past decades, significant breakthroughs have occurred in new applied aspects of quantum physics, such as secure communication networks, sensitive sensors for biomedical imaging, and fundamentally new computing paradigms. In order to accelerate the development impact of innovation, many quantum manifestos, memoranda, strategies and other roadmaps have been developed and adopted in recent years. Thus, it is important to provide a quality learning process for the next generation of technicians, engineers, scientists and application developers in the field of quantum technologies.

[0003] Quantum key distribution is a key transfer technique that uses quantum phenomena to ensure secure communication. This method allows two parties connected through an open communication channel to create a common random key, which only they know, and use it to encrypt and decrypt messages. [0004] A device for teaching quantum information transfer processes is known from the prior art (patent CN 206379044 U, 04.08.2017). A well-known educational demonstration device for quantum cryptographic communication contains a signal interceptor via an optical fiber, a receiving and transmitting units providing for sending a video signal in the form of an optical signal, and receiving on the receiving side, the interceptor unit transmitting to the receiving side of the video signal through a light splitting module. Corresponding quantum key distribution devices are provided at the transmitting side and the receiving side for distributing keys for encrypting the transmission of the video signal, and the sending side, the receiving side and the listening side are provided with displays for displaying the received video signals.

[0005] The disadvantage of this installation is its limited functionality, which consists in the lack of a universal ability to configure the receiving and transmitting modules, with which you can generate various experimental processes for the distribution of the quantum key.

DISCLOSURE OF THE INVENTION

[0006] The present invention is aimed at solving a technical problem associated with the creation of a training setup for conducting experiments in the field of quantum cryptography, which will provide a simple and convenient interaction in terms of its configuration and the formation of the required process of performing the experiment.

[0007] The technical result is to expand the functionality by supplying the transmitting and receiving units with a combination of control and functional FPGAs connected to computing computer devices, providing a convenient and quick setting of the required parameters of the experiment using the installed optical scheme.

[0008] An additional effect is to provide ease of control (programming) of the system by using a combination of control and functional FPGAs in each of the blocks.

[0009] The claimed technical result is achieved due to the design of a training setup for performing experiments in quantum optics for the purpose of studying quantum cryptography protocols, which contains a transmitting and receiving unit connected by a quantum data transmission channel, while the transmitting unit contains a Faraday mirror, a phase modulator connected through a storage line with a variable optical attenuator, a beam splitter and a synchronizing detector; the receiving unit contains a laser module connected to a circulator, which is connected to the first single photon detector (DOP), a beam splitter connected to the second DOP, a phase modulator connected through a delay line to a polarizing beam splitter; moreover the transmitting unit contains a functional FPGA connected to the control FPGA connected to the first control computing device, while the functional FPGA provides a delay of electrical impulses to ensure their arrival to the phase modulator at the required time; the receiving unit contains a functional FPGA connected to the control FPGA connected to the first control computing device, while the functional FPGA provides shortening to several nanoseconds of the electrical pulse sent by the laser module; and the first and second computing devices control the quantum key distribution process.

[0010] In one of the particular examples of the installation execution, the control FPGAs provide programming of functional FPGAs using the Lab View interface.

[OOP] In another particular example of the installation, the computing devices of the transmitting and receiving units are interconnected by means of a data transmission channel.

[0012] In another particular example of the implementation of the installation in the transmitting unit, the functional FPGA is connected to the phase modulator and the synchronizing detector through the corresponding expansion boards.

[0013] In another particular example of the implementation of the installation in the receiving unit, the functional FPGA is connected to the phase modulator and the laser module through the corresponding expansion boards.

BRIEF DESCRIPTION OF DRAWINGS

[0014] FIG. 1 illustrates the claimed system.

[0015] FIG. 2 illustrates timing diagrams of optical pulse sequences.

[0016] FIG. 3 illustrates the distribution of the number of photons for different values of m. [0017] FIG. 4 illustrates a histogram of DOP trigger statistics.

[0018] FIG. 5 illustrates a graph of the delay time.

[0019] FIG. 6 illustrates a plot of delay time intervals.

[0020] FIG. 7 illustrates a graph of DOP counting in a given window. CARRYING OUT THE INVENTION

[0021] FIG. 1 shows a functional diagram of the claimed setup (10) for performing experiments in quantum optics. The main elements of the installation are the transmitting unit (100), called Alice, and the receiving unit (200), Bob. Each of

[0022] Alice's transmitting unit (100) contains a functional FPGA (110), through which it is connected to the first control computing device (101) through the corresponding control FPGA (102). FPGA (110) provides delay of electrical impulses to ensure their arrival to Alice's phase modulator (114) at the required time.

[0023] Alice's block (100) also contains an optical circuit that includes a Faraday mirror (115), a phase modulator (114) connected through a storage line (117) with a variable optical attenuator (PO A, 116), a beam splitter (118) and synchronizing detector (113).

[0024] The receiving unit (200) Bob contains a functional FPGA (210), through which it is connected to the second control computing device (201) through the corresponding control FPGA (202). Bob's FPGA (210) shortens the electrical pulse sent by the laser module (213) to a few nanoseconds.

[0025] Bob's optical scheme (200) contains a laser module (213) connected to a circulator (214), which is connected to a first single photon detector (DOP, 216), a beam splitter (219), connected to a second DOP (217), phase a modulator (215) connected through a delay line (218) to a polarizing beam splitter (220).

[0026] Computing devices (101, 102) provide control over the process of quantum key distribution and are performed, as a rule, in the form of computing devices, for example, personal computers, laptops, etc.

[0027] Blocks of Alice (100) and Bob (200) are interconnected through a quantum data channel (20) formed by means of optical fiber.

In this case, the computing devices (101, 102) are connected to each other via an Ethernet data transmission channel.

[0028] Preparation of quantum states is carried out on the side of Alice's block (100), which performs a phase shift f _{A of the} pulses by the value: 0, p / 2, p, 3p / 2 in a random order. In this case, Alice's block (100) associates the phase shifts 0 and p / 2, the logical unit p and 3p / 2, with the value of a logical zero. Bob's Block (200) measures the quantum states of the arrived photons, performing a random phase shift f _B by 0 and p / 2.

[0029] Direct generation of multiphoton optical pulses and detection of attenuated optical pulses (measurement) is performed by Bob (200), phase encoding of optical pulses and their attenuation to a single-photon state (preparation) is performed by Alice (100).

[0030] The operation of the installation (10) is carried out during the interaction of the blocks of Alice (100) and Bob (200), in the following sequence.

[0031] The laser (213) in the Bob unit (200) generates a sequence of n linearly polarized multiphoton optical pulses - train, with a laser pulse repetition rate no (pulse repetition period in the train, To - cycle). The period of the trains' repetition T _{t is} calculated in such a way that all n pulses of the train have time to pass the distance along the optical channel from "Bob" (200) to "Alice" (100) and back. In total, a package of N trains is formed during the session.

[0032] FIG. 2 shows the timing diagrams of sequences of optical pulses at the output of the laser (213) and polarization beam splitter (220) of the Bob unit (200). The generated train of laser pulses is fed to a circulator (214), which directs laser radiation to a beam splitter (219), which divides each of the pulses in half in intensity and directs them to the long and short arms of the interferometer.

[0033] In the long arm of the interferometer there is a delay line (DL, 218) of a predetermined length, for example, 10 m, respectively, the train of laser pulses hitting this arm is delayed with respect to the train in the short arm of the interferometer by the time T _DL . Thus, at the output of the polarization beam splitter (200), a doubled sequence of laser pulses is obtained as compared to the initial train (Fig. 2), while the pulses that have overcome the long arm of the interferometer have a lower intensity due to losses arising in the delay line (218 ) and a phase modulator (215), thus the pulses in the train form pairs of linearly polarized mutually orthogonal pulses.

[0034] Next, the train of pulses leaves the Bob block (200) and through the optical quantum channel (30) is sent to the Alice block (100). At this stage, the phase modulator (215) in the long arm of the interferometer is not activated.

[0035] In Alice's block (100), the train of laser pulses is divided by a beam splitter (118) in intensity in a 1: 1 ratio, part of the optical power is supplied to synchrodetector (113), which fixes the moment of arrival of the train of pulses for the functional FPGA (PO), which calculates, according to the signal of the synchrodetector (113), the moment of issuing the control signal to the driver of the phase modulator (111) of Alice's block (100) The train is used further to form the sequence and is fed to the variable optical attenuator (PO A, 116) and then to the storage line (SL, 117), designed to exclude the influence of parasitic reflections from the elements of the optical circuits of Alice's unit (100) and Bob's unit (200) on Bob's DOP (216, 217).

[0036] After the accumulation line (117), the train of pulses is fed to the phase modulator (114), the driver of which (111), in accordance with the calculated FPGA (PO) time delay, issues a command about the phase shift of the first pulse of each pair. The phase shift value 0, p / 2, p, 3p / 2 is determined by the FPGA (software) randomly. After the phase modulator (114), the pulse is reflected from the Faraday mirror (115) (in this case, it changes its polarization to orthogonal) and again enters the phase modulator (114). The data on the applied phase shifts to the train of the FPGA pulses (110) are sent by Alice's block through the control FPGA (102) to Alice's computing device (101) for further processing.

[0037] Next, the train of optical pulses sequentially enters the storage line (SL, 117) and to the PAO (116), which again attenuates the train of pulses before entering the quantum channel (30), now to a single-photon state.

[0038] After passing through the quantum channel (30), the train of pulses arrives at the polarization beam splitter (220) of the Bob unit. Since the pulses in the train, being mutually pairwise orthogonally polarized, changed their polarization to the opposite when reflected from the Faraday mirror (115), now the polarizing beam splitter (220) directs the pulse to the long arm, if it passed through the short arm before, and vice versa, in the short, if before that he passed the long. Passing along the long arm of the interferometer, optical pulses successively enter the delay line (SL, 218) and the phase modulator (215). The functional FPGA of Bob's block (210) issues commands to the phase modulator driver (212) about the phase shift by 0 or p / 2 values for measurements.

[0039] Trains of optical pulses passing along the long and short arms interfere in the beam splitter (219) and are directed to the DOP inputs (216) through the circulator (214), which count the pulses (trigger) depending on the value of the phase shifts on the phase modulators Alice's block (114) and Bob's block (215).

[0040] If the phase difference of the optical pulses is 0 or p, then we can talk about the compatibility of the bases "Alice" and "Bob" in the preparation and measurement of single photons, and the measurement results will be determined. In the case when the phase difference of the pulses will be k / 2, or 3k / 2, we can talk about the incompatibility of the bases of "Alice" and "Bob", while the DOF (216, 217) will be triggered randomly. So, for example, if both phase modulators of Alice (114) and Bob (215) blocks do not apply a shift, then constructive interference is realized on DOP (217) (it works), and on DOP (216) - destructive (does not work).

[0041] The information transfer procedure can be schematically summarized in Table 1.

[0042] Данные о срабатываниях ДОФ (216, 217) и значения сдвигов фаз импульсов трейнов функциональной ПЛИС Боба (210) выдаёт через управляющую ПЛИС (202) в вычислительное устройство блока Боба (201) для дальнейшей обработки. Table 1

[0042] The data on the DOF triggers (216, 217) and the values of the phase shifts of the pulses of the trains of the functional FPGA of Bob (210) are sent through the control FPGA (202) to the computing device of the Bob block (201) for further processing.

[0043] Thus, on the sides of the blocks "Alice" (100) and "Bob" (200) is formed

Raw sequence. In those cases when the Alice (100) and Bob (200) blocks used the same basis for preparing and measuring optical pulses, they should receive the same data bit values (zeros and ones).

[0044] Formed as a result of interaction of Alice (100) and Bob (200) blocks

The "raw" sequence is then subjected to a "sieving" procedure. The computer of the Bob block (201) sends the computer of the Alice block (101) via an open information channel (for example, Ethernet) a data array with the selected measurement bases.

[0045] Based on the received data array, the computer of the Alice block (101) determines for which pulses the measurement basis selected by the Bob (200) block was correct (coincided with the cooking basis), and generates a data table containing the numbers of pulses within the packet and the values of the transmitted bits , which is, in fact, the final sequence. To the computer of the Bob block (201), the computer of the Alice block (101) reports only the numbers of the packet pulses, according to which it forms the final sequence on its side.

[0046] The scheme of operation of the installation (10) is based on a two-pass automatic compensation scheme Plug & Play, the principle of operation of which was disclosed by the authors earlier in patent RU 2671620 C1 (MCKT LLC, 02.11.2018). The application of the principles of constructing this optical scheme is the most convenient from the point of view of the simplicity of setting the parameters for the distribution of the quantum key.

[0047] Next, consider the sequence of events that occur with optical pulses during quantum key distribution in the Plug & Play scheme. The laser module (213), located in the Bob unit (200), generates a sequence of identical multiphoton optical pulses, called a train. The wavelength of optical pulses is, as a rule, l = 1550 nm, the pulses are linearly polarized. The laser pulse repetition rate is one of the main characteristics of the quantum key distribution. Let's designate this value as n o, the corresponding period - To, these time intervals are ticks. In a train, n impulses and trains follow with a period Tt - the period of the train (Fig. 2). In total, a package of N trains is formed during the session. The interval between the trains is such that the train manages to travel the distance from Bob (200) to Alice (100) and back, and only after that the next one starts.

где -длина линии задержки hq ~1.47, D1 - коэффициент преломления, со - скорость света.[0048] The Bob block (200) is assembled from a polarization-retaining fiber (PM fiber). The train of laser pulses hits the circulator (214), which directs the laser radiation to the beam splitter (219). As a result of the passage of the beam splitter (219), each pulse is split in half and directed to the corresponding arms of the interferometer. [0049] In one of the arms of the interferometer there is a delay line (218) and a phase modulator (215), therefore, the pulses passed through this arm (220) arrive later, therefore, at the exit from the polarization beam splitter (220) it is obtained in 2 times more impulses than in the original train. The lag time is described by the formula:

where is the length of the delay line hq ~ 1.47, D1 is the refractive index, and ω is the speed of light.

In practice, it is convenient to use the relation arising from this formula: in 5 ns, a light pulse passes 1 m of optical fiber, i.e. it turns out that the speed of light in an optical fiber is approximately equal to 2x10 ⁸ m / s. Due to the described bias, the pulses are conveniently viewed in pairs. The pulses that have crossed the larger shoulder have a lower intensity due to losses arising in the delay line and phase modulator. Note that the pulses emanating from the Bob's polarization beam splitter (220) are linearly polarized, and their polarization is pairwise orthogonal. Then the train leaves Bob (200) and goes through the quantum channel (30) to Alice (100). The phase modulator (215) in the long arm of Bob's interferometer (200) is not activated when the pulses pass towards Alice (100).

[0050] Arriving at Alice's optical circuit (100), the laser pulses are divided into two unequal parts: most of the optical power goes to the synchronizing detector (113), and a smaller part is used further for processing. The entire optical scheme of Alice (100), like the quantum channel (30), consists of a single-mode fiber that does not support polarization.

здесь LSL длина накопительной линии. При расчете количества импульсов в трейне необходимо округлять значения в меньшую сторону. Период трейна ограничен снизу длиной квантового канала + накопительная линия. Это ограничение накладывается по той же причине: не должно быть наложения импульсов идущих туда и обратно. Т.е. необходимо дождаться последнего импульса в трейне и только потом посылать новый. При расчете периода трейна необходимо округлять значения в большую сторону. [0051] The sync detector (113) serves to synchronize all fast processes, in particular phase modulation. The sync detector (113) indicates the time of arrival of the laser pulses on the optical circuit of Alice (100), which ensures the timely supply of electrical signals to the phase modulator (114) of Alice with an appropriate delay. [0052] Before Alice's phase modulator (114), the pulses still have to pass the accumulator line (117). Accumulative line (117) serves to accommodate all train impulses. The length of the accumulative line (117) determines the number of impulses in the train. When the train moves from Bob (200) to Alice (100), multiple reflections inevitably occur on optical inhomogeneities, mainly on optical connectors. These reflections go back to Bob (200), and can cause false alarms of photon detectors (216, 217), but only the signal obtained when the train is reflected from the Faraday mirror (115) is informative. Therefore, to protect against false alarms, it is necessary to separate in time the parasitic reflections and the reflection from the Faraday mirror (115), while there should be no overlap, for example, of reflections from the tail impulses of the train and the already returning head impulses of the train. For this, the accumulative line (117) is used, and the train can be up to 2 times longer than the accumulative line:

here LSL is the accumulation line length. When calculating the number of impulses in a train, it is necessary to round the values down. The period of the train is limited from below by the length of the quantum channel + storage line. This limitation is imposed on that the same reason: there should be no overlap of impulses going back and forth. Those. it is necessary to wait for the last impulse in the train and only then send a new one. When calculating the train period, it is necessary to round up the values.

[0053] Between the storage line (117) and the Faraday mirror (115), a phase shift is applied to one of the pulses of each pair by Alice's phase modulator (114), then the pulses reflected by the Faraday mirror (115) are returned back to the storage line (117). Reflecting from the Faraday mirror (115), the laser pulse changes its polarization to orthogonal.

[0054] It should be noted that the phase shift needs to be superimposed on only one pulse in a pair, but this shift must be superimposed on both the round trip path. This is dictated by the peculiarities of the operation of the phase modulator (114), since the phase modulator (114) shifts only one polarization component, and the polarization state of laser pulses can be unpredictable, since they passed through the quantum channel (30), underwent various polarization distortions, and it is not known what state of polarization falls into the phase modulator of Alice's block (100). [0055] After reflection from the Faraday mirror (115), the polarization changes to orthogonal, and if the phase modulator (114) has shifted one polarization component in its basis, then on the way back it will shift the other component. These circumstances dictate the following limitation: between the input of the phase modulator (114) and the Faraday mirror (115), there should be only one pulse from a pair. Thus, it turns out that the time the pulse travels the distance from the input of the phase modulator (114) and back should not exceed the time between optical pulses in a pair. If, for example, the repetition period of laser pulses T0 = 100 ns, and in this case, after passing the larger arm of the Bob interferometer (200), the second pulse is shifted by half a period, then the length of the optical fiber between the phase modulator and the Faraday mirror (115) should not exceed 5 meters.

[0056] From Alice (100), the train of pulses is sent through the quantum channel (30) to Bob

(200). In this case, the polarization transformations that the train experienced on the way to Alice are compensated. This is the advantage of the Plug & Play circuit - there is no need to constantly tune the optical circuit to compensate for distortion. But this scheme also has drawbacks, for example, due to the need to transmit by trains, the key distribution speed significantly decreases.

[0057] Since the pulses in the train, being mutually pairwise orthogonally polarized, changed their polarization to the opposite when reflected from the Faraday mirror (115), now they each pass along a different arm of the interferometer Bob (200). A polarizing beam splitter (220) directs a pulse to the long arm if it has passed the short arm before, and vice versa. If both phase modulators, Alice and Bob, do not apply a shift, then constructive interference is realized at DOP (217), and destructive interference at DOP (216). Two bases ensure the secrecy of key transmission. Alice uses a phase modulator (215) to apply four phase shifts: 0, p / 2, p and 3 p / 2. It turns out that the phase shifts 0 and p correspond to the same basis, for example +, and the shifts π / 2 and 3 s / 2 correspond to the basis X. Bob decides in which basis to measure, either without shifting the phase (basis +), or shifting the phase by p / 2, then this will be a measurement in basis X. Both Bob and Alice choose the basis of measurement and the basis of preparation quite randomly.

где п_{р -} число фотонов в импульсе. [0058] Quantum key distribution based on the BB84 protocol requires the optical pulses to be single-photon. If the pulses contain more than one photon, then the eavesdropping unit (Eve) can carry out an attack by splitting the number of photons and the key will be compromised. Due to the complexity of creating an effective source of single photons, in real conditions, quantum states are not single-photon. In practice, weakened laser pulses are used instead of single-photon states; coherent states with an average number of photons m = 0.1-0.5 photons / pulse. At low intensity, the distribution of the number of photons in a pulse obeys the statistics

where n _{p is the} number of photons in the pulse.

[0059] FIG. 3 shows the distribution of the number of photons for different values of m. The figure shows that if we have an average of 1 photon per pulse, then about 37% of our pulses do not contain a single photon, the same amount of one photon each, almost 20% - 2 photons, more than 5% - 3 photons, and even for four photon pulses, the contribution is very noticeable. This state of affairs is completely unacceptable, since there are too many pulses to carry out a photon splitting attack. If we use pulses with m = 0.1, then, as can be seen from the graph, the number of pulses containing 2 photons and more is negligible.

[0060] Laser pulses of the required power are obtained using a polarizing optical attenuator (POA, 116), which attenuates the train two times: on the way back and forth. In order to set the necessary attenuation on the attenuator, for example, such that pulses come out from Alice on average, say, 0.1 photon / pulse, it is necessary to use a power meter to determine the power of laser pulses leaving Alice's beam splitter (118). But this measurement is complicated by the fact that the pulses are initially weak, and even more attenuated pulses are needed, and all this goes beyond the measurements of the power meter. Therefore, it is necessary to measure more powerful pulses, and then additionally reduce the power by the required number of times.

где Pi_{n -} мощность излучения, входящего в оптическую схему, P_{out -} мощность излучения на выходе. [0061] You can consider the following sequence of actions. It is imperative to turn off both single photon detectors (216, 217) so as not to damage them. For reliability, it is best to disconnect the optical connectors. Using a continuous laser light source, measure the attenuation coefficients in the optical schemes of Bob (200) and Alice (100). To calculate the attenuation coefficient in the optical scheme, the formula is used:

where Pi _{n is the} power of the radiation entering the optical scheme, P _{out is the} power of the radiation at the output.

[0062] Because there are multiple optical connections between the beam splitter and Alice's attenuator, some of the power is reflected from these connectors. Since the output train is very weak after the pulses are attenuated at PO A (116), the contribution of the reflected signal becomes significant, and when calculating the attenuation coefficient necessary to achieve the required number of photons per pulse, one will have to take into account the reflection coefficient in this section. To determine the reflection coefficient, a large attenuation value (60 dB) is set at the PHA (116) so that nothing comes out, and the power of the reflected signal Pr is measured. Then the minimum attenuation is set at the PHA (116) and the output power P _out is also measured. Let the coming to

, где множитель 2 учитывает проход туда-обратно, а_с - неизменное ослабление канала, a_n - некоторое значение аттенюации. VOA power P _m , then the output

, where the multiplier 2 takes into account the round trip, and _c is the constant attenuation of the channel, and _n is some value of the attenuation.

где l - длина волны лазерного излучения, равная 1550 нм. [0064] Для быстрого построения различных оптических экспериментов, предложенная установка (10), в которой используется две функциональные ПЛИС (110, 210) в каждом из блоков Алисы (100) и Боба (200). ПЛИС (110, 210) выполняются на отдельно разработанной материнской плате, осуществляющий маршрутизацию сигналов на плате и с заложенным функционалом по укорачиванию электрических сигналов (например, для лазера) и задержки сигналов на сотни микросекунд с точностью до наносекунд. Управляющие сигналы, формируются, собираются и передаются с помощь управляющих ПЛИС (102, 202) на вычислительные компьютерные устройства (101, 201), Управляющие ПЛИС (102, 202) могут выполняться в виде плат NI (National Instruments). [0063] The power at which the pulse contains the required average number of photons m, is calculated by the formula:

where l is the laser wavelength equal to 1550 nm. [0064] To quickly build various optical experiments, the proposed setup (10), which uses two functional FPGAs (110, 210) in each of the blocks Alice (100) and Bob (200). FPGAs (110, 210) are executed on a separately developed motherboard, which carries out signal routing on the board and with the built-in functionality for shortening electrical signals (for example, for a laser) and delaying signals by hundreds of microseconds with an accuracy of nanoseconds. Control signals are generated, collected and transmitted with the help of control FPGAs (102, 202) to computing devices (101, 201). Control FPGAs (102, 202) can be executed in the form of NI boards (National Instruments).

[0065] Functional FPGAs on the motherboards of Alice (110) and Bob (210) have a USB-U ART serial interface for exchanging data with external devices, a USB-JTAG converter for FPGA programming, FLASH memory for storing values of tunable parameters, connectors VHDCI for connecting an NI board such as PCIe-7841R. On the motherboard there is a short circuit (short circuit) protection unit for power supply.

[0066] The form factor of the motherboard is designed in such a way that it can be installed in a standard computer case, the footprints correspond to the standard ATX format. To control opto-electrical components and to collect data from external devices, the motherboard has the required number of SMA connectors (for example, 10) and a number of universal connectors of the PC1e x4 format (for example - 12). Expansion cards such as laser driver, DAC, etc. can be installed in the universal connectors. A number of these connectors are designed for fast processes (for example, supply voltage to a phase modulator) and contain both parallel and serial interfaces. A number of connectors - for connecting expansion cards with a serial interface and digital inputs. Used FPGAs (software, 210) as part of motherboards have sufficient potential to implement all the functionality.

[0067] As expansion cards in the present installation (10), a laser module driver (211), phase modulator drivers (111, 212) and a sync detector driver (112) are used. The laser module (213) generates laser pulses with a length corresponding to the length of the input signal, the current through the laser diode is stabilized. The laser module (213) contains a Peltier element, so that the temperature of the laser diode of the module (213), and as a consequence, the wavelength is also stabilized. The phase modulator driver (111, 212) is a 10-channel DAC designed to supply voltage to the corresponding electro-optical components. Driver synchrodetector (112) is used to register high-power (multiphoton) laser pulses that perform an auxiliary function of synchronizing processes in the optical scheme.

[0068] The main problem of manufacturing and tuning a quantum cryptography device is the synchronization of all optical and electrical pulses in the device. As mentioned above, the design of the claimed setup (10) is based on the control FPGAs (102, 202) of National Instruments boards programmed in Lab VIEW. Control FPGAs (102, 202) are connected to each functional FPGA (software, 210) of motherboards using a special cable with VHDCI connectors. The FPGA (110) of the motherboard in Alice's block (100) monitors the supply current of the expansion boards, delays electrical impulses to ensure that they arrive at the phase modulator (114) at the right time, and in Bob's block (200) FPGA (210) shortens the electrical pulse sent to the laser driver to a few nanoseconds (211). [0069] The developed software (software) for control of the educational installation (10) of the research complex for quantum key distribution represents several software parts-projects. Alice (100) and Bob (200) blocks are involved in key generation, respectively. There are two software projects for each device. One project is the firmware of functional FPGAs (110, 210). Another project is a process automation and data collection system developed in the Lab VIEW environment and using a PC1e R-series board from National Instruments to implement control FPGAs (102, 202).

[0070] The LabVIEW project can be divided into 2 parts. The first part is software for execution on an R-series board, for example, PC1e 7841R - firmware for the FPGA board. The second part is the host-level program, organizes the user interface, provides TCP / IP connection of Alice (100) and Bob (200) blocks, accesses the memory of the NI (102, 202) control FPGA boards, sifts the quantum key.

[0071] The firmware of Bob's NI board (202) provides the following functionality: the formation of laser pulses, the supply of a digital signal to the phase modulator (215) (this organizes the control signals of the DAC of the phase modulator - CLK and WRT), the choice of the phase shift value - random or user-defined polling of single photon detectors. When executing the program logic, the time (cycle number of the corresponding cycle) of triggering the detectors of single photons is entered into the FIFO memory. The board (202) stores the value of the phase shift in the interferometric circuit and the fact of triggering / non-triggering of single photon detectors into memory. [0072] For the firmware of the control board NI Alice (102), the corresponding software is used, which provides digital and control signals to the DAC of the phase modulator (114) and stores the supplied value in the FIFO memory. In addition, Alice's firmware transmits to the motherboard FPGA (110) the value of the delay time for signaling the phase modulator (114). The delay time value is transmitted via the SPI interface, and Alice's firmware generates all the necessary control signals.

[0073] Quantum key distribution requires precise synchronization of optical and electrical pulses throughout the Plug & Play scheme. This leads to the need to pre-configure the entire circuit. Also, after a while, it may be necessary to adjust the tuning parameters, for example, due to a change in the length of the fiber-optic channels with temperature fluctuations.

[0074] Setting the time of the expected arrival of pulses to Bob's detectors. As already mentioned, when optical pulses move from Bob's (200) blocks to Alice (100), multiple reflections arise, which trigger the DOP (216, 217) of Bob (200). It is preferable to register DOP (216, 217) triggers only in those time windows when the arrival of pulses from Alice's block (100) is expected. The Host (Channel Test) .vi program is located in the Lab View project on Bob's computer (201) and is used to determine the cycle number in which the pulse returns to Bob (200). Clock numbers are counted from the moment the train is sent to Alice (100).

[0075] Channel Test works as follows. Single-pulse trains leave the Bob (200) block; after passing through the optical circuits of Bob (200) and Alice (100) blocks, they are registered by DOP (216, 217). The statistics of the DOP triggering (216, 217) is collected by the numbers of ticks and a histogram is plotted using the data obtained (Fig. 4), where the numbers of ticks are plotted along the x-axis, and the number of detector triggers is plotted along the y-axis. The last pulse in the histogram corresponds to the reflection from the outermost element in the optical circuit - from the Faraday mirror (115).

[0076] In the software interfaces of computer devices (101, 201), the found value will correspond to the opening of a large window for DOP (216, 217). For the case shown in the histogram of Fig. 4, the required number will be equal to 1818. The pulses in the histogram can double, this is because the registration of the counts of the single photon detectors occurs in smaller time windows than the histogram scale displays, so that the signal can be recorded in 2 different windows. Here, for a given measurement accuracy, both paired pulses correspond to one laser pulse. To confirm the correctness of the chosen reflex, you can carry out the following actions: change the voltage on the phase detector (DAC Level), change the number of impulses in the train, while the number of reflexes should increase, you can make sure that the cycle number is determined correctly by changing the window to be torn off. The resulting measure number should be set in all subsequent subroutines in the Open Window field. The time window closes after the arrival of the last impulse in the train, i.e. in the Close Window field, enter the number obtained by summing the number of impulses in the train with Open Window.

[0077] Adjusting the voltage across the phase modulators Bob (215) and Alice (114).

[0078] Bob's Host subroutine (PM Level Test) .vi is in the firmware of Bob's functional FPGA (210) and is used to find the voltage value that must be applied to the phase modulator (215) to shift the optical pulse by phase p. When the program starts, a series of trains is sent (it is better to send several tens of pulses per train), while the voltage supplied to the phase modulator (215) each time changes by a small amount (DAC interval). If at some DOF-e (216, 217), in the absence of voltage across the modulators, an interference minimum was realized, then by gradually increasing the voltage at one of the modulators, a maximum should be realized.

[0079] The procedure for adjusting the voltage corresponding to the phase shift by p is carried out for the phase modulators of Alice (114) and Bob (215) separately. For this, different, but similar in terms of interface, subroutines are used. After that, in the corresponding fields of the Alice and Bob programs for the distribution of the key, it will be necessary to set the obtained values. This voltage value in arbitrary units is called in other programs DAC Level, DAC - Digital-to-Analog Converter.

[0080] Setting the amount of delay in applying voltage to Alice's phase modulator (114).

[0081] The time of arrival of the optical pulses is recorded by Alice's sync detector (113). But before the pulse is shifted on the phase modulator (114), it still has to pass the storage line (117), spending some time on this. To find the exact value of the time pause, which must be maintained before applying voltage to the phase modulator (114), it is necessary to run the Host (Alice Delay Line Test) programs of the same name on the Alice (100) and Bob (200) blocks simultaneously. The interface of these programs is different: Bob's program (200) sends laser pulses and sends Alice (100) information about the DOP triggering. Alice's block (100) changes the delay time (SPI Interval) with a certain step and builds a graph (Fig. 5). [0082] The appearance of this graph explains FIG. 6. As long as the delay is insufficient (delay tl), the electrical pulses supplied to Alice's phase modulator (114) are ahead of the optical pulses and a maximum is realized at the DOP (217), as it should be in the absence of modulation. As the delay increases, the last electrical impulse covers the first optical one, so we see the first minimum on the graph. Then we continue to increase the delay time, passing through a series of minima, until the entire electrical sequence of pulses is aligned with the optical one. This will correspond to the desired delay, in the graph of FIG. 6 about 24648. If you use another DOF (216) of Bob for measurement, then the graph will be inverted - the minimums and maximums will swap places.

[0083] To search for the value of the delay time, you first need to estimate what the value of the delay time will be, based on the length of the quantum channel (30). Then you need to select the number of impulses in the train for the initial search for this value. It is reasonable to choose such a quantity that the search result is stretched by 1000 - 2000 units along the graph scale. According to these estimates, set the start and end points of the search - DL SPI (min) and DL SPI (check). Here DL is the delay line (218), and the control of the delay line is carried out via the SPI interface by Alice's computer (101). Next, you need to select the search step - SPI Interval. The unit search step corresponds to the increment of the delay value by the division value. It is recommended to choose a sufficiently large step, but not a multiple of the pulse repetition period. Thus, it is necessary to repeat the procedure several times, each time decreasing the step, the number of pulses, the scanning interval, until the exact value of the delay is found. The last measurement is made for 3-5 impulses per train.

[0084] Fine tuning the DOP time windows (216, 217).

[0085] As mentioned above, the time window per clock is a rather rough value. In order to reduce the dark count, which makes the main contribution to the QBER (the magnitude of the error), it is reasonable to reduce the time window for recording the arriving signals for each DOF (216, 217). In the developed software for each DFO (216, 217), a time window with a width of 10 ns is individually configured. To perform this setting, multi-pulse trains are sent and statistics are collected for each of the small windows using Bob's (200) Host (Detector Window Test) .vi program (Fig. 7). For a pulse repetition period of 200 ns, there are 40 windows ranging from 0 to 39. FIG. 7 that the maximum DOP count (216) corresponds to the 7th window, and it is not much more than the count in the 6th window. The value of the tuning parameter set in Bob's program for the quantum key distribution is 7. [0086] It should also be noted that to find the window number for D1, it is necessary to set the offset voltage corresponding to the offset p on Bob's phase modulator (215). To find the window number for DOF (217), on the contrary, you need to make sure that no voltage is applied to Bob's phase modulator (215).

[0087] Setting the timing of the signal feed to Alice's phase modulator (114).

[0088] As previously described, the phase shift needs to be applied to only one pulse in a pair, but this shift must be superimposed on the path to and from the Faraday mirror (115). To set the voltage on and off times on the phase modulator (114), it is necessary to vary the parameters in the “PM Data on” and “PM Data off” fields, which indicate the number of clock cycles counted from the moment the synchronizing signal arrives. To control these parameters, it is necessary to connect the oscilloscope probes to the CLK data pins of the DAC and, by changing the parameters in the “PM Data on” and “PM Data off” fields, achieve the optimal duration of the pulse supplied to the phase modulator (114).

[0089] The ability to generate a quantum key using one single photon detector.

[0090] The BB84 protocol allows quantum key distribution using only one DOP (216 or 217). To do this, Bob's block (200) needs to randomly select not only the measurement basis, but what he is going to check in this basis - "0" or "1". If he guessed correctly with both the basis and the bit, then the detector will click. If you have not guessed correctly with the basis, then the detector will click with a probability of 50%. If Bob guessed correctly with the basis, but did not guess with the bit, then the detector will not click, but such cases will still be thrown out.

[0091] In the proposed implementation of the "Plug and play" scheme, Bob's phase modulator (215) is randomly fed not two voltage values, as in the case of working with one DOF, and also, like Alice, four values corresponding to phase shifts 0, p / 2, s and 3 1/2. For those events in which Bob recorded a click, he tells Alice via an open channel the basis of the measurement: basis I is shifts 0 or p, basis II is shifts p / 2 or 3 p / 2. Bob keeps the information which bit was checked in the given basis secret. After that, Alice and Bob keep the bits in the coincident bases for themselves and thus get the sifted key. When distributing a key using one DOF, the key is at least 2 times shorter than when using 2 DOFs. [0092] The tuning procedures for working with one DOF do not fundamentally differ from those for two DOFs. The fundamental difference is that fine tuning of the time window is carried out only for one DOF, for example, (216). For start key generation, both Alice and Bob put the host program into the mode of operation with one DOF.

[0093] Start key generation.

[0094] To start key generation, you need to use Bob and Alice's programs Bob.vi and Alice.vi using the software interface of computer devices (101, 102). For a successful launch, you must enter the previously found parameters in the corresponding fields. If all parameters were found correctly, the key transfer will start. The program will display the key length and QBER. By shifting the DOP windows (216, 217), the maximum key length can be achieved, and by changing the delay time, the QBER value can be reduced.

[0095] Thus, the claimed setup (10) allows a wide range of quantum optics experiments to be carried out for the purpose of studying quantum cryptography protocols.

[0096] The presented description of the claimed solution discloses only preferred examples of its implementation and should not be construed as limiting other, particular examples of its implementation, not going beyond the scope of legal protection, which are obvious to a specialist in the relevant field of technology.

Claims

FORMULA 1. An educational setup for performing experiments in quantum optics for the purpose of studying quantum cryptography protocols, containing a transmitting and receiving unit connected by a quantum data transmission channel, while the transmitting unit contains a Faraday mirror, a phase modulator connected through a storage line with a variable optical attenuator, a beam splitter and a sync detector; the receiving unit contains a laser module connected to a circulator, which is connected to the first single photon detector (DOP), a beam splitter connected to the second DOP, a phase modulator connected through a delay line to a polarizing beam splitter; moreover, the transmitting unit contains a functional FPGA connected to the control FPGA connected to the first control computing device, while the functional FPGA provides a delay of electrical pulses to ensure their arrival to the phase modulator at the required time; the receiving unit contains a functional FPGA connected to the control FPGA connected to the first control computing device, while the functional FPGA provides shortening to several nanoseconds of the electrical pulse sent by the laser module; and the first and second computing devices control the quantum key distribution process. 2. Installation according to claim 1, characterized in that the control FPGAs provide programming of functional FPGAs using the LabView interface. 3. Installation according to i.2, characterized in that the computing devices of the transmitting and receiving units are interconnected by means of an information data transmission channel. 4. Installation according to claim 1, characterized in that the functional FPGA in the transmitting unit is connected to the phase modulator and the synchronizing detector through the corresponding expansion boards. 5. Installation according to claim 1, characterized in that the functional FPGA in the receiving unit is connected to the phase modulator and the laser module through the corresponding expansion boards.