A quantum random number generation system and method based on light quantum walk

Abstract

The application discloses a quantum random number generation system and method based on light quantum walk, which comprises a pump light source, a nonlinear crystal, a single photon performance measurement module, a fiber transmission module, a polarization state forming module and a quantum walk network.

Description

Technical Field

[0001] This invention relates to optical quantum information processing technology, and more particularly to a quantum random number generation system and method based on optical quantum walks. Background Technology

[0002] Random numbers are an important resource, required to varying degrees in applications ranging from cryptography and lotteries to simulations and computer applications. When classical algorithm generators produce random number sequences, these sequences are completely deterministic and exhibit a huge, finite period, thus being considered pseudo-random numbers. In contrast, quantum random number generators utilize the wave nature and inherent randomness of quantum physics systems to generate random numbers, guaranteed by the superposition principle and measurement principle of quantum mechanics, resulting in truly random numbers. There are various methods for generating quantum random numbers, including: quantum random number generators based on radioactive decay, noise-based quantum random number generators, and optical quantum random number generators. Optical quantum random number generators further include those based on laser phase noise, photon arrival time, and path branching. Among these, path branching quantum random number generators are widely used due to their simple construction, stability, reliability, and ease of implementation. However, typical path-branching quantum random number generators based on multiple beam splitters typically have multiple built-in Mach-Zehnder interferometers. The uncertainty in the beam splitting ratio of the beam splitters and the variation in the optical length between the two arms of the Mach-Zehnder interferometer with the external environment both lead to high interferometer noise and low fidelity, making it difficult to generate quantum random number sequences that meet practical requirements. Furthermore, quantum random number generators need to consider various probability distributions, including the uniform distribution commonly used in quantum key distribution, and the Gaussian distribution required for Monte Carlo and noise simulation applications. Solving these technical problems is key to further expanding the applications of quantum random number generators. Summary of the Invention

[0003] Purpose of the invention: This invention addresses the problems existing in the prior art by providing a quantum random number generation system and method based on optical quantum walks that can be distributed on demand and has high fidelity and low noise.

[0004] Technical solution: The quantum random number generation system based on photonic quantum walks described in this invention includes:

[0005] Pump light source, used to emit pump light;

[0006] Nonlinear crystals are used to generate a pair of parametric photons with orthogonal polarization from incident pump light, namely a signal photon and an idler photon;

[0007] The single-photon performance measurement module is used to measure the single-photon performance of the system, extract the signal photon delay as a prediction, and transmit idler photons;

[0008] The fiber optic transmission module is used to transmit idler photons through optical fibers, thereby achieving beam shaping and compensating for the phase generated by the idler photons in the optical fiber.

[0009] The polarization state forming module is used to prepare phase-compensated idler photons into idler photons with the desired polarization state.

[0010] A quantum walk network has multiple output terminals, which are used to output idler photons of the desired polarization state from each output terminal according to the target probability distribution;

[0011] A single-photon detector is connected to each output terminal of the quantum walk network. It is used to perform a coincidence measurement between the signal photon extracted by the single-photon performance measurement module and the idler photon output by the current output terminal. When the coincidence measurement is successful, the output terminal identifier number is output as a random number.

[0012] Furthermore, the nonlinear crystal is specifically a type II quasi-phase-matched PPKTP crystal with a domain periodicity reversal structure, located at the waist of the pump beam.

[0013] Furthermore, a long-pass filter is placed after the nonlinear crystal.

[0014] Furthermore, the single-photon performance measurement module includes a first polarization beam splitter, a first half-wave plate, a second polarization beam splitter, a first fiber coupler, a second fiber coupler, and a third fiber coupler. The first half-wave plate is located at the transmission end of the first polarization beam splitter, the first fiber coupler is located at the reflection end of the first polarization beam splitter, the second polarization beam splitter is located behind the first half-wave plate, and the second fiber coupler and the third fiber coupler are located sequentially at the reflection end and the transmission end of the second polarization beam splitter.

[0015] Furthermore, the optical fiber transmission module includes a single-mode optical fiber, a fourth optical fiber coupler connected to the receiving end of the single-mode optical fiber and arranged sequentially according to the optical path propagation direction, a first quarter-wave plate, a second half-wave plate, and a second quarter-wave plate.

[0016] Furthermore, the polarization state forming module includes a third polarization beam splitter, a third half-wave plate, and a third quarter-wave plate arranged sequentially according to the optical path propagation direction.

[0017] Furthermore, the quantum walk network includes n quantum walk branches and n+1 output terminals. The i-th quantum walk branch includes one beam deflector and i half-wave plates. The i half-wave plates are located behind the i beams emitted from the previous quantum walk branch, and the beam deflector is located behind the i half-wave plates, thereby deflecting the beams transmitted from the i half-wave plates to form i+1 beams. i = 1, ..., n, where n is a positive integer greater than or equal to 1. The parallelogram optical path between every two adjacent beam deflectors in the quantum walk network is an MZ interferometer, and the upper and lower arms of each MZ interferometer have equal optical path lengths.

[0018] Furthermore, the system also includes:

[0019] The splitting ratio calculation module is used to calculate the optimal splitting ratio of each half-wave plate of the quantum walk network according to a preset algorithm. The optimal splitting ratio makes the quantum random number output by the quantum walk network conform to the target probability distribution.

[0020] Furthermore, the spectrophotometry calculation module is used to perform the following steps during execution:

[0021] Step A: Establish the probability distribution P of the i-th output terminal of the quantum walk network. i Regarding the beam splitting ratio {θ} of each half-wave plate j The functional relationship of}: P i =F({θ j |j=1,…,m}),i=1,..,n+1,m is the number of half-wave plates;

[0022] Step B: Randomly select a value from 0 to 1 as the splitting ratio θ of the half-wave plate in the quantum walk network. j The initial value;

[0023] Step C: Using the functional relationship established in Step A, calculate the probability distribution {P} of the quantum walk network based on the current splitting ratio of the half-wave plate. i |i=1,…,n+1};

[0024] Step D: Calculate the target probability distribution {T} i The probability distribution {P} of the current quantum walk network i Losses between}

[0025] Step E: Based on the principle of error gradient descent, update the splitting ratio value of each half-wave plate as follows: Where η is the learning rate, and its value ranges from 0 to 1. The derivative of the loss L with respect to the splitting ratio of each half-wave plate is obtained by the chain rule.

[0026] Step F: Iteratively repeat steps C to E until the loss is less than the set value, then stop the iteration process and output the split ratio of the half-wave plate at this point as the optimal split ratio.

[0027] This invention also provides a method for generating quantum random numbers based on photonic quantum walks, comprising:

[0028] Step 1: Pump light is incident on a nonlinear crystal. A small portion of the pump photons generate a pair of parametric photons with orthogonal polarization, namely a signal photon and an idler photon. After passing through the nonlinear crystal, a long-pass filter is applied, and most of the pump photons are reflected and filtered out by the long-pass filter.

[0029] Step 2: Place a single-photon performance measurement module after the long-pass filter to measure the single-photon performance of the system, extract the signal photon delay as a prediction, and transmit idler photons;

[0030] Step 3: Transmit the idler photons through the optical fiber to achieve beam shaping and compensate for the phase generated by the idler photons in the optical fiber;

[0031] Step 4: Prepare the phase-compensated idler photons into idler photons of the desired polarization state;

[0032] Step 5: Use a quantum walk network to output the idler photons of the desired polarization state from each output terminal according to the target probability distribution. The quantum walk network has multiple output terminals.

[0033] Step 6: Collect idler light at the output of the quantum walk network using single-photon detector 18, and perform a coincidence measurement between the signal photon extracted by the single-photon performance measurement module and the idler photon output at the current output. When the coincidence measurement is successful, output the current output identifier number as a random number.

[0034] Beneficial effects: Compared with the prior art, the significant advantages of this invention are: (1) It adopts a multi-step quantum walk network, and by setting different parameters of the quantum walk network, any desired probability distribution can be generated, including uniform distribution (see Figure 4 ) and Gaussian distribution (see Figure 5 (2) A high-fidelity, low-noise quantum random number generator can be achieved using only linear optical elements. The method is convenient, stable, and reliable. For example, the uniform distribution and left-hand circular polarization fidelity are 95.8% (see Figure 4 The right-hand circular polarization fidelity is 96.5% (see...). Figure 4 Gaussian distribution, left-handed circular polarization fidelity of 95.8% (see...). Figure 5 The right-hand circular polarization fidelity is 94.1% (see...). Figure 5(3) Set up a split ratio calculation module to calculate the optimal parameter value of the quantum walk network (the optimal split ratio of each half-wave plate), so that the quantum random number output by the quantum walk network increases the degree of conformity with the target probability distribution. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the structure of the quantum random number generation system based on photonic quantum walk provided by the present invention;

[0036] Figure 2 It is a second-order time correlation function g 2 The graph shows a coincidence time of 3ns, g 2 The minimum value of the function is 0.02860 ± 0.00001;

[0037] Figure 3 This is a probability distribution diagram of four types of polarized light after passing through a four-step quantum walk network. The order of the bars from left to right is horizontally polarized, vertically polarized, right-handed circularly polarized, and left-handed circularly polarized light. The horizontal lines on the bars represent theoretical values.

[0038] Figure 4 This is a probability distribution diagram of the quantum random number sequence output after passing right-handed and left-handed circularly polarized light through a 4-step quantum walk network, with a uniform distribution as the target probability distribution.

[0039] Figure 5 This is a probability distribution diagram of the quantum random number sequence output after passing right-handed and left-handed circularly polarized light through a 4-step quantum walk network, with a Gaussian distribution as the target probability distribution. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. The reference to "embodiment" herein means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0042] This embodiment provides a quantum random number generation system based on optical quantum walks, such as... Figure 1 As shown, the system includes: a pump source for emitting pump light; a nonlinear crystal for generating a pair of orthogonally polarized parametric lights (signal light and idler light) from the incident pump light; a single-photon performance measurement module for measuring the system's single-photon performance, extracting the delayed signal photon as a prediction, and transmitting the idler photon; an optical fiber transmission module for transmitting the idler light through an optical fiber, achieving beam shaping, and compensating for the phase generated by the idler photon in the optical fiber; a polarization state forming module for preparing the phase-compensated idler light into the desired polarization state; a quantum walk network with multiple output terminals for outputting the idler photon of the desired polarization state from each output terminal according to the target probability distribution; a single-photon detector for performing a coincidence measurement between the signal photon extracted by the single-photon performance measurement module and the idler photon output from the output terminal of the quantum walk network, outputting the current output terminal identifier as a random number when the coincidence measurement is successful; and a splitting ratio calculation module for calculating the optimal splitting ratio of each half-waveplate of the quantum walk network according to a preset algorithm. Each module is described in detail below.

[0043] The pump source emits pulsed pump light, which in this embodiment is pulsed light with a wavelength of 397.49 nm.

[0044] The nonlinear crystal is specifically a type II quasi-phase-matched PPKTP nonlinear crystal 1 with a domain period inversion structure, a structural period of 8.8 μm, located at the waist of the pump beam. When the pulsed pump light is incident orthogonally on the PPKTP nonlinear crystal 1, a spontaneous parametric down-conversion process occurs, generating a pair of orthogonally polarized parametric beams, namely a signal photon and an idler photon. Both the front and rear end faces of the PPKTP nonlinear crystal 1 are coated with antireflection films for the pump and parametric beams, with transmittance greater than 99.8%. The center wavelength of the parametric beams is 794.98 nm.

[0045] A long-pass filter 2 is placed after the nonlinear crystal to filter out pump light that does not participate in the spontaneous parametric down-conversion.

[0046] The single-photon performance measurement module includes a first polarization beamsplitter 3, a first half-wave plate 4, a second polarization beamsplitter 5, a first fiber coupler 6, a second fiber coupler 7, and a third fiber coupler 8. The first half-wave plate 4 is located at the transmission end of the first polarization beamsplitter 3, the first fiber coupler 6 is located at the reflection end of the first polarization beamsplitter 3, the second polarization beamsplitter 5 is located behind the first half-wave plate 4, and the second fiber coupler 7 and the third fiber coupler 8 are located at the reflection end and transmission end of the second polarization beamsplitter 5, respectively. The first polarization beamsplitter 3 reflects vertically polarized signal light and transmits horizontally polarized idler photons. The signal photons are collected by the first fiber coupler 6 and, after a delay of 59 ns, are used as a prediction signal. The front and rear ends of the first polarization beamsplitter 3 and the second polarization beamsplitter 5 are coated with parametric antireflection films, with transmittance greater than 99.5%. The front and rear ends of the first half-wave plate 4 are also coated with parametric antireflection films, with transmittance greater than 99.8%. Experimentally, the arrival time of a single photon at the detector can be precisely measured using an electronic system. If the signal photon and the idler photon are detected simultaneously, meaning the time difference between their arrival at the detector is within 3 ns, it is considered a two-body coincidence event and is recorded by the electronic system. Predicting a single-photon source means that by precisely measuring the arrival time of the signal photon, the temporal or spatial position of the idler photon can be predicted, thus generating a sequence of single photons. Its single-photon nature is characterized by a second-order time correlation function, implemented using a single-photon performance measurement module. The second-order time correlation function is also known as g... 2 Function: g 2 =P(678)P(6) / [P(67)P(68)], where P(678) refers to the probability of three-body coincidence events where fiber couplers 6, 7, and 8 simultaneously detect photons, P(6) refers to the probability of single-body coincidence events where the first fiber coupler 6 detects photons, and P(67) and P(68) refer to the probabilities of two-body coincidence events where 6 and 7 and 6 and 8 detect photons, respectively. At the delay time τ = 0, g 2 The closer the function value is to 0, the better the single-photon property (see...). Figure 2 ).

[0047] The fiber optic transmission module includes a single-mode fiber and a fourth fiber coupler 9, a first quarter-wave plate 10, a second half-wave plate 11, and a second quarter-wave plate 12, sequentially arranged in the optical propagation direction at the receiving end of the single-mode fiber. The single-mode fiber can achieve beam shaping of the idler light. The first quarter-wave plate 10, the second half-wave plate 11, and the second quarter-wave plate 12 are combined to compensate for the phase generated by the idler light in the single-mode fiber. The front and rear ends of the above three wave plates are all coated with parametric antireflection films, and the transmittance is greater than 99.8%.

[0048] The polarization state forming module includes a third polarization beamsplitter 13, a third half-wave plate 14, and a third quarter-wave plate 15 arranged sequentially according to the optical path propagation direction. The angles of the third half-wave plate 14 and the third quarter-wave plate 15 can be rotated to convert the horizontal polarization state emitted from the polarization beamsplitter 13 into arbitrary polarization states (including horizontal polarization, vertical polarization, left-hand circular polarization, and right-hand circular polarization). Both the front and rear ends of the third polarization beamsplitter 13 are coated with parametric antireflection films, with transmittance greater than 99.5%. Similarly, the front and rear ends of the third half-wave plate 14 and the third quarter-wave plate 15 are coated with parametric antireflection films, with transmittance greater than 99.8%.

[0049] The quantum walk network consists of n quantum walk branches and n+1 output terminals. The i-th quantum walk branch includes one beam deflector and i half-wave plates. The i half-wave plates are located behind the i beams emitted from the previous quantum walk branch. The beam deflector is located behind the i half-wave plates, thus deflecting the beams transmitted from the i half-wave plates to form i+1 beams; i = 1, ..., n, where n is a positive integer greater than or equal to 1. The parallelogram optical path between every two adjacent beam deflectors in the quantum walk network is an MZ interferometer, and the upper and lower arms of each MZ interferometer have equal optical path lengths. For example... Figure 1 The diagram shows a 4-step quantum walk network with 5 outputs, comprising 10 half-wave plates and 4 beam deflectors. From left to right, the first half-wave plate 17 and the first beam deflector 16 form the first quantum walk branch; the second and third half-wave plates 17 and the second beam deflector 16 form the second quantum walk branch; the fourth, fifth, and sixth half-wave plates 17 and the third beam deflector 16 form the third quantum walk branch; and the seventh, eighth, ninth, and tenth half-wave plates 17 and the fourth beam deflector 16 form the fourth quantum walk branch. Increasing the number of half-wave plates 17 and beam deflectors 16 further expands the network to include more steps. Beam deflectors 16 transmit vertically polarized and horizontally polarized light respectively, with the horizontally polarized light exhibiting a lateral shift of approximately 4 mm relative to the vertically polarized light. The half-wave plate 17 has a central aperture of approximately 5 mm, with the remainder being a glass substrate, enabling polarization manipulation of a single beam. The combination of beam deflectors 16 and half-wave plates 17 allows adjustment of the splitting ratio between vertically and horizontally polarized light. All ten half-wave plates 17 have parametric anti-reflection coatings on their front and rear ends, achieving a transmittance greater than 99.8%. Similarly, the four beam deflectors 16 have parametric anti-reflection coatings on their front and rear ends, achieving a transmittance greater than 99.2%. In the 4-step quantum walk network, the parallelogram optical path between every two adjacent beam deflectors forms one MZ interferometer. The 4-step quantum walk network forms a total of six MZ interferometers. The tilted beam deflectors and the half-wave plates placed within the interferometers ensure that the upper and lower arms of each MZ interferometer have equal optical path lengths.

[0050] The splitting ratio calculation module is used to calculate the optimal splitting ratio of each half-wave plate of the quantum walk network according to a preset algorithm. This optimal splitting ratio ensures that the quantum walk network outputs quantum random numbers that conform to the target probability distribution. During execution, it performs the following steps: Step A: Establish the probability distribution P of the i-th output of the quantum walk network. i Regarding the beam splitting ratio {θ} of each half-wave plate j The functional relationship of}: P i =F({θ j |j=1,…,m}),i=1,..,n+1,m is the number of half-wave plates; Step B: Randomly select values ​​from 0 to 1 as the splitting ratio θ of the half-wave plates in the quantum walk network. j Initial value; Step C: Using the functional relationship established in Step A, calculate the probability distribution {P} of the quantum walk network based on the current splitting ratio of the half-wave plate. i |i=1,…,n+1};Step D: Calculate the target probability distribution {T} i The probability distribution {P} of the current quantum walk network i Losses between} Step E: Based on the principle of error gradient descent, update the splitting ratio value of each half-wave plate as follows: Where η is the learning rate, and its value ranges from 0 to 1. The derivative of the loss L with respect to the splitting ratio of each half-wave plate is obtained using the chain rule. Step F: Iteratively repeat steps C to E until the loss is less than a set value, then stop the iteration process and output the splitting ratio of the half-wave plate at this point as the optimal splitting ratio. After the optimal splitting ratio is calculated, the polarization state is set by rotating the angle of each half-wave plate in the quantum walk network. Then, the splitting ratio of the vertically polarized and horizontally polarized light obtained by the beam deflector reaches the optimal splitting ratio.

[0051] Each output of the quantum walk network is connected to a single-photon detector 18, which is used to perform a coincidence measurement between the signal photon extracted by the single-photon performance measurement module (output of the first fiber coupler 6) and the idler photon output at the current output. When the coincidence measurement is successful, the output identifier number at this time is output as a random number.

[0052] This invention system forms a predictive single-photon source, outputting a pair of photons (signal photon and idler photon) each time. One signal photon is extracted by the first fiber coupler 6 as a prediction signal, while the other idler photon enters the quantum walk network and ultimately exits from one of multiple output terminals. Each output terminal is connected to a single-photon detector 18 to perform a coincidence measurement between the signal photon and the idler photon. If the difference between the detected signal photon and the idler photon is within 3 ns, the coincidence measurement is considered successful. The identifier of the currently successful output terminal is then output as a random number. For example... Figure 1 The five output terminals shown are numbered 1, 2, 3, 4, and 5 from top to bottom. If the idler photon is output from the third output terminal (marked as 3), and is detected by the single-photon detector and the measurement is successful, a random number 3 is output and recorded by the electronics system. The single-photon source is then predicted to continuously send photon pairs, thus continuously recording a set of data, for example: 3, 2, 5, 2, 2, 3, 5, ... This is the quantum random number sequence. Because the probability distribution of the output terminals of the quantum walk network conforms to the target probability distribution, the quantum random number sequence also conforms to the target probability distribution.

[0053] This embodiment also provides a quantum random number generation method based on optical quantum walks. This method is implemented using the above-mentioned system and specifically includes:

[0054] Step 1: Pulsed pump light with a wavelength of 397.49 nm is incident on PPKTP nonlinear crystal 1. A small portion of the pump photons generate a pair of parametric photons with orthogonal polarization, namely a signal photon and an idler photon. After passing through the long-pass filter 2 of PPKTP nonlinear crystal 1, the center wavelength of the parametric photons is 794.98 nm. Most of the pump photons are reflected and filtered out by the long-pass filter 2.

[0055] Step 2: Place a single-photon performance measurement module after the long-pass filter 2 to measure the single-photon performance of the system, extract the signal photon delay as a prediction, and transmit idler photons. The single-photon performance measurement module includes a first polarization beamsplitter 3, a first half-wave plate 4, and a second polarization beamsplitter 5 placed sequentially. A first fiber coupler 6 is located at the reflection end of the first polarization beamsplitter 3, and a second fiber coupler 7 and a third fiber coupler 8 are located at the reflection end and transmission end of the second polarization beamsplitter 5, respectively. The combined module is used to measure the second-order time correlation function of the predicted single-photon source.

[0056] Step 3: Transmit the idler photons through optical fiber to achieve beam shaping and compensate for the phase generated by the idler photons in the fiber. In specific implementation, the idler photons are shaped by the single-mode fiber and then emitted through the fourth fiber coupler 9. A combined waveplate consisting of the first quarter-wave plate 10, the second half-wave plate 11, and the second quarter-wave plate 12 is placed sequentially after the fourth fiber coupler 9 to compensate for the phase generated by the idler photons in the single-mode fiber.

[0057] Step 4: Prepare the phase-compensated idler photons into idler photons with the desired polarization state. Specifically, an arbitrary polarization state can be generated by combining the third polarization beam splitter 13, the third half-wave plate 14, and the third quarter-wave plate 15.

[0058] Step 5: A quantum walk network is used to output the idler photons of the desired polarization state from each output terminal according to the target probability distribution. The quantum walk network has multiple output terminals. Based on the target probability distribution, the quantum walk network uses a preset algorithm to calculate the splitting ratio of each half-wave plate in the quantum walk network, and sets the angle of each half-wave plate according to the splitting ratio, so that the idler photons of the desired polarization state are output from each output terminal according to the target probability distribution.

[0059] Step 6: Collect idler light at the output of the quantum walk network using single-photon detector 18, and perform a coincidence measurement between the signal photon extracted by the single-photon performance measurement module and the idler photon output at the current output. When the coincidence measurement is successful, output the current output identifier number as a random number.

[0060] Measurements were performed on this invention using the third half-wave plate 14 and the third quarter-wave plate 15 to generate horizontally polarized, vertically polarized, right-handed circularly polarized, and left-handed circularly polarized light, respectively, which were then incident on the quantum walk network, resulting in four quantum walk distributions. (See...) Figure 3 From left to right. Right-handed and left-handed circularly polarized light are incident on the quantum walk network. When the target probability distribution is uniform, the quantum random number sequence output by this invention is shown below. Figure 4 It can be seen that it conforms to a uniform distribution. When the target probability distribution is a Gaussian distribution, the quantum random number sequence output by this invention is shown in the figure. Figure 5 As can be seen, it conforms to a Gaussian distribution. Therefore, it can be concluded that the quantum random number sequence output by this invention conforms to the target probability distribution.

Claims

1. A quantum random number generation system based on photonic quantum walks, characterized in that, include: Pump light source, used to emit pump light; Nonlinear crystals are used to generate a pair of parametric photons with orthogonal polarization from incident pump light, namely a signal photon and an idler photon; The single-photon performance measurement module is used to measure the single-photon performance of the system, extract the signal photon delay as a prediction, and transmit idler photons; The fiber optic transmission module is used to transmit idler photons through optical fibers, thereby achieving beam shaping and compensating for the phase generated by the idler photons in the optical fiber. The polarization state forming module is used to prepare phase-compensated idler photons into idler photons with the desired polarization state. A quantum walk network has multiple output terminals, which are used to output idler photons of the desired polarization state from each output terminal according to the target probability distribution; A single-photon detector is connected to each output terminal of the quantum walk network. It is used to perform a coincidence measurement between the signal photon extracted by the single-photon performance measurement module and the idler photon output by the current output terminal. When the coincidence measurement is successful, the current output terminal identifier number is output as a random number. The splitting ratio calculation module is used to calculate the optimal splitting ratio of each half-wave plate of the quantum walk network according to a preset algorithm. The optimal splitting ratio makes the quantum walk network output quantum random numbers conform to the target probability distribution. The spectrophotometer ratio calculation module is used to perform the following steps during execution: Step A: Establish the probability distribution of the i-th output terminal of the quantum walk network. Regarding the splitting ratio of each half-wave plate Functional relationship: ,i=1,..,n+1, m is the number of half-wave plates; Step B: Randomly select a value from 0 to 1 as the splitting ratio of the half-wave plate in the quantum walk network. The initial value; Step C: Using the functional relationship established in Step A, calculate the probability distribution of the quantum walk network based on the current splitting ratio of the half-wave plate. ; Step D: Calculate the target probability distribution The probability distribution of the current quantum walk network Losses between ; Step E: Based on the principle of error gradient descent, update the splitting ratio value of each half-wave plate as follows: ,in The learning rate, with a value between 0 and 1. The derivative of the loss L with respect to the splitting ratio of each half-wave plate is obtained by the chain rule. Step F: Iteratively repeat steps C to E until the loss is less than the set value, then stop the iteration process and output the split ratio of the half-wave plate at this point as the optimal split ratio.

2. The quantum random number generation system based on photonic quantum walks according to claim 1, characterized in that: The nonlinear crystal is specifically a type II quasi-phase-matched PPKTP crystal with a domain period inversion structure, located at the waist of the pump beam.

3. The quantum random number generation system based on photonic quantum walks according to claim 1, characterized in that: A long-pass filter is placed after the nonlinear crystal.

4. The quantum random number generation system based on photonic quantum walks according to claim 1, characterized in that: The single-photon performance measurement module includes a first polarization beamsplitter, a first half-wave plate, a second polarization beamsplitter, a first fiber coupler, a second fiber coupler, and a third fiber coupler. The first half-wave plate is located at the transmission end of the first polarization beamsplitter, the first fiber coupler is located at the reflection end of the first polarization beamsplitter, the second polarization beamsplitter is located behind the first half-wave plate, and the second fiber coupler and the third fiber coupler are located at the reflection end and transmission end of the second polarization beamsplitter, respectively.

5. The quantum random number generation system based on photonic quantum walks according to claim 1, characterized in that: The optical fiber transmission module includes a single-mode optical fiber, a fourth optical fiber coupler connected to the receiving end of the single-mode optical fiber and arranged sequentially according to the optical path propagation direction, a first quarter-wave plate, a second half-wave plate, and a second quarter-wave plate.

6. The quantum random number generation system based on photonic quantum walks according to claim 1, characterized in that: The polarization state forming module includes a third polarization beam splitter, a third half-wave plate, and a third quarter-wave plate arranged sequentially according to the optical path propagation direction.

7. The quantum random number generation system based on photonic quantum walks according to claim 1, characterized in that: The quantum walk network includes n quantum walk branches and n+1 output terminals. The i-th quantum walk branch includes one beam deflector and i half-wave plates. The i half-wave plates are located behind the i beams emitted from the previous quantum walk branch. The beam deflector is located behind the i half-wave plates, thereby deflecting the beams transmitted from the i half-wave plates to form i+1 beams. i = 1, ..., n, where n is a positive integer greater than or equal to 1. The parallelogram optical path between every two adjacent beam deflectors in the quantum walk network is an MZ interferometer, and the upper and lower arms of each MZ interferometer have equal optical path lengths.

8. A method for a quantum random number generation system based on photonic quantum walks according to claim 1, characterized in that, include: Step 1: Pump light is incident on a nonlinear crystal. A small portion of the pump photons generate a pair of parametric photons with orthogonal polarization, namely a signal photon and an idler photon. After passing through the nonlinear crystal, a long-pass filter is applied, and most of the pump photons are reflected and filtered out by the long-pass filter. Step 2: Place a single-photon performance measurement module after the long-pass filter to measure the single-photon performance of the system, extract the signal photon delay as a prediction, and transmit idler photons; Step 3: Transmit the idler photons through the optical fiber to achieve beam shaping and compensate for the phase generated by the idler photons in the optical fiber; Step 4: Prepare the phase-compensated idler photons into idler photons of the desired polarization state; Step 5: Use a quantum walk network to output the idler photons of the desired polarization state from each output terminal according to the target probability distribution. The quantum walk network has multiple output terminals. Step 6: Collect idler light at the output of the quantum walk network using a single-photon detector, and perform a coincidence measurement between the signal photon extracted by the single-photon performance measurement module and the idler photon output at the current output. When the coincidence measurement is successful, output the current output identifier number as a random number.

Images