A chip quantum random number generation device and a generation method

Abstract

The application discloses a chipped quantum random number generation device, which comprises a spontaneous emission light source, a cascade light splitting waveguide, a weak light detector and a control circuit, the spontaneous emission light source is used for emitting spontaneous photons based on a power supply signal; the cascade light splitting waveguide has single-input and multiple-output; the weak light detector is a detector array, each detector corresponds to one output of the cascade light splitting waveguide; the control circuit receives the number of photons detected by each detector in a predetermined light emission period respectively, and generates quantum random numbers based on the number of photons. Through the Y-shaped cascade light splitting waveguide design, the use of low-cost silicon-based light detectors and the generation rate of photon counting quantum random numbers can be considered.

Description

Technical Field

[0001] This invention relates to the field of quantum optics, and more specifically to a silicon photonic chip and a corresponding method for generating quantum random numbers using quantum optics. Background Technology

[0002] Quantum random numbers are random numbers generated using certain random characteristics of quantum systems. Their randomness originates from the principles of quantum mechanics and is inherently random. Because the randomness of quantum random numbers is supported by explicit quantum mechanical theory, it can be guaranteed that random numbers possess the principle of unpredictability. Furthermore, compared to classical random number devices, quantum random number devices have the characteristics of low power consumption and insensitivity to environmental variables.

[0003] Quantum random number generators can be classified into the following categories according to the different sources of randomness:

[0004] 1. Quantum random number generators based on optical systems, such as photon path selection, photon arrival time, photon vacuum state fluctuation, laser phase fluctuation, and laser tube radiation mode competition.

[0005] 2. Non-optical quantum random number generators, such as tunneling-based decay, pn junction shot noise, and random walks of neutral atoms.

[0006] 3. Quantum random number generator based on radioactive nuclear decay.

[0007] Of the above technical approaches, the third approach has issues with radiation safety and stability, while the second approach suffers from implementation complexity, making it difficult to miniaturize or even chip-scale. Currently, most quantum random number devices are based on the first type of optical system quantum random number generator. For example, SKT-IDQ has already developed a consumer-grade quantum random number chip, which is already installed in the Samsung Galaxy Quantum smartphone.

[0008] According to quantum optics theory, photon number detection involves quantum random shot noise, which follows a Poisson distribution. In a Poisson distribution, when the photon number is N, its uncertainty (i.e., the standard deviation of the Poisson distribution) is the square root of N. Therefore, the shot noise of a photon is the square root of the number of photons. Since this shot noise originates entirely from the quantum properties of photons, it belongs to quantum noise. Converting this quantum noise into counting results in quantum random numbers.

[0009] However, due to the Gaussian background in the photon number distribution of the light source, the quantum noise varies greatly depending on the photon number distribution. For example, the quantum noise of 100 photons is... That is, a photon count fluctuation within the range of 10% compared to the total photon count can be converted into a truly random number between 1 and 10. However, when the number of photons from the light source is large, such as 10,000 photons, the quantum noise, although... However, compared to the total photon count fluctuation of only 1%, it is difficult to distinguish.

[0010] Therefore, extracting a quantum random number with a Gaussian background requires the array detector to remain sensitive to pixels with fewer photons at the edges of the Gaussian distribution. This necessitates that the detector's sensitivity approach the resolution of a single photon, increasing costs. Alternatively, detection can be limited to locations with a higher number of photons within the Gaussian distribution, sacrificing the efficiency of quantum random number generation. Summary of the Invention

[0011] Purpose

[0012] To eliminate the background differences caused by the Gaussian distribution of photon numbers and better extract quantum random numbers from photon number noise, this invention proposes a quantum random number generator using silicon photonics technology. This device comprises four parts: a spontaneous emission source, a cascaded beam splitter waveguide, a silicon-based weak light detector, and a control (digital circuitry). By utilizing the quantum randomness of photon path selection in the cascaded beam splitter waveguide and the quantum randomness of photon arrival time, a truly random binary number incorporating both quantum randomnesses is generated. Simultaneously, the cascaded beam splitter waveguide eliminates the Gaussian distribution of photon numbers in free space, resulting in a more uniform distribution of photon numbers reaching the detector.

[0013] Specifically, the present invention provides a quantum random number generator, characterized in that the quantum random number generator comprises: a spontaneous emission light source, a cascaded optical waveguide, a weak light detector, and a control circuit.

[0014] The spontaneous emission light source is used to emit spontaneous light particles based on a power supply signal;

[0015] The cascaded optical waveguide has a single input and multiple outputs. The single input is used to receive photons emitted by the spontaneous emission source. There is a cascaded Y-shaped waveguide between the single input and the multiple outputs. The number of channels of the waveguide increases in the form of 2 to the power of N, where N is a positive integer greater than or equal to 2, representing the number of stages between the single input and the multiple outputs.

[0016] The weak light detector is a detector array, with each detector corresponding to one output of the cascaded optical waveguide;

[0017] The control circuit receives the number of photons detected by each detector within a predetermined emission period and generates a quantum random number based on the number of photons.

[0018] In a preferred implementation, the spontaneous emission source periodically emits light pulses in the form of pulses to the input of the cascaded optical waveguide.

[0019] In another preferred implementation, the exposure integration time of the low-light detector is matched with the emission time of the spontaneous emission source.

[0020] In another preferred implementation, the spectral broadening of the spontaneous emission source is less than the speed of light divided by the longest optical path distance between the spontaneous emission source and the weak light detector. Preferably, the spontaneous emission source, the cascaded beam splitter, the weak light detector, and the control circuit are all fabricated on the same silicon wafer, and the spontaneous emission source and the cascaded beam splitter are arranged sequentially to each other.

[0021] In another preferred implementation, the control circuit converts the voltage signal received by each detector into a binary digital signal and extracts the low bits of the digital signal as a quantum random number.

[0022] In another preferred implementation, the control circuit converts the voltage signal received by each detector into an 8-bit binary digital signal and extracts the lower 4 bits of the digital signal as a quantum random number.

[0023] On the other hand, the present invention provides a method for generating quantum random numbers, characterized in that the method includes:

[0024] Photons are emitted using a light source;

[0025] The photons emitted by the light source are cascaded and split, with the output of each split doubling the original number of input channels.

[0026] The output photons in each channel after cascaded beam splitting are detected;

[0027] The number of photons detected by each channel within a predetermined emission period is received, and a quantum random number is generated based on the number of photons.

[0028] In a preferred embodiment, the light source is a spontaneously emitting light source that periodically emits light pulses in a pulsed manner.

[0029] In another preferred implementation, the spectral broadening of the light source is less than the speed of light divided by the longest optical path distance between the light source and the detector that probes the split channel.

[0030] In another preferred implementation, the voltage signal generated based on the detected photons is converted into a binary digital signal, and the low bits of the digital signal are extracted as quantum random numbers.

[0031] It should be noted that the "chip-based" approach mentioned in this invention can refer to the quantum random book generator being formed by directly etching onto a single silicon wafer, or it can refer to the process of fabricating different silicon light sources, cascaded waveguides, detectors, etc., on different silicon wafers and then setting them up in sequence.

[0032] Invention principles and technical effects

[0033] At the light source, the emission power of a silicon spontaneous emission source is relatively weak, and its quantum efficiency is much lower than that of a conventional LED. Since quantum random numbers require a weak light source, and this invention utilizes a silicon spontaneous emission source, there is no need to add optical attenuation devices at the light source; the weak light source required for quantum random numbers can be achieved simply by adjusting the coupling efficiency. Furthermore, there are numerous spontaneous emission modes. Since the beam splitter is a silicon optical waveguide, considering the characteristics of silicon optical waveguides, only the TE0 mode (parallel to the beam splitter's optical axis) and the TM0 mode (perpendicular to the optical axis) can enter the beam splitter for propagation. Therefore, the beam splitter also acts as an attenuator, retaining only the power of the TE0 and TM0 modes and isolating other optical power outside the beam splitter's optical waveguide.

[0034] In the optical waveguide, this invention, through a Y-type cascaded optical waveguide design, can simultaneously achieve the use of low-cost silicon-based photodetectors and the generation rate of photon counting quantum random numbers, whereas existing quantum random number chips that detect the Gaussian distribution of photon numbers in free space cannot simultaneously achieve both. Photons from a silicon-based spontaneous emission source enter the Y-type cascaded optical waveguide and propagate in TEO mode (TMO mode attenuates too quickly and is not considered here), as shown in the attached diagram. Figure 2 At the bifurcation point, a quantum randomness occurs, with each photon heading towards one of the two paths of the split waveguide with a 1 / 2 probability. This continues in the next stage, with each photon heading towards one of the two paths of the split waveguide with a 1 / 2 probability, and so on. Although spontaneously emitted photons have a Gaussian distribution in free space, this distribution is disrupted once they enter the waveguide; the Gaussian distribution only exists within each waveguide. The overall light distribution becomes a uniform distribution, meaning the power at each exit port of the Y-type cascaded split waveguide is approximately equal. Quantum noise based on this becomes the source of quantum random numbers.

[0035] Photons are identical particles in quantum mechanics; photons emitted simultaneously by a single-frequency light source at the same time are indistinguishable when their quantum numbers are identical. However, the light source in this invention is a silicon spontaneous emission source, which exhibits certain broadening in both its frequency spectrum and emission time. According to the coherence length formula:

[0036] L=c / Δν

[0037] Where c is the speed of light in vacuum, and Δv is the spectral broadening. The spectral broadening of a spontaneous emission source generally exceeds 100 GHz, so the corresponding coherence length L is less than 3 mm. Furthermore, in this invention, the spectral broadening of the spontaneous emission source is set to be less than the speed of light divided by the longest optical path distance between the spontaneous emission source and the weak light detector, thus making it suitable for non-identical particles.

[0038] Since photons behave as non-identical particles within the chip, the quantum randomness in this invention has two sources. One is the path-selection quantum noise unique to this invention, i.e., the photon in... Figure 2 The randomness of each path selection in a Y-type cascaded waveguide ultimately manifests as a random distribution of photon numbers at multiple exit points of the cascaded waveguide. Since non-identical particles conform to probability theory, the more cascaded the Y-type waveguide, the more this distribution approximates a spatial Poisson distribution, denoted as:

[0039]

[0040] Where m is the average expected number of photons in each path after a light pulse passes through a Y-type cascaded waveguide, n is the actual number of photons, and P(n) is the probability of that number of photons. According to the Poisson distribution principle, the standard deviation of the number of photons in each path is... The spatially distributed quantum noise that can be obtained is

[0041] Secondly, there is quantum noise in the photon arrival time. By configuring the detectors and setting a predetermined time window, the number of photons arriving at each detector within that time window follows a Poisson distribution over time, specifically:

[0042] Where n is the average expected value of the number of photons in each path during the detector's activation period, y is the actual number of photons in each path, and P(y) is the probability of that number of photons. According to the Poisson distribution principle, the standard deviation of the number of photons in each path is... That is, the number of photons in the time distribution that can be obtained is quantum noise.

[0043] In this invention, spatially distributed quantum noise originates from the Y-type cascaded optical waveguide, while temporally distributed photon quantum noise originates from the light source and detector themselves; these two are completely independent and can be superimposed. When the emission window and detection window are equal, and this window is taken as a unit time, m = n can be considered. Therefore, the usable photon quantum noise is... That is, the source of quantum random numbers is the same type of photon noise obtained by only using a detector, which is not twice that of the cascaded split waveguide method.

[0044] In a preferred implementation, all the aforementioned devices are fabricated directly on a silicon wafer. For example, the three optical components—the light source, the cascaded beam splitter, and the detector—use commonly used silicon photonics processes such as CMOS, while the digital circuitry utilizes silicon transistor technology common to integrated circuits. All three components—the light source, the cascaded beam splitter, and the detector—are arranged sequentially on the same silicon wafer, and the digital circuitry controls them. This implementation significantly simplifies the packaging process of quantum random number chips, eliminating the need to package silicon-based and non-silicon-based microdevices together, thus improving the overall manufacturing efficiency of quantum random number chips. Another advantage of all-silicon integration is one-time manufacturing, eliminating instability caused by mechanical drift and thermal expansion and contraction of optical components, thereby eliminating classical noise and ensuring that random numbers originate entirely from the quantum random process within the chip.

[0045] This invention enables the creation of a random number generator with dual quantum randomness by combining a low-cost linear array detector with a silicon photonic spontaneous emission source and a cascaded waveguide, and the random number generation efficiency is significantly higher than that of existing technologies. Attached Figure Description

[0046] The accompanying drawings provided below are primarily for illustrative purposes to further understand the present invention and constitute a part of this invention. The illustrative embodiments and descriptions of the present invention are used to explain the invention and do not constitute an improper limitation of the invention. In the drawings:

[0047] Figure 1 This refers to the system components of a quantum random number generator.

[0048] Figure 2 This is a schematic diagram of a silicon-based cascaded beam splitter.

[0049] Figure 3 This is a schematic diagram of a method for extracting quantum random numbers from signals from a weak light detector. Detailed Implementation

[0050] The present invention will be further described in detail below with reference to the embodiments and accompanying drawings, but the embodiments of the present invention are not limited thereto.

[0051] The specific embodiments of the present invention are mainly provided to facilitate a further understanding of the present invention and constitute a part of the present invention. The illustrative embodiments and descriptions of the present invention are used to explain the present invention and do not constitute an improper limitation of the present invention.

[0052] like Figure 1 As shown, this embodiment presents a quantum random number generator. This quantum random number generator comprises four main parts:

[0053] (1) Silicon-based spontaneous emission light source: A silicon-based spontaneous emission light source is placed at the front end of the equipment to emit light spontaneously based on a power supply signal. For example, silicon-based light-emitting diodes (LEDs) can be fabricated on a silicon wafer using CMOS or other processes, or other types of spontaneous emission light sources can be integrated. The fabrication of the silicon-based spontaneous emission light source is an existing technology in this field and will not be detailed here. The power supply to the spontaneous emission light source can be controlled by digital circuits to control the emission of light. High level emits light, low level does not emit light. The design requires a voltage stability better than 1%, so that the light emission stability of the light source is better than 1%.

[0054] (2) Cascaded Optical Splitting Waveguide: A cascaded optical splitter waveguide is installed downstream of the silicon-based spontaneous emission light source. The cascaded optical splitter waveguide adopts a single-input, multi-output cascaded structure. The number of waveguides in each stage is twice that of the previous stage. That is, as shown in the attached diagram. Figure 2 As shown, the single-input point is a Y-shaped waveguide that splits from 1 to 2. Each channel of the waveguide is further split into two channels, and so on, from 1 to 2, 2 to 4, 4 to 8, and so on, creating a cascaded structure. For example, a 9-stage cascade has 2 to the power of 9 channels (512 channels), and a 10-stage cascade has 2 to the power of 10 channels (1024 channels), and so on. After photons from the spontaneous emission source enter the cascaded waveguide, quantum randomness occurs in path selection. In a preferred implementation, the cascaded waveguide and the silicon-based spontaneous emission source are mounted on the same silicon wafer and manufactured using semiconductor technology, placing them upstream and downstream, respectively.

[0055] (3) Low-light detector: A low-light detector is set in the light-emitting direction of the cascaded optical waveguide. For example, a silicon photodiode array (a linear array in this embodiment) is set at the light-emitting port of the cascaded optical waveguide. The array is located at the exit of the cascaded optical waveguide, and each silicon photodiode corresponds one-to-one with each exit of the cascaded optical waveguide. Each silicon photodiode has a light intensity resolution of about 100 photons. After detecting a photon, a photocurrent is formed and sent to the analog-to-digital converter (ADC) part of the control circuit. The detection window of the low-light detector is set so that it matches the emission period of the light source on the one hand, and the time length of the detection window is adjusted so that it can receive about 100 (or less) photons within the window period under the current illumination adjustment. Preferably, the above-mentioned silicon photodiode array is fabricated on a silicon wafer and is set on the same silicon wafer as the cascaded optical waveguide and the silicon-based spontaneous emission light source, arranged in sequence.

[0056] (4) Control Circuit: The control circuit adopts digital circuitry, which can be used to provide power based on an external clock signal for light emission control. That is, it receives an external clock signal and outputs a timing voltage signal through a counter to power the spontaneous emission light source, thereby controlling the silicon-based spontaneous emission light source to emit light. In addition, the control circuit also includes a signal processing module, which converts the analog voltage signal of the weak light detector into a digital signal (ADC) and extracts the low-order bits as random number output.

[0057] The specific working process of the quantum random number generator of the present invention includes the following steps:

[0058] 1. Power the quantum random number generator with a standard voltage of 3.3V or 5V. Use an external clock signal input to the control circuit of the quantum random number generator. The control circuit contains an internal counter (see attached diagram). Figure 1 The counter, according to firmware settings, generates pulse voltages to the silicon-based spontaneous emission light source. For example, with a clock speed of 10MHz, the counting unit time is 100ns (100 nanoseconds). If the counting period is set to 100us (100 microseconds), then one cycle counts 1000. The counter outputs voltage to the silicon-based spontaneous emission light source during the first 100 counts, providing power for 10us (10 microseconds). The idle time is from count 101 to 1000, i.e., 900us. In other words, the counter causes the silicon-based spontaneous emission light source to emit a 10us light pulse every 100us, with a duty cycle of 1:9.

[0059] 2. Light pulses emitted by silicon-based spontaneous emission light sources enter the surrounding area. Figure 2 The Y-type cascaded beam splitter waveguide allows for adjustment of coupling efficiency based on the waveguide port size and distance from the light source. Preferably, the coupling efficiency between the light source and the beam splitter port is no higher than 5%, 3%, or 1%. For example, when the coupling efficiency is set to 1%, only 1% of the photons from a 10µs optical pulse enter the cascaded beam splitter waveguide, resulting in a 20dB attenuation. If the power of the silicon-based spontaneous emission light source is 100nW, the power after entering the cascaded beam splitter waveguide is 1nW, and the energy of the 10µs optical pulse is approximately 1fJ (1 femtojoule). After further attenuation of approximately 20dB within the Y-type cascaded beam splitter waveguide, the total energy reaching the detector from each output port is approximately 10aJ (10 atjoules), equivalent to the energy of 100 photons (the energy of a 1µm photon is approximately 0.2aJ).

[0060] 3. Each exit of the Y-type cascaded waveguide is equipped with a silicon-based weak light detector. Within one cycle, the detector's exposure integration time is matched to the emission time of the silicon-based spontaneous emission source. For example, in a 1ms cycle, if the silicon-based spontaneous emission source emits a 10µs light pulse, the detector window is also set to 10µs. The photon arrival time follows a Poisson distribution, and the quantum noise caused by the arrival time is within the square root of the photon number. Simultaneously, the fluctuation in the photon number due to path selection also follows a Poisson distribution. Therefore, in addition to the quantum noise caused by the arrival time, a quantum noise within the square root of the photon number is added. Thus, the relationship between the quantum noise and the particle number in this invention is: That is, when each detector receives about 100 photons in the detection window, the quantum noise range is about 20, as shown in the attached figure. Figure 3 .

[0061] 4. The control circuit converts the voltage signal received by each detector in the detector array into an 8-bit binary digital signal. (See attached diagram) Figure 3 As shown, the binary digital signal corresponding to 100 photons is "01100100", the binary digital signal corresponding to 127 photons is "11111111", and the binary digital signal corresponding to 80 photons is "01010000". Since quantum noise only accounts for about 20% of the background signal, the higher bits of this 8-bit number represent the background signal with weak randomness. Therefore, quantum random numbers need to extract the lower bits of this 8-bit number. For example, if the lower 4 bits account for 1 / 8 of the background, it can be used as a completely random quantum number.

[0062] 5. In the case of a 100µs period, a 10µs emission and detection window, 64 outputs of a cascaded waveguide, 40dB of coupling and splitting attenuation, and the lower 4 bits of the analog-to-digital converter (ADC), the quantum random number generation rate can reach 10000 × 4 bits × 64 channels = 2.56MHz, or 2.56 million quantum random numbers per second. This rate can be further increased by reducing the period and increasing the light intensity of the silicon-based spontaneous emission source, the number of stages in the cascaded waveguide, and the accuracy of the ADC. For example, with a 5µs period, a 1µs emission and detection window, 1024 outputs of a cascaded waveguide, 60dB of coupling and splitting attenuation, and the lower 8 bits of the 16-bit ADC, the quantum random number generation rate can reach 200000 × 8 bits × 1024 channels = 1.6GHz.

[0063] This invention enables the creation of a random number generator with dual quantum randomness using a low-cost linear array detector.

[0064] Although the principles of the present invention have been described in detail above with reference to preferred embodiments, those skilled in the art should understand that the above embodiments are merely illustrative explanations of the implementation of the present invention and are not intended to limit the scope of the present invention. The details in the embodiments do not constitute a limitation on the scope of the present invention. Any obvious changes, such as equivalent transformations or simple substitutions, based on the technical solutions of the present invention without departing from the spirit and scope of the present invention fall within the protection scope of the present invention.

Claims

1. A chip-based quantum random number generator, characterized in that, The quantum random number generator includes: a spontaneous emission light source, a cascaded optical waveguide, a weak light detector, and a control circuit. The spontaneous emission light source is used to emit spontaneous light particles based on the power supply signal; the spectral broadening of the spontaneous emission light source is less than the speed of light divided by the longest optical path distance between the spontaneous emission light source and the weak light detector; The cascaded optical waveguide has a single input and multiple outputs. The single input is used to receive photons emitted by the spontaneous emission source. There is a cascaded Y-shaped waveguide between the single input and the multiple outputs. The number of channels of the waveguide increases in the form of 2 to the power of N, where N is a positive integer greater than or equal to 2, and N represents the number of stages of the cascaded Y-shaped waveguide between the single input and the multiple outputs. The weak light detector is a detector array, in which each detector corresponds to one output of the cascaded optical waveguide; The control circuit receives the number of photons detected by each detector within a predetermined emission period and generates a quantum random number based on the number of photons. The control circuit converts the voltage signal received by each detector into a binary digital signal and extracts the low bits of the binary digital signal as a quantum random number.

2. The chip-based quantum random number generator according to claim 1, characterized in that, The spontaneous emission source periodically emits light pulses in the form of pulses into the input of the cascaded optical waveguide.

3. The chip-based quantum random number generator according to claim 1, characterized in that, The low-light detector is a linear array detector, and the exposure integration time of the linear array detector is matched with the emission time of the spontaneous emission source.

4. The chip-based quantum random number generator according to claim 1, characterized in that, The spontaneous emission light source, cascaded optical waveguide, weak light detector, and control circuit are all fabricated on the same silicon wafer, with the spontaneous emission light source and cascaded optical waveguide arranged sequentially to each other.

5. The chip-based quantum random number generator according to claim 1, characterized in that, The control circuit converts the voltage signal received by each detector into an 8-bit binary digital signal and extracts the lower 4 bits of the 8-bit binary digital signal as a quantum random number.

6. A method for generating quantum random numbers, characterized in that, The method is implemented using a chip-based quantum random number generator according to any one of claims 1 to 5, comprising: The spontaneous emission light source emits spontaneous light particles based on the power supply signal; the spectral broadening of the spontaneous emission light source is less than the speed of light divided by the longest optical path distance between the spontaneous emission light source and the weak light detector; The self-emitting electrons are cascaded and beam-splittered via cascaded optical waveguides. Each cascaded optical waveguide has a single input and multiple outputs. The single input receives the self-emitting electrons. A cascaded Y-shaped waveguide connects the single input and the multiple outputs. The number of channels in the waveguide increases in the form of powers of 2, where N is a positive integer greater than or equal to 2, and N represents the number of stages of the cascaded Y-shaped waveguide between the single input and the multiple outputs. The weak light detector is a detector array, in which each detector detects the output photons in the corresponding channel after cascaded beam splitting. The control circuit receives the number of photons detected by each detector within a predetermined emission period and generates a quantum random number based on the number of photons. The control circuit converts the voltage signal received by each detector into a binary digital signal and extracts the low bits of the binary digital signal as a quantum random number.

7. The method according to claim 6, characterized in that, The spontaneous emission light source emits light pulses periodically in pulse form.

8. The method according to claim 6, characterized in that, The voltage signal generated based on the detected photons is converted into a binary digital signal, and the low bits of the binary digital signal are extracted as quantum random numbers.

Images