Superconducting Physical Random Number Generator

Abstract

To provide a superconductivity physical random number generating device capable of adjusting an operation point of a superconductivity physical random number generator to an optimum operation point.SOLUTION: A superconductivity physical random number generating device includes a superconductivity physical random number generator for generating a physical random number, and a toggle type flip-flop using a superconductive circuit, inputs each of branch signals in which a common clock signal is branched into each of the superconductivity physical random number generator and the toggle type flip-flop, and inputs each of branch currents in which a common constant current is branched into the superconductivity physical random number generator as an operation point control current, and into the toggle type flip-flop as a bias current.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to a superconducting physical random number generator. [Background technology]

[0002] Random numbers are used in various fields, including information and communications. There are two types of random numbers: pseudo-random numbers, which are generated based on an algorithm, and physical random numbers (true random numbers), which are generated using an essentially random physical phenomenon. When physical random numbers are generated, unpredictable physical phenomena that are the source of randomness include noise, jitter, or quantum measurement.

[0003] Although pseudorandom numbers can be generated easily and quickly by a normal computer, the generated pseudorandom number sequence always has periodicity, and if the algorithm used is identified, the pseudorandom number sequence can be predicted. In order to generate a good pseudorandom number sequence quickly, a lot of computing resources and a lot of computing time are required.

[0004] In contrast, physical random numbers generated by dedicated elements have no periodicity or correlation between bits and are unpredictable, but because the generation principle is based on weak physical phenomena, it can be difficult to generate physical random numbers quickly. A known example of a random number generator that generates physical random numbers is a superconducting random number generator that uses a superconducting circuit to exploit the properties of a superconductor. A superconducting random number generator can achieve high random number generation speed while reducing the power required to generate physical random numbers.

[0005] Non-Patent Document 1 describes the frequency synchronization phenomenon of Josephson oscillators. Specifically, Non-Patent Document 1 describes a technique for completely matching the frequencies of the clocks output from multiple superconducting single flux quantum clock generating circuits by biasing the multiple superconducting single flux quantum clock generating circuits with a common current source (constant current) without using an on-chip resistor (see Non-Patent Document 1). Such frequency synchronization is based on Faraday's law of electromagnetic induction, which is a basic property of quantized magnetic flux (magnetic flux quantum), which is the information carrier of superconducting circuits. [Prior art documents] [Non-patent literature]

[0006] [Non-Patent Document 1] Yuki Yamanashi, Ryo Kinoshita, and Nobuyuki Yoshikawa, "Frequency Synchronization Phenomenon of Josephson Oscillators," 68th Spring Meeting of the Japan Society of Applied Physics, March 18, 2021 Summary of the Invention [Problem to be solved by the invention]

[0007] However, in general, when a superconducting physical random number generator is used, it is necessary to control the operating point of the superconducting physical random number generator extremely accurately to an operating point where the output probability of each of "0" and "1" is 0.5 (i.e., 50%). However, in reality, the operating point changes over time due to deviations in the operating point caused by noise or power supply fluctuations (e.g., voltage fluctuations), making it difficult to generate high-quality physical random numbers stably and quickly for a long period of time. In such cases, the quality or generation speed of physical random numbers generated by the superconducting physical random number generator decreases.

[0008] The present disclosure has been made in consideration of these circumstances, and has an objective of providing a superconducting physical random number generator that can adjust the operating point of a superconducting physical random number generator to an optimal operating point. [Means for solving the problem]

[0009] One aspect of the present disclosure is a superconducting physical random number generating device comprising a superconducting physical random number generator that generates physical random numbers and a toggle flip-flop using a superconducting circuit, in which each branch signal branched from a common clock signal is input to the superconducting physical random number generator and the toggle flip-flop, respectively, and each branch current branched from a common constant current is input to the superconducting physical random number generator as an operating point control current and to the toggle flip-flop as a bias current. Effect of the Invention

[0010] According to the present disclosure, in a superconducting physical random number generator, the operating point of the superconducting physical random number generator can be adjusted to an optimal operating point. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a diagram showing an example of a schematic configuration of a superconducting physical random number generator according to an embodiment (first embodiment). [Diagram 2] FIG. 2 is a diagram illustrating an example of a configuration (equivalent circuit) of a random number generator according to the embodiment. [Diagram 3] 1 is a diagram showing an example of a characteristic showing a relationship between an operating point control current and an output probability in a random number generator according to an embodiment. FIG. [Figure 4] FIG. 4 is a diagram illustrating an example of a clock signal input to a random number generator according to an embodiment. [Diagram 5] FIG. 2 is a diagram showing an example of a physical random number signal output from a random number generator according to an embodiment. [Figure 6] FIG. 11 is a diagram showing an example of frequency synchronization (immediately after starting operation) according to the embodiment; [Figure 7] FIG. 2 is a diagram illustrating an example of frequency synchronization (steady state) according to the embodiment. [Figure 8] FIG. 13 is a diagram showing an example of a simulation result of a physical random number output from a random number generator according to an embodiment. [Figure 9]FIG. 1 is a diagram showing an example of a schematic configuration of a superconducting physical random number generator according to an embodiment (second embodiment). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

[0013] (First embodiment) A first embodiment will be described.

[0014] [Superconducting physical random number generator] FIG. 1 is a diagram showing an example of a schematic configuration of a superconducting physical random number generator 1 according to an embodiment (first embodiment). The superconducting physical random number generator 1 includes a random number generator 11, which is a generator of physical random numbers using superconducting properties (superconducting physical random number generator), a TFF 12, which is a toggle-type flip-flop using superconducting properties, and a Josephson Transmission Line (JTL) 13.

[0015] An input terminal (not shown) of a clock signal a1 in the superconducting physical random number generator 1 is connected to a clock input terminal of the random number generator 11 and a clock input terminal of the TFF 12, respectively. TFF12 and JTL13 are connected. An input terminal (not shown) of a constant current b1 in the superconducting physical random number generator 1 is connected to an operating point control current input terminal of the random number generator 11. The input terminal (not shown) of this constant current b1 is also connected to a bias current input terminal of the TFF 12 via a JTL 13. In this embodiment, the JTL 13 is used as a delay transmission line. The constant current b1 is supplied, for example, from a constant current source (not shown). A physical random number output terminal of the random number generator 11 is connected to an output terminal (not shown) of the physical random number in the superconducting physical random number generator 1.

[0016] A clock signal a1 supplied from a superconducting oscillator circuit (not shown), which is a clock generator using superconducting characteristics, is input to a superconducting physical random number generator 1 and then branched into two, with the clock signal of one branch (one branch signal) being input to a random number generator 11 and the clock signal of the other branch (the other branch signal) being input to a TFF 12. These clock signals are composed of a pulse train c1 such as, for example, "111111...". In this embodiment, the clock signal a1 supplied from a superconducting oscillator circuit (not shown) is a common clock signal for the random number generator 11 and the TFF 12. In this embodiment, the frequency of the clock signal a1 is described as f [Hz]. f is a positive value and is not particularly limited. As the superconducting oscillator circuit (not shown) that is the clock generator, for example, a ring oscillator circuit using superconducting characteristics may be used.

[0017] The TFF 12 generates a superconducting single flux quantum (SFQ) pulse train c2 based on the input clock signal. In the example of FIG. 1, a configuration example in which the SFQ pulse train c2 is transmitted between the TFF 12 and the JTL 13 is shown. Here, the TFF 12 has a function of a 1 / 2 frequency divider and divides the clock signal by 1 / 2. The SFQ pulse train c2 is composed of a pulse train such as "101010...". The frequency of the SFQ pulse train c2 is f / 2 [Hz].

[0018] A constant current b1 supplied from a constant current source (not shown) is input to the superconducting physical random number generator 1 and then branched into two, with the current of one branch (one branch current) being input to the random number generator 11 as an operating point control current b2, and the current of the other branch (the other branch current) being input to the TFF 12 via the JTL 13 as a bias current b3. In this embodiment, the constant current source and the constant current b1 are common to the random number generator 11 and the TFF 12. In the random number generator 11, the operating point of the random number generator 11 is controlled by the operating point control current b2. TFF12 is biased by bias current b3.

[0019] The random number generator 11 generates and outputs physical random numbers based on a pulse train c1 based on a clock signal a1 and an operating point control current b2. As a result, a physical random number sequence c3, which is a sequence of the physical random numbers, is output from the superconducting physical random number generator 1. In this embodiment, the physical random number sequence c3 is composed of a pulse train of random physical random numbers such as "101001...". In this embodiment, the average frequency of the physical random number sequence c3 is f / 2 [Hz].

[0020] In this embodiment, for ease of explanation, a device (superconducting physical random number generator 1) that integrates the components shown in Figure 1 (random number generator 11, TFF 12, and JTL 13) will be described. However, the superconducting physical random number generator 1 does not necessarily have to be configured to house the components shown in Figure 1 in a single housing, and also includes a configuration in which the components shown in Figure 1 are each arranged and connected at necessary locations.

[0021] In addition, the superconducting physical random number generator 1 may include other components, such as one or both of a superconducting oscillator circuit (not shown), which is a clock generator that uses superconducting properties, and a constant current source (not shown) that supplies a constant current b1.

[0022] <Random number generator> FIG. 2 is a diagram showing an example of a configuration (equivalent circuit) of the random number generator 11 according to the embodiment. The random number generator 11 has a configuration in which a thermal noise source is directly connected to a comparator using an SFQ element. The random number generator 11 generates physical random numbers by utilizing the thermal noise current of a resistor (for example, resistor 153 in the example of FIG. 2). The random number generator 11 according to this embodiment is a circuit that utilizes the characteristics of a superconducting single flux quantum circuit, which uses the presence or absence of a flux quantum in a superconducting ring to represent "1" and "0".

[0023] The random number generator 11 includes a coil 111, a coil 112, a coil 113, a Josephson junction 132 and a resistor 152, and a coil 114 between an input terminal T1 and an output terminal T2. The random number generator 11 also includes a Josephson junction 131 and a resistor 151 between a point between the coils 111 and 112 and the ground terminal G1. The random number generator 11 also includes a Josephson junction 133 and a resistor 153 between a point between the Josephson junction 132 and the coil 114 and the ground terminal G1. The random number generator 11 also includes a constant current source 171 connected between the coil 112 and the coil 113. In this embodiment, the constant current source 171 supplies a constant current of 133 [μA] to a point between the coil 112 and the coil 113.

[0024] Here, each of the coils 111 to 114 may be called an inductor. Also, the Josephson junction 131 and the resistor 151 may be an integral circuit part. Similarly, the Josephson junction 132 and the resistor 152 may be an integral circuit part, and the Josephson junction 133 and the resistor 153 may be an integral circuit part.

[0025] In the example of Figure 2, a pulse sequence c1 based on the clock signal a1 shown in Figure 1 is input to the input terminal T1 (clock input terminal) of the random number generator 11, a physical random number sequence c3 shown in Figure 1 is output from the output terminal T2 (physical random number output terminal) of the random number generator 11, and an operating point control current b2 (Ictl shown in Figure 2) shown in Figure 1 is input to the connection point of the Josephson junction 133 of the random number generator 11.

[0026] The configuration of the random number generator 11 shown in FIG. 2 is well known, and a detailed description thereof will be omitted in this embodiment. Furthermore, the random number generator 11 is not necessarily limited to the example in FIG. 2, and other configurations may be used.

[0027] [Random number generator operating point] FIG. 3 is a diagram showing an example of a characteristic 1011 representing the relationship between the operating point control current and the output probability in the random number generator 11 according to the embodiment. In the graph shown in FIG. 3, the horizontal axis represents the operating point control current [μA], and the vertical axis represents the output probability. The graph also shows a characteristic 1011 that indicates the relationship between the operating point control current and the output probability. In this embodiment, the operating point of the random number generator 11 in the superconducting physical random number generator 1 is automatically controlled to a point where the output probability of "1" out of "0" and "1" is 0.5 (i.e., 50%). As a result, in this embodiment, the random number generator 11 in the superconducting physical random number generator 1 automatically generates good quality physical random numbers.

[0028] [Example of clock input and physical random number output of a random number generator] FIG. 4 is a diagram showing an example of a clock signal 1021 input to the random number generator 11 according to the embodiment. In the graph shown in FIG. 4, the horizontal axis represents time [ps], and the vertical axis represents voltage [μV]. Also shown in the graph is a clock signal 1021 . In the example of FIG. 4, clock signal 1021 is composed of a pulse train such as "1111111111..." when converted into a digital value (a pulse train equivalent to pulse train c1 shown in FIG. 1).

[0029] FIG. 5 is a diagram showing an example of a physical random number signal 1031 output from the random number generator 11 according to the embodiment. In the graph shown in FIG. 5, the horizontal axis represents time [ps], and the vertical axis represents voltage [μV]. In addition, a physical random number signal 1031 is shown in the graph. 5, the physical random number signal 1031 is composed of a pulse train such as "1010011001..." when converted into a digital value (a pulse train equivalent to the physical random number train c3 shown in FIG. 1). Note that in this embodiment, "1" and "0" are each output with a 50% probability in the physical random number signal 1031, and the sequences of "1" and "0" shown in FIG. 1 and FIG. 5 are examples for explanation purposes.

[0030] [Frequency synchronization] An example of frequency synchronization will be shown with reference to FIG. 6 and FIG. The principle of frequency synchronization is similar to the technology described in Non-Patent Document 1, and a detailed description thereof will be omitted in this embodiment.

[0031] FIG. 6 is a diagram showing an example of a state of frequency synchronization (immediately after the start of operation) according to the embodiment. FIG. 6 shows a random number generator 11 and a TFF 12. Immediately after the superconducting physical random number generator 1 starts operating, for example, a constant current of (2×Ib) magnitude (current corresponding to constant current b1 shown in FIG. 1) is divided almost equally, and a branch current of about Ib magnitude is input to the random number generator 11 as an operating point control current (operating point control current b2 shown in FIG. 1), and a branch current of about Ib magnitude is input to the TFF 12 as a bias current (bias current b3 shown in FIG. 1). Here, Ib represents a predetermined current amount (current magnitude) and is a positive value. In the example of FIG. 6, the delay transmission line JTL 13 is omitted from the illustration.

[0032] In the example of FIG. 6, immediately after the start of operation, the random number generator 11 outputs a pulse train 2011 with an average frequency f1, and the TFF 12 outputs a pulse train 2012 with a frequency f2. In this example, the case is shown where the average frequency f1 and the frequency f2 are different immediately after the start of operation. In this embodiment, the frequency f2 of the TFF 12 is controlled to f / 2.

[0033] As shown in FIG. 6, immediately after the start of operation, the output average voltage (f1·Φ0) of the random number generator 11 and the output average voltage (f2·Φ0) of the TFF 12 are different. Here, Φ0 is the magnetic flux quantum (2.068×10 -15 [Wb]) and serves as the information carrier in superconducting circuits. Furthermore, the output average voltage is based on the law of electromagnetic induction.

[0034] FIG. 7 is a diagram showing an example of frequency synchronization (steady state) according to the embodiment. FIG. 7 shows the state in FIG. 6 (the state immediately after the start of operation) when a steady state is reached after a certain period of time has elapsed.

[0035] FIG. 7 shows a random number generator 11 and a TFF 12. In the steady state of the superconducting physical random number generator 1, for example, a constant current of (2×Ib) (current corresponding to the constant current b1 shown in FIG. 1) is branched into two at a predetermined unequal ratio, and a branch current in one branch that is greater than Ib (or a branch current smaller than Ib) is input to the random number generator 11 as an operating point control current (operating point control current b2 shown in FIG. 1), and a branch current in the other branch that is smaller than Ib (or a branch current larger than Ib) is input to the TFF 12 as a bias current (bias current b3 shown in FIG. 1). In the example of FIG. 7, the delay transmission line JTL 13 is omitted from the illustration.

[0036] In the example of FIG. 7, in the steady state, the random number generator 11 outputs a pulse train 2111 having an average frequency fsync, and the TFF 12 outputs a pulse train 2112 having a frequency fsync. Here, in this example, in a steady state, the average frequency fsync of the physical random number sequence output from the random number generator 11 and the frequency fsync of the pulse sequence output from the TFF 12 are adjusted to the same frequency by frequency synchronization. Furthermore, in this embodiment, the frequency fsync of the TFF 12 is controlled to f / 2, that is, the same as the frequency f2 shown in Fig. 6. As a result, in the steady state, the average frequency fsync of the physical random number sequence output from the random number generator 11 becomes f / 2.

[0037] 7, in the steady state, the output average voltage (fsync·Φ0) of the random number generator 11 matches the output average voltage (fsync·Φ0) of the TFF 12. Thus, in the steady state, the average frequency fsync of the output of the random number generator 11 matches the frequency fsync of the output of the TFF 12. That is, in this embodiment, since the frequency fsync of the output of the TFF 12 is f / 2, the average frequency fsync of the output of the random number generator 11 is also f / 2.

[0038] In this manner, in a situation where a constant current b1 is supplied and branched into an operating point control current b2 to the random number generator 11 and a bias current b3 to the TFF 12, the operating point control current b2 to the random number generator 11 and the bias current b3 to the TFF 12 are self-adjusted so that the output average voltage of the random number generator 11 and the output average voltage of the TFF 12 become equal, thereby self-adjusting the operating point of the random number generator 11 to an optimal operating point. Here, the output average voltage is proportional to the operating frequency.

[0039] In this case, the number of "1"s in the physical random number output from random number generator 11 is f / 2 [times] per second on average. At this time, in random number generator 11, the magnetic flux output is (f / 2)·Φ0 [Wb] per second, and the average output voltage is (f / 2)·Φ0 [V]. In this case, the frequency of the output of "1" in TFF12 is f / 2 [Hz]. The random number generator 11 and the TFF 12 are frequency synchronized.

[0040] [Example of frequency synchronization simulation results] FIG. 8 is a diagram showing an example of a simulation result of a physical random number output from the random number generator 11 according to the embodiment. In the graph shown in FIG. 8, the horizontal axis represents time, and the vertical axis represents voltage. The graph also shows characteristics 1111 of the simulation results of the physical random number output from the random number generator 11.

[0041] As shown in Figure 8, during period P1 during initial operating point adjustment, the physical random number output from random number generator 11 is not random (good quality physical random numbers), but during steady state period P2 after operating point self-adjustment is completed by frequency synchronization, the physical random number output from random number generator 11 becomes random (good quality physical random numbers).

[0042] [Regarding the first embodiment] As described above, in the superconducting physical random number generator 1 according to this embodiment, the operating point of the superconducting physical random number generator (random number generator 11) can be adjusted to an optimal operating point. Furthermore, in the superconducting physical random number generator 1 according to this embodiment, the circuit area can be reduced by using a superconducting physical random number generator (random number generator 11).

[0043] The superconducting physical random number generator 1 of this embodiment has the function of self-adjusting the operating point of the random number generator 11 to an operating point with the optimal output probability (0.5) by utilizing frequency synchronization of the superconducting circuit, and can generate, for example, high-quality physical random numbers stably and quickly. In the superconducting physical random number generator 1 of this embodiment, the operating point of the random number generator 11 can be self-adjusted to an optimal operating point without external control, except for the input of, for example, a clock signal a1 and a constant current b1.

[0044] 1 shows a configuration example in which the bias current b3 is input to the TFF 12 via the JTL 13 in the superconducting physical random number generator 1, but in another configuration example, the JTL 13 may not be provided and the bias current b3 may be directly input to the TFF 12. However, in an actual circuit, the JTL 13 is often provided from a practical standpoint. Usually, the longer the transmission line of the JTL 13, the larger the amount of bias current b3 can be. Therefore, it is possible to adjust the amount of bias current b3 by adjusting the length of the JTL 13. This makes it possible to set, for example, the magnitude of the bias current b3 to be approximately the same as the magnitude of the operating point control current b2.

[0045] Note that the technology of Non-Patent Document 1 does not disclose adjusting the operating point of the superconducting physical random number generator (random number generator 11) or using TFF 12 equivalent to a 1 / 2 frequency divider, and differs from the technology of the superconducting physical random number generator 1 of this embodiment.

[0046] Second embodiment A second embodiment will be described. The configuration and operation of the superconducting physical random number generator of this embodiment differs from the configuration and operation of the superconducting physical random number generator of the first embodiment in that multiple superconducting physical random number generators (random number generators) are provided in parallel.

[0047] [Superconducting Physical Random Number Generator] FIG. 9 is a diagram showing an example of a schematic configuration of a superconducting physical random number generator 301 according to an embodiment (second embodiment). The superconducting physical random number generator 301 includes multiple random number generators 311-1 to 311-N (N in the example of FIG. 9, where N is an integer equal to or greater than 2) that are superconducting physical random number generators, a TFF 312 that is a toggle-type flip-flop that uses superconducting properties, and a Josephson transmission line (JTL) 313.

[0048] An input terminal (not shown) of a clock signal a11 in the superconducting physical random number generator 301 is connected to the clock input terminals of the multiple random number generators 311-1 to 311-N and the clock input terminal of the TFF12. TFF312 and JTL313 are connected. An input terminal (not shown) of a constant current b11 in the superconducting physical random number generator 301 is connected to each of the operating point control current input terminals of the multiple random number generators 311-1 to 311-N. In addition, the input terminal (not shown) of this constant current b11 is connected to a bias current input terminal of the TFF312 via a JTL313. In this embodiment, the JTL 313 is used as a delay transmission line. The constant current b11 is supplied from, for example, a constant current source (not shown). The physical random number output terminals of the multiple random number generators 311-1 to 311-N are connected to the multiple physical random number output terminals (not shown) of the superconducting physical random number generator 1, respectively.

[0049] A clock signal a11 supplied from a superconducting oscillator circuit (not shown), which is a clock generator using superconducting properties, is input to a superconducting physical random number generator 301 and then branched (N+1), with the N branched clock signals (N branched signals) being input to each of a plurality of random number generators 311-1 to 311-N, and one branched clock signal (one branched signal) being input to a TFF 312. Here, in this embodiment, the clock signal a11 supplied from a superconducting oscillator circuit (not shown) is a common clock signal for the multiple random number generators 311-1 to 311-N and the TFF 312. In this embodiment, as in the first embodiment, the frequency of the clock signal a11 will be described as f [Hz]. As the superconducting oscillator circuit (not shown) that is the clock generator, for example, a ring oscillator circuit using superconducting characteristics may be used.

[0050] The TFF 312 generates a superconducting single flux quantum (SFQ) pulse train based on the input clock signal. In the example of Fig. 9, a configuration example in which the SFQ pulse train is transmitted between the TFF 312 and the JTL 313 is shown. Here, the TFF 312 has a function of a 1 / 2 frequency divider and divides the frequency of the clock signal by 2. The frequency of the SFQ pulse train becomes f / 2 [Hz].

[0051] A constant current b11 supplied from a constant current source (not shown) is input to a superconducting physical random number generator 301 and then branched into (N+1), with each of the N branch currents (N branch currents) being input to each of the multiple random number generators 311-1 to 311-N as an operating point control current b12, and the current of one branch (one branch current) being input to TFF312 via JTL313 as a bias current b13. In this embodiment, the constant current source and the constant current b11 are common to the multiple random number generators 311-1 to 311-N and the TFF 312. In each of the random number generators 311-1 to 311-N, the operating point of each of the random number generators 311-1 to 311-N is controlled by the operating point control current b12. In TFF312, bias is provided by bias current b13. In this embodiment, for convenience of explanation, the operating point control currents input to each of the multiple random number generators 311-1 to 311-N are collectively referred to as operating point control current b12.

[0052] Each of the random number generators 311-1 to 311-N generates and outputs a physical random number based on a pulse train based on the clock signal a11 and an operating point control current b12. As a result, a physical random number sequence, which is a sequence of the physical random numbers, is output from the superconducting physical random number generator 301. In this embodiment, the average frequency of the physical random number sequence is f / 2 [Hz].

[0053] In this embodiment, for ease of explanation, a device (superconducting physical random number generator 301) that integrates the components shown in Figure 9 (random number generators 311-1 to 311-N, TFF 312, and JTL 313) will be described. However, the superconducting physical random number generator 301 does not necessarily have to be configured to house the components shown in Figure 9 in a single housing, and also includes a configuration in which the components shown in Figure 9 are each arranged and connected at necessary locations.

[0054] In addition, the superconducting physical random number generator 301 may include other components, such as one or both of a superconducting oscillator circuit (not shown), which is a clock generator that uses superconducting properties, and a constant current source (not shown) that supplies a constant current b11.

[0055] [Regarding the second embodiment] As described above, in the superconducting physical random number generator 301 according to this embodiment, the operating points of the multiple superconducting physical random number generators (random number generators 311-1 to 311-N) can be adjusted to optimal operating points. Furthermore, in the superconducting physical random number generator 301 according to this embodiment, the circuit area can be reduced by using the superconducting physical random number generators (random number generators 311-1 to 311-N).

[0056] In the superconducting physical random number generator 301 of this embodiment, even when multiple random number generators 311-1 to 311-N are parallelized, the device has a function of self-adjusting the operating points of the random number generators 311-1 to 311-N to the operating point with the optimal output probability (0.5) by utilizing frequency synchronization of the superconducting circuits, and can generate, for example, good-quality physical random numbers stably and quickly. In the superconducting physical random number generator 301 of this embodiment, for example, except for the input of the clock signal a11 and the constant current b11, the operating points of the random number generators 311-1 to 311-N can be self-adjusted to optimal operating points without external control of the individual random number generators 311-1 to 311-N.

[0057] In the superconducting physical random number generator 301 according to this embodiment, the multiple random number generators 311-1 to 311-N are arranged in parallel, so that physical random numbers can be generated at higher speeds than in the first embodiment. Furthermore, in the superconducting physical random number generator 301 of this embodiment, it is possible to maintain the operating points of multiple random number generators 311-1 to 311-N at optimal operating points by supplying an operating point control current b12 branched off from one constant current b11.

[0058] In the superconducting physical random number generator 301 according to this embodiment, even if there is variation in the characteristics of the individual random number generators 311-1 to 311-N, parallelization of the multiple random number generators 311-1 to 311-N can be easily realized. For example, in the superconducting physical random number generator 301 of this embodiment, increasing the number of random number generators 311-1 to 311-N connected in parallel reduces the effect of deviations in the operating points of the individual random number generators 311-1 to 311-N on the other random number generators, stabilizing the generation of physical random numbers.

[0059] 9 shows a configuration example in which the bias current b13 is input to the TFF312 via the JTL313 in the superconducting physical random number generator 301, but in another configuration example, the JTL313 may not be provided and the bias current b13 may be directly input to the TFF312. However, in an actual circuit, the JTL313 is often provided from a practical standpoint. Usually, the longer the transmission path of the JTL 313, the larger the amount of the bias current b13 can be. Therefore, it is possible to adjust the amount of the bias current b13 by adjusting the length of the JTL 313. As a result, even when a plurality of random number generators 311-1 to 311-N are used, it is also possible to set the magnitude of the bias current b13 to be approximately the same as the magnitude of the operating point control current b12, for example.

[0060] Note that the technology of Non-Patent Document 1 does not disclose adjusting the operating points of multiple superconducting physical random number generators (random number generators 311-1 to 311-N) or using TFF 312 equivalent to a 1 / 2 frequency divider, and differs from the technology of the superconducting physical random number generator 301 of this embodiment.

[0061] [Application examples of the above embodiments] The superconducting physical random number generator 1, 301 according to the above embodiments may be applied to various fields in which random numbers are used (various fields using superconducting circuits), such as key generation for one-time pads in classical cryptographic communications, cryptographic key generation or basis selection in quantum cryptographic communications, optimization using Monte Carlo calculations, various physical simulations, fitting, Fourier transform, machine learning, or probabilistic computing hardware (e.g., stochastic computing circuits), and as a specific example, may be applied to stochastic computing using a multi-channel random number generator.

[0062] [Modifications of the above embodiment] In the above embodiments, a TFF (TFF12 in the first embodiment, TFF312 in the second embodiment) equivalent to a 1 / 2 frequency divider is used as a circuit for frequency synchronization with a random number generator (random number generator 11 in the first embodiment, random number generators 311-1 to 311-N in the second embodiment), but as another configuration example, a circuit that outputs a frequency other than half the frequency (f / 2 in the embodiments) may be used. For example, a configuration for frequency synchronization with a frequency shifted from half the frequency (f / 2 in the embodiments) in practical use may be used.

[0063] [Configuration example according to the above embodiment] As an example configuration, a superconducting physical random number generator (superconducting physical random number generator 1 in the example of FIG. 1, and superconducting physical random number generator 301 in the example of FIG. 9) includes a superconducting physical random number generator (random number generator 11 in the example of FIG. 1, and random number generators 311-1 to 311-N in the example of FIG. 9) that generates physical random numbers, and a toggle-type flip-flop using a superconducting circuit (TFF12 in the example of FIG. 1, and TFF312 in the example of FIG. 9). The superconducting physical random number generator inputs each of the branched signals of a common clock signal (clock signal a1 in the example of Figure 1, and clock signal a11 in the example of Figure 9) into a superconducting physical random number generator and a toggle-type flip-flop, respectively. The superconducting physical random number generator inputs each branch current branched from a common constant current (constant current b1 in the example of Figure 1, and constant current b11 in the example of Figure 9) to the superconducting physical random number generator as an operating point control current (operating point control current b2 in the example of Figure 1, and operating point control current b12 in the example of Figure 9), and also inputs each branch current to a toggle flip-flop as a bias current (bias current b3 in the example of Figure 1, and bias current b13 in the example of Figure 9).

[0064] As one configuration example, the superconducting physical random number generator further includes a Josephson transmission line (JTL13 in the example of FIG. 1, JTL313 in the example of FIG. 9). The superconducting physical random number generator inputs a bias current to a toggle flip-flop via a Josephson transmission line.

[0065] As an example configuration, a superconducting physical random number generator (superconducting physical random number generator 301 in the example of FIG. 9) includes a plurality of superconducting physical random number generators (random number generators 311-1 to 311-N in the example of FIG. 9). The superconducting physical random number generator inputs each of the branched signals of a common clock signal (clock signal a11 in the example of Figure 9) into multiple superconducting physical random number generators and toggle flip-flops (TFF312 in the example of Figure 9). The superconducting physical random number generator inputs each branch current branched from a common constant current (constant current b11 in the example of Figure 9) to each of multiple superconducting physical random number generators as an operating point control current (operating point control current b12 in the example of Figure 9), and also inputs it to a toggle flip-flop as a bias current (bias current b13 in the example of Figure 9).

[0066] Although the embodiments of this disclosure have been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and designs that do not deviate from the gist of this disclosure are also included. [Explanation of symbols]

[0067] 1, 301...Superconducting physical random number generator, 11, 311-1 to 311-N...Superconducting physical random number generator (random number generator), 12, 312...Toggle flip-flop (TFF), 13, 313...Josephson transmission line (JTL), 111 to 114...Coil, 131 to 133...Josephson junction, 151 to 153...Resistor, 171...Constant current source, 1011, 11 11...characteristics, 1021, a1, a11...clock signal, 1031...physical random number signal, 2011, 2012, 2111, 2112, c1...pulse train, G1...ground terminal, P1, P2...period, T1...input terminal, T2...output terminal, b1, b11...constant current, b2, b12...operating point control current, b3, b13...bias current, c2...SFQ pulse train, c3...physical random number sequence

Claims

1. A superconducting physical random number generator for generating physical random numbers; A toggle-type flip-flop using a superconducting circuit, Equipped with inputting each of the branched signals of a common clock signal into the superconducting physical random number generator and the toggle-type flip-flop, Each branch current branched from a common constant current is input to the superconducting physical random number generator as an operating point control current, and is also input to the toggle-type flip-flop as a bias current. Superconducting physical random number generator.

2. Furthermore, a Josephson transmission line is provided, The bias current is input to the toggle flip-flop through the Josephson transmission line.

2. The superconducting physical random number generator according to claim 1.

3. A plurality of the superconducting physical random number generators are provided, inputting each of the branched signals of the common clock signal to each of the plurality of superconducting physical random number generators and the toggle-type flip-flops; Each of the branch currents branched from the common constant current is input to each of the plurality of superconducting physical random number generators as the operating point control current, and is also input to the toggle-type flip-flop as the bias current.

3. The superconducting physical random number generator according to claim 1 or 2.

Images