A true random number generator based on liquid molecular brownian motion

Abstract

The application discloses a true random number generator, which comprises a random number collecting unit and a random number generating unit. The random number collecting unit comprises a high-pressure-resistant closed transparent liquid pool, which contains a mixture of particles in Brownian motion; and a high-speed microscopic imaging device, which periodically takes pictures of a section inside the mixture to obtain multiple frames of pictures. The random number generating unit comprises a digital processor, which processes the multiple frames of pictures to obtain a preliminary random number string A0; and a postprocessor, which processes A0 to obtain a random number sequence.

Description

[0001] This application is a divisional application of Chinese Patent Application No. CN201710326230, entitled "A True Random Number Generator Based on Liquid Molecular Brownian Motion". TECHNICAL FIELD

[0002] The present application relates to a random number generator, and more particularly, to a true random number generator based on liquid molecular Brownian motion. BACKGROUND

[0003] Random numbers play an important role in the fields of cryptography, simulation, integrated circuit testing, etc. A random number sequence is usually a combination of random binary sequences, and can be divided into pseudo-random numbers and true random numbers according to whether it is reproducible and predictable. Pseudo-random numbers are not truly random, and they are usually calculated by various electronic devices such as computers, so once the generation method and some parameters (such as random number seeds) are known, the same pseudo-random numbers can be obtained, thereby facing the risk of being deciphered. Therefore, in order to ensure the security of communication, a generator capable of generating true random numbers is needed.

[0004] True random numbers can usually only be obtained through physical phenomena. Common entropy sources of true random numbers include, but are not limited to, circuit noise, cosmic noise, radioactive decay, etc. Current true random number generators usually generate true random numbers based on circuit noise, etc., but the noise is usually small and difficult to directly and accurately measure and quantify. The common practice is to measure it after amplification. However, since the amplifier is not absolutely linear, the randomness of the current signal obtained after amplification of the small noise will inevitably be affected. Other true random number generation methods, such as using a flip-flop in metastable state to generate true random numbers, usually have problems such as insufficient randomness.

[0005] In addition, with the increase of communication rate, there is also a need for high-speed true random number generators to meet the increasing requirements of communication rate. SUMMARY

[0006] The present application provides a high-speed true random number generator, which takes liquid Brownian motion as an entropy source, and provides a high-speed true random number sequence with sufficient randomness by periodically shooting Brownian motion and digitally processing the shot pictures.

[0007] The true random number generator comprises a random number collecting unit and a random number generating unit. The random number collecting unit comprises a high-pressure-resistant closed transparent liquid pool containing a mixture formed by a colorless transparent colloidal liquid and particles doing Brownian motion in the colorless transparent colloidal liquid, and a vacuum sealed space for accommodating the colorless transparent colloidal liquid to expand under high pressure; and a high-speed microscopic imaging device periodically taking internal sections of the mixture to obtain multiple frames of pictures. The random number generating unit comprises a digitizing processor placing each frame of picture obtained by the high-speed microscopic imaging device each time in a plane rectangular coordinate system, drawing a grid, traversing all intersection points in the grid and analyzing the gray scale around the intersection points to obtain a random number represented by the intersection point, and thereby obtaining a preliminary random number string A0 corresponding to the each frame of picture; and a post-processor processing the preliminary random number string A0 processed by the digitizing processor to obtain a random number sequence; wherein the high-speed microscopic imaging device is aligned with the middle part of the mixture in the transparent liquid pool on one side of the transparent liquid pool, the opposite side of the high-speed microscopic imaging device has a white gasket attached to the pool wall of the transparent liquid pool, which helps the shooting, and wherein the high-speed microscopic imaging device is electrically connected to the random number generating unit. BRIEF DESCRIPTION OF DRAWINGS

[0008] Figure 1 is a diagram of a high-pressure-resistant closed transparent liquid pool 100 in the random number collecting unit;

[0009] Figure 2 is a diagram of a high-pressure-resistant closed transparent liquid pool 100 in the random number collecting unit, and a mixture 210 and a vacuum sealed space 220 in it.

[0010] Figure 3 is an internal section 300 of the mixture 210 in the closed transparent liquid pool 100 periodically taken by the high-speed microscopic imaging device in the random number collecting unit.

[0011] Figure 4 is a diagram 400 of placing the middle part of a frame of picture taken by the high-speed microscopic imaging device in a plane rectangular coordinate system for digitizing processing to obtain a preliminary random number string A0 by a digitizing processor in the random number generating unit.

[0012] Figure 5 is a process diagram 500 of processing the preliminary random number string A0 by a post-processor in the random number generating unit to generate a final true random number sequence.

[0013] Figure 6 is a whole diagram 600 of the true random number generator.

[0014] Those skilled in the art will appreciate that, for clarity, not every device is shown in the drawings. Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and not of limitation. DETAILED DESCRIPTION

[0015] In order to provide a high-speed and sufficiently random true random number sequence, the present application provides a high-speed true random number generator that uses liquid Brownian motion as an entropy source, and provides a high-speed true random number sequence by periodically capturing Brownian motion and digitally processing the captured images. The present application will be described in detail below with reference to the accompanying drawings.

[0016] Figure 1 is a diagram of a high-pressure resistant closed transparent liquid cell 100 in a random number collection unit. In Figure 1 , the liquid cell 100 is empty and in an unsealed state. The liquid cell 100 is transparent to facilitate capturing by a high-speed microscopic imaging device. In addition, although the transparent liquid cell is shown as having a certain shape in Figure 1 , those skilled in the art will appreciate that the transparent liquid cell can be any kind and shape of transparent liquid cell that achieves the purpose of the present application.

[0017] Figure 2 is a diagram of a high-pressure resistant closed transparent liquid cell 100 and a mixture 210 therein in a random number collection unit. In this diagram, the liquid cell 100 contains a mixture 210 of a colorless transparent gelatinous liquid and particles to a target scale, the particles performing Brownian motion in the colorless transparent gelatinous liquid. The example is achieved by adding the particles to half of the target scale in the empty liquid cell 100 shown in Figure 1 , and then adding the gelatinous liquid to the target scale in the liquid cell. The step of first adding the particles to half of the target scale and then adding the gelatinous liquid to the target scale is to make the volume of the particles in the closed liquid cell equal to half of the sum of the volumes of the particles and the gelatinous liquid, so that the chances of any point in the mixture being occupied by the particles are equal.

[0018] In the example shown in Figure 2 , after the mixture 210 reaches the target scale, the transparent liquid cell 100 is evacuated and sealed. In the above example, the target scale is lower than the height of the closed liquid cell, and the portion of the closed liquid cell 100 above the target scale is a vacuum sealed space 220 for accommodating expansion of the colorless transparent gelatinous liquid under high pressure.

[0019] In the above example, the particles can have the same density as the colorless transparent gummy liquid. The particles can also be black water-insoluble particles having a diameter of 0.01-0.1 mm. For example, the black water-insoluble particles can be diamond dust obtained, for example, by sieving.

[0020] The random number collection unit can also include a thermometer that can measure the temperature of the mixture inside the transparent liquid cell. The thermometer can be built-in or wireless.

[0021] The random number collection unit can also include a heater that can heat the closed transparent liquid cell and the mixture inside it so that the mixture inside keeps a target temperature that is lower than the boiling point of the colorless transparent gummy liquid.

[0022] In addition, the random number collection unit includes a high-speed microscopic imaging device that periodically takes pictures of the internal section of the mixture inside the closed transparent liquid cell to obtain multiple frames of pictures.

[0023] The closed transparent liquid cell and the mixture inside it, the thermometer, the heater and the high-speed microscopic imaging device described above constitute the random number collection unit of the true random number generator. Although for the sake of clarity, Figure 1 and Figure 2 The thermometer, the heater and the high-speed microscopic imaging device are not shown in the above-mentioned figures, but it is understood by those skilled in the art that, for the purposes of the present application, the true random number generator can include any type of thermometer, heater and high-speed microscopic imaging device described above that can achieve the purposes of the present application.

[0024] With reference to Figure 3 , Figure 3is an internal section 300 of the mixture in the closed transparent liquid pool periodically photographed by the high-speed microscopic imaging device in the random number collection unit. The high-speed microscopic imaging device can periodically photograph the internal section 300 of the mixture 210, and each time of photographing can obtain multiple pictures. The photographing by the high-speed microscopic imaging device can be high-speed to help the true random number generator to quickly generate a true random number sequence. For example, the high-speed microscopic imaging device can photograph once every N seconds, where N < 0.1. The speed of photographing by the high-speed microscopic imaging device can be adjusted based on the requirement of the communication rate to meet the requirement of the increasing communication rate. The high-speed microscopic imaging device can be aligned to the middle part of the mixture in the transparent liquid pool on one side of the transparent liquid pool. The opposite side of the high-speed microscopic imaging device can have a white gasket attached to the pool wall of the transparent liquid pool to avoid irrelevant objects being photographed by the high-speed microscopic imaging device. The high-speed microscopic imaging device can also be electrically connected to the random number generation unit to transmit the photographed picture data to the random number generation unit for processing.

[0025] Figure 4 is a diagram 400 of the photographed picture in which the middle part is selected after removing the surrounding part and placed in a planar rectangular coordinate system for digital processing to obtain a preliminary random number sequence A0by a digital processor in the random number generation unit. The middle part is selected after removing the surrounding part of the picture to exclude the influence of the pool wall on the Brownian motion of the liquid molecules, so as to ensure that the motion of the molecules in the picture part for digital processing is truly random.

[0026] In the present application, the device for performing the digital processing can be referred to as a digital processor (not shown). The digital processor can perform digital processing on each of the multiple pictures obtained by the high-speed microscopic imaging device each time of photographing to obtain a preliminary random number sequence A0corresponding to the each picture by the following steps: placing the middle part of the each picture selected after removing the surrounding part in a planar rectangular coordinate system, drawing a grid with a spacing of 0.1 mm, and selecting the intersection points of the grid as the i-th value point x i of the random number sequence A0(i > 0, i ∈ N); setting the initial values of the x i to be all 0; repeatedly traversing all the intersection points in the selected middle part and analyzing the gray scale within a range of 0.01 mm in diameter around the intersection point, and if the gray scale is greater than 50%, setting the x i= 1. Those skilled in the art should understand that the non-repetitive traversal of all the intersection points in the selected intermediate portion can be performed in any order that can non-repetitively traverse all the intersection points. After the above operation, a preliminary random number string A0 corresponding to each frame of picture is obtained.

[0027] Subsequently, the preliminary random number string A0 can be post-processed. In the present application, the device performing the post-processing can be referred to as a post-processor (not shown). The digital processor described above and the post-processor constitute a random number generation unit in the true random number generator described in the present application. The digital processor and the post-processor can be various commercially available computer processors or microprocessors that can perform the functions described in the present application, such as INTEL Core i5 series or i7 series processors, or STM32 series single-chip microprocessor.

[0028] Figure 5 is a process diagram 500 of processing the preliminary random number string A0 by the post-processor in the random number generation unit to generate a final true random number sequence. The post-processor can process the preliminary random number string A0 by using von Neumann correction on the random number string A0, discarding the data string A1' with no output, generating a random number sequence A1; performing XOR chain correction on the discarded data string A1' after discarding the first number to obtain a random number sequence A2; and splicing the random number sequence A2 to the random number sequence A1 to obtain a random number sequence A3 corresponding to each frame of picture. The number of stages of the XOR chain can depend on the random number requirement. For example, if higher uniformity and randomness are required, the number of stages can be increased to 8, and if the efficiency of random number generation is required to be higher, the number of stages can be appropriately reduced.

[0029] After the above processing of the post-processor, a random number sequence A3 corresponding to each frame of picture is obtained. However, the high-speed microscopic imaging device can obtain multiple frames of picture for each shooting, and each frame of picture corresponds to a random number sequence A3, so the post-processor can further splice multiple random number sequences A3 corresponding to multiple frames of picture obtained for each shooting to obtain a random number sequence A4 corresponding to each shooting. Those skilled in the art should understand that the splicing is performed in any order that can non-repetitively traverse each frame of picture in the multiple frames of picture obtained for each shooting.

[0030] In addition, as described above, the photographing of the internal cross section of the mixture by the high-speed microscopic imaging device can be performed periodically, and each time the photographing is performed, a plurality of pictures can be obtained. The photographing performed by the high-speed microscopic imaging device can be high-speed photographing, so as to help the true random number generator to quickly generate a true random number sequence. For example, the high-speed microscopic imaging device can perform photographing once every N seconds, where N < 0.1. Therefore, the post-processor can further splice a plurality of random number sequences A4 corresponding to each photographing in the periodic photographing to obtain a random number sequence A5. It should be understood by those skilled in the art that the splicing is performed in any order capable of non-repeatedly traversing each photographing in the periodic photographing.

[0031] In addition, in order to further improve the randomness of the random number sequence A5 obtained above, the post-processor can further perform XOR periodic sequence correction on the random number sequence A5 to obtain a random number sequence A6. The random number sequence A6 is the final output of the true random number generator described in the present application.

[0032] Figure 6 is a whole diagram of the true random number generator 600. In the diagram, it can be seen that the opposite side of the high-speed microscopic imaging device 620 has a white gasket 650 attached to the wall of the transparent liquid pool 100, so as to avoid irrelevant objects being photographed by the high-speed microscopic imaging device; the high-speed microscopic imaging device 620 can also be electrically connected to the random number generation unit 610, so as to transmit the picture data obtained by photographing to the random number generation unit for processing. As shown, there can be 2 or more high-speed microscopic imaging devices 620, so as to double the random number production efficiency. The whole diagram of the true random number generator 600 also shows a thermometer 630 and a heater 640, which operate as described above with reference to the Figure 2

[0033] The true random number generator described in the present application can quickly generate a true random number sequence to meet the requirement of continuously increasing communication speed, and the true random number sequence generated by the true random number generator takes Brownian motion of liquid molecules as an entropy source and is corrected in the process of generating the final output random number sequence, so as to overcome problems such as insufficient randomness, while the digital loss is small and the random number production efficiency is high (which can reach 5-10 GBPS).​

Claims

1. A true random number generator (600), comprising: a random number collecting unit and a random number generating unit (610), the random number collecting unit comprises: a high-pressure-resistant closed transparent liquid pool (100) containing a mixture (210) formed by a colorless transparent colloidal liquid and particles performing Brownian motion in the colorless transparent colloidal liquid, and a vacuum sealed space (220) for accommodating expansion of the colorless transparent colloidal liquid under high pressure; a high-speed microscopic imaging device (620) periodically capturing cross sections inside the mixture to obtain multiple frames of pictures; the random number generating unit (610) comprises: a digitizing processor placing each of the multiple frames of pictures obtained by each capture of the high-speed microscopic imaging device in a plane rectangular coordinate system, drawing a grid, traversing all intersection points in the grid and analyzing the grayscale around the intersection points to obtain random numbers represented by the intersection points, and further obtaining a preliminary random number string A0 corresponding to each frame of picture; a post-processor processing the preliminary random number string A0 processed by the digitizing processor to obtain a random number sequence; the digitizing processor is configured to: digitize each of the multiple frames of pictures obtained by each capture of the high-speed microscopic imaging device to obtain a preliminary random number string A0 corresponding to each frame of picture: place the middle part of each frame of picture after removing the surrounding part in a plane rectangular coordinate system, draw a grid with a spacing of 0.1 mm, and select the intersection points of the grid as the positions of the i-th value point xi (i>0, i∈N) in the random number string A0; set the initial values of the xi to be all 0; non-repeatedly traverse all intersection points in the selected middle part and analyze the grayscale within a range of 0.01 mm in diameter around the intersection points, and if the grayscale is greater than 50%, set the xi=1; wherein the non-repeatedly traversing all intersection points in the selected middle part is performed in any order that can non-repeatedly traverse all the intersection points; the post-processor is configured to: A von Neumann correction is applied to the random number string A0, and the data string A with no output is discarded 1′ to generate a random number string A1; A discarded data string A 1′ After the first number is removed, the XOR chain correction is performed to obtain a random number sequence A2; splice the random number sequence A2 to the random number sequence A1 to obtain a random number sequence A3 corresponding to each frame of picture; wherein the number of stages of the XOR chain depends on the random number demand.

2. The true random number generator (600) according to claim 1, wherein the random number collecting unit further comprises: a thermometer (630) measuring the temperature of the mixture (210) inside the closed transparent liquid pool (100), wherein the thermometer (630) is built-in or wireless; a heater (640) heating the closed transparent liquid pool (100) and the mixture (210) inside it, so that the internal mixture (210) maintains a target temperature lower than the boiling point of the colorless transparent colloidal liquid.

3. The true random number generator (600) according to claim 1, wherein the volume of the particles in the closed transparent liquid pool (100) in the random number collecting unit is equal to half the sum of the volumes of the particles and the colloidal liquid, so that the chances of any point in the mixture (210) being occupied by the particles are equal.

4. The true random number generator (600) of claim 1, wherein, The particles in Brownian motion in the liquid pool of the random number collection unit are black water-insoluble particles with a diameter of 0.01-0.1 mm and the same density as the colorless transparent gelatinous liquid.

5. The true random number generator (600) according to claim 4, wherein The black water-insoluble particles are corundum dust.

6. The true random number generator (600) of claim 1, wherein, The shooting cycle of the high-speed microscopic imaging device (620) is N seconds, where N < 0.

1.

7. The true random number generator (600) according to claim 1, wherein The postprocessor is further configured to: The postprocessor is further configured to process the preliminary random number string A0 by: splicing multiple random number sequences A3 corresponding to multiple pictures obtained in each shooting to obtain a random number sequence A4 corresponding to each shooting; wherein the splicing is performed in any order capable of non-repeatedly traversing each picture in multiple pictures obtained in each shooting.

8. The true random number generator (600) of claim 1, wherein, The postprocessor is further configured to: The postprocessor is further configured to process the preliminary random number string A0 by: splicing multiple random number sequences A4 corresponding to each shooting in periodic shooting to obtain a random number sequence A5; wherein the splicing is performed in any order capable of non-repeatedly traversing each shooting in the periodic shooting.

9. The true random number generator (600) according to claim 8, wherein The postprocessor is further configured to: The postprocessor is further configured to process the preliminary random number string A0 by: XORing the random number sequence A5 with a periodic sequence to obtain a random number sequence A6.

10. The true random number generator (600) of claim 2, wherein, The random number collection unit is further configured to: adding the particles to half of a target scale in the empty liquid pool, and then adding the gelatinous liquid to the target scale in the liquid pool, so that the volume of the particles in the closed liquid pool (100) in the random number collection unit is equal to half of the sum of the volume of the particles and the volume of the gelatinous liquid; subsequently, vacuumizing the closed liquid pool and sealing it; wherein the target scale is lower than the height of the closed liquid pool (100), and the part of the closed liquid pool (100) higher than the target scale is the vacuum sealed space (220) used to accommodate the colorless transparent gelatinous liquid expanded under high pressure.

Images