CN-122027115-A - Parallel random number generation method and device based on multi-transverse mode laser
Abstract
The application provides a parallel random number generation method and device based on a multi-transverse mode laser, and relates to the technical field of random number application. The method comprises the steps of generating parallel and independent multipath chaotic signals through a multi-transverse mode laser, and performing signal processing on the multipath chaotic signals to obtain a random number sequence, wherein the random number sequence comprises a plurality of parallel random numbers. The device comprises a multi-transverse mode laser and a processing component, wherein the multi-transverse mode laser is used for generating parallel and independent multi-channel chaotic signals, and the processing component is used for carrying out signal processing on the multi-channel chaotic signals to obtain a random number sequence, wherein the random number sequence comprises a plurality of parallel random numbers. The application utilizes the multi-transverse mode characteristic of the multi-transverse mode laser, generates space-time chaos by introducing the gain competition of the multi-transverse modes, constructs parallel independent chaos sources from space dimension, thereby obtaining a chaos signal, and samples the chaos signal and processes corresponding digital signals to finish the generation of multi-channel and parallel random numbers.
Inventors
- LUO XIANGANG
- CHEN QIANG
- ZHANG FEIFEI
- TANG XI
- XU MINGFENG
- PU MINGBO
- ZHANG YIQUN
Assignees
- 中国科学院光电技术研究所
Dates
- Publication Date
- 20260512
- Application Date
- 20260212
Claims (10)
- 1. A parallel random number generation method based on a multi-transverse mode laser, the method comprising: generating parallel and independent multipath chaotic signals through a multi-transverse-mode laser; Carrying out signal processing on the plurality of chaotic signals to obtain a random number sequence; wherein the random number sequence comprises a plurality of parallel random numbers.
- 2. The method of claim 1, wherein the performing signal processing on the plurality of chaotic signals to obtain a random number sequence comprises: preprocessing a plurality of chaotic signals to obtain digital signals; Carrying out optimization processing on the digital signal to obtain the random number sequence; The optimization processing comprises filtering processing and nonlinear probability distribution shaping processing.
- 3. The method of claim 2, wherein the preprocessing the plurality of chaotic signals to obtain digital signals comprises: performing parallel photoelectric conversion on the plurality of chaotic signals to obtain initial electric signals; And acquiring the initial electric signal through a digital oscilloscope to obtain the digital signal.
- 4. The method of claim 2, wherein optimizing the digital signal to obtain the random number sequence comprises: Performing multistage filtering processing on the digital signal through a cascade narrow-band notch filter to obtain filtering data; Carrying out nonlinear probability distribution shaping treatment on the filtering data to obtain shaped data; Carrying out quantization processing on the shaping data to obtain quantized data; And selecting N effective bits from the multi-bit quantized data, and performing logic exclusive OR processing on the N effective bits to obtain the random number sequence.
- 5. The method of claim 4, wherein N is determined based on a support test condition for a random number performance test.
- 6. The method of any of claims 1-5, wherein the multi-transverse mode laser is designed in a manner that includes: determining a required sequence length of the random number sequence; Determining the number of required paths of the multipath chaotic signals based on the length of the required sequence; and determining the working parameters of the multi-transverse-mode laser based on the number of required paths and independent requirements among the plurality of chaotic signals.
- 7. The method of claim 6, wherein the operating parameters include a number of holes in a reticle in the multi-transverse mode laser and a number of transverse modes supported by a chaotic light field of the multi-transverse mode laser; the number of holes of the mask plate in the multi-transverse-mode laser is larger than or equal to the required number of paths; The number of the transverse modes supported by the chaotic light field of the multi-transverse mode laser is greater than or equal to 900.
- 8. The method according to any one of claims 1-5, further comprising: and carrying out random number performance test on the random number sequence.
- 9. A parallel random number generation device based on a multi-transverse mode laser is characterized by comprising a multi-transverse mode laser and a processing component; the multi-transverse mode laser is used for generating parallel and independent multipath chaotic signals; the processing component is used for carrying out signal processing on the plurality of chaotic signals to obtain a random number sequence; wherein the random number sequence comprises a plurality of parallel random numbers.
- 10. The apparatus of claim 9, wherein the multi-transverse mode laser comprises a gain medium, a pump source, a mask, a resonant cavity, and an output coupling; the pump source is used for carrying out side pumping on the gain medium; The resonant cavity is used for transmitting a chaotic light field, wherein the resonant cavity comprises a plurality of optical elements; the mask plate is used for carrying out spatial discretization on the output chaotic light field; The output coupling is used for outputting a plurality of chaotic signals.
Description
Parallel random number generation method and device based on multi-transverse mode laser Technical Field The application relates to the technical field of random number application, in particular to a parallel random number generation method and device based on a multi-transverse mode laser. Background Random numbers have wide application in the fields of secret communication, cryptography, quantum simulation, and the like. Currently, random number generators are largely divided into pseudo-random number generators and true random number generators. The method is based on deterministic algorithm generation, has periodicity and is easy to break, the safety is limited, and the method can generate true random numbers with high randomness by depending on physical entropy sources such as a chaotic laser, thermal noise and the like, but can only realize single-channel output. However, with the rapid development of information security and intelligent computing technologies, single-channel random number generators have been difficult to meet the requirements of high-speed, multi-tasking systems. On one hand, in high-speed encryption communication, the key updating rate needs to be matched with the Tbps-level data transmission rate, a single-path random source is limited by the bandwidth and the back-end sampling rate, and cannot support real-time encryption of multi-channel or multi-user links at the same time, and on the other hand, in parallel computing and random modeling, a large number of independent random sequences are used for Monte Carlo simulation, optimization algorithm or neural network weight initialization, and single-path output cannot meet the requirement of a large-scale parallel task on independent entropy flow. Furthermore, a single channel architecture also presents a potential risk in terms of system security that an overall random system will be faced with complete failure once the entropy source is disturbed or eavesdropped. Therefore, the current true random number generator cannot realize the parallel random number generation function of multiple channels and mutually independent, and cannot meet the use requirement of the current parallel random number. Disclosure of Invention In view of the above, an object of the embodiments of the present application is to provide a parallel random number generation method and apparatus based on a multi-transverse mode laser, so as to solve the problem that the parallel random number generation function of multiple channels and independent of each other cannot be realized in the prior art. In order to solve the above problems, in a first aspect, an embodiment of the present application provides a parallel random number generation method based on a multi-transverse mode laser, the method including: generating parallel and independent multipath chaotic signals through a multi-transverse-mode laser; Carrying out signal processing on the plurality of chaotic signals to obtain a random number sequence; wherein the random number sequence comprises a plurality of parallel random numbers. In the implementation process, the multi-transverse mode laser is used as a signal source of the true random number generator, the multi-transverse mode characteristic of the multi-transverse mode laser can be utilized, space-time chaos is generated by introducing gain competition of the multi-transverse mode laser, parallel independent chaos sources are constructed from space dimensions, so that a multi-channel chaos signal with extremely high randomness is obtained, and a plurality of parallel and mutually independent true random numbers with high randomness can be obtained by sampling the chaos signal and processing corresponding digital signals. The number and randomness of the generated random numbers are effectively improved, and the use requirement of parallel random numbers in various scenes is met. Optionally, the signal processing is performed on the plurality of chaotic signals to obtain a random number sequence, including: preprocessing a plurality of chaotic signals to obtain digital signals; Carrying out optimization processing on the digital signal to obtain the random number sequence; The optimization processing comprises filtering processing and nonlinear probability distribution shaping processing. In the implementation process, the chaotic signal can be preprocessed firstly to be converted into a digital signal which is convenient to process, and then the digital signal is subjected to various optimization processes such as filtering process, nonlinear probability distribution shaping process and the like, so that the randomness of the random number is effectively improved. Optionally, the preprocessing the plurality of chaotic signals to obtain digital signals includes: performing parallel photoelectric conversion on the plurality of chaotic signals to obtain initial electric signals; And acquiring the initial electric signal through a digital oscilloscope