A Method for Generating True Random Numbers With Multiple Distribution Characteristics

Abstract

In this paper, a method of generating true random numbers obeying multiple distribution characteristics is proposed. First, two resistance-capacitance (RC) self-excited oscillation circuits are used to generate two jittered, periodically unstable square wave signals, and then a text high-precision and high-frequency quartz crystal oscillator is used to sample and measure these two jittered signals to obtain their periods. Due to the randomness of the external environment, there is Gaussian white noise in the circuit, so the periods of these two signals are random variables subject to the Gaussian distribution. Finally, we use a text field-programmable gate array (FPGA) to perform secondary processing on these two random signals, which can generate true random number sequences that conform to the distribution characteristics of Gaussian distribution, Poisson distribution, 0–1 distribution, uniform distribution, etc. The random numbers generated by the circuit can be applied to the field of intelligent control related to automatic control and machine learning. This method uses a physical entropy source to generate high-quality true random sequences, which are easy to build, low in circuit cost, and small in size. The experimental results show that the method is low-cost, highly reliable and easy to integrate, providing an effective solution for the generation of true random numbers in electronic systems.

Keywords

• True random number
• RC self-excited oscillation
• random distribution
• FPGA
• intelligent control