Design And Implementation Of True Random Number Generator Based On FPGA

With the rapid development of the Internet of Things,cloud computing and big data,the scale of communication is exponentially increasing.The issue of information security attracts more and more attention.The encryption system is the guarantee of information security.The true random number generator plays extremely important role in the encryption system.The increase of communication speed leads to the increase of the demand of true random numbers and still shows an upward trend.The generation speed of random numbers,power consumption and reliability are always the focus of this kind of research.FPGA is an ideal experimental platform for secure cryptographic hardware implementations,and the demand for true random numbers in FPGA-related applications is also growing.The difficulty of routing and the influence of process deviation make the existing true random number generator less applicable to the FPGA and it is difficult to meet the demand of the related application for the true random number.In this master's thesis,a high-entropy random number generator based on FPGA is proposed,which uses non-traditional latch structure combined with a new random number acquisition method to obtain true random number.Compared with the true random number generator widely used in FPGA,This method has an adjustable function that eliminates the need for manual wiring and overcomes the effects of process variations,and has a low resource consumption and high random number generation rate,providing a new reference for related research on FPGA.Compared with other existing methods,the proposed method firstly uses a highly reliable structure to ensure its stable performance and secondly improves the data generation rate through the improved method,which makes the real random number generator increased the rate on the basis of high reliability.Through the PVT test,the experimental results show that the true random number generator has high robustness to temperature(20?~80?),voltage(0.9~1.1V)and process variation,and the true random numbers can pass the NIST randomness test.The random number generation rate is 14.2M/s under normal working conditions.