Research On The Quantum Random Number Generator Based On Non-uniform Quantization

Random numbers are important resources in science technology and engineering applications,and play a vital role in fields such as secure communication,basic theory testing,and computer simulation.At present,the acquisition of random numbers is mainly realized through deterministic algorithms,but because pseudo-random numbers can be predicted in nature,there are certain security risks.With the development of quantum information technology,it is becoming more and more feasible to extract randomness from quantum systems to generate random numbers.This scheme is essentially based on the unpredictability of measurement collapse in the quantum world,and the basic principles of quantum mechanics can guarantee the randomness of the random numbers it generates.This kind of random number generator based on quantum mechanics is called a quantum random number generator(QRNG).Because the random numbers generated by QRNG are highly random and unpredictable,QRNG has gradually become a research hotspot in the field of quantum information.This paper mainly studies the QRNG based on non-uniform quantization.The main contents include:1.The min-entropy study of the QRNG based on non-uniform quantization.In the traditional QRNG scheme,an analog-to-digital converter(ADC)is mainly used to uniformly quantize the output signal of the detector to generate the original random number sequence.However,the extractable randomness of the original random number sequence generated by ADC single sampling is relatively small,and the utilization rate of the initial randomness of the original random number sequence is insufficient.Aiming at the above problems,we propose a new non-uniform quantization method,and apply this method to the QRNG scheme based on vacuum fluctuations.In this scheme,we use the min-entropy to quantify the extractable randomness of the original random number sequence,and compare the min-entropy of the uniform quantization method and the non-uniform quantization method.The simulation result proves that when the quantum-to-classical-noise ratio(QCNR)is relatively high,the non-uniform quantization method we propose can extract more randomness from the original random number sequence,and the higher the QCNR,the more obvious the advantage.At the same time,because the ADC only supports a uniform quantization method,we designed a non-uniform quantization method using a multi-channel voltage comparator.2.Finite-size analysis of the QRNG based on non-uniform quantization.The min-entropy is a commonly used randomness quantification method,but the entropy authentication method of the source-independent QRNG protocol is a more rigorous and more secure randomness quantification method.Therefore,we use the above-mentioned randomness quantification method to quantify the extractable randomness of the original random number sequence in the non-uniform quantization QRNG scheme.At the same time,the research on quantum random numbers usually only considers the quantum randomness in the ideal case of infinite-size.However,the finite-size effect will also affect the quantum randomness in practical applications.Therefore,we compare the extractable randomness of the original random number sequence in the ideal case of infinite-size and the actual case of finite-size,and prove the influence of finite-size effect on quantum randomness.