Semi-self-testing Quantum Random Number Generator with CMOS Sensors

宋萧天, 李宏伟, 银振强 +5 more

Yan Li, Jing Wang, Yanjie Zhang +3 more

Random numbers are extensively utilized in cryptography, where the application of security systems demands random number generators of high reliability. In the scheme of generating quantum random numbers based on photon number counting—LED light source, CMOS image sensor(CIS), and post-processing module are integrated together. However, property of LED and CIS are significantly influenced by the operating environment temperature, limiting its applications in practice. This paper proposes the self-regulating module to regulate the flux of LED light and exposure time of CIS during operation, which enlarges working temperature of the QRNG chip. The novel self-regulating module can adjust the LED driven current, CIS exposure time based on the data of self-test. The mean value of readout number in extreme temperature is stabilized at around 550. Technical noise is under 8 electrons and quantum noise minentropy is stabilized at around 0.53 bits/bit in extreme temperatures. Tests results demonstrate the output random number bits satisfy the criteria of standards at extreme temperature −40 °C and 85 °C.

Albert Capdevila Ortiz

Maksim Iavich, Tamari Kuchukhidze, Giorgi Iashvili +1 more

The subject matter of the article is pseudo-random number generators. Random numbers play the important role in cryptography. Using not secure pseudo-random number generators is a very common weakness. It is also a fundamental resource in science and engineering. There are algorithmically generated numbers that are similar to random distributions but are not random, called pseudo-random number generators. In many cases the tasks to be solved are based on the unpredictability of random numbers, which cannot be guaranteed in the case of pseudo-random number generators, true randomness is required. In such situations, we use real random number generators whose source of randomness is unpredictable random events. Quantum Random Number Generators (QRNGs) generate real random numbers based on the inherent randomness of quantum measurements. The goal is to develop a mathematical model of the generator, which generates fast random numbers at a lower cost. At the same time, a high level of randomness is essential. Through quantum mechanics, we can obtain true numbers using the unpredictable behavior of a photon, which is the basis of many modern cryptographic protocols. It is essential to trust cryptographic random number generators to generate only true random numbers. This is why certification methods are needed which will check both the operation of the device and the quality of the random bits generated. The goal of the research is also to develop the model of a hybrid semi self-testing certification method for quantum random number generators (QRNG). The tasks to be solved are to create the mathematical model of a random number generator, which generates the fast random numbers at a lower cost. To create the mathematical model of a hybrid semi self-testing certification method for quantum random number generators. To integrate a hybrid semi self-testing certification method to the hybrid random number generator. the methods used are mathematical optimization and simulation. The following results were obtained: we present the improved hybrid quantum random number generator, which is based on QRNG, which uses the time of arrival of photons. The model of a hybrid semi self-testing certification method for quantum random number generators (QRNG) is offered in the paper. This method combines different types of certification approaches and is rather secure and efficient. Finally, the hybrid certification method is integrated into the model of the new quantum random number generator. Conclusions. The scientific novelty of the results obtained is as follows: 1. The hybrid quantum random number generator is offered, which is based on QRNG, which uses the time of the arrival of photons. It uses the simple version of the detectors with few requirements. The hybrid QRNG produces more than one random bit per the detection of each photon. It is rather efficient and has a high level of randomness. 2. The hybrid semi self-testing certification method for quantum random number generators (QRNG) is offered. The Self-testing, as well as device-independent quantum random number generation methods, are analyzed. The advantages and disadvantages of both methods are identified. Based on the result the hybrid method is offered. 3. The hybrid semi self-testing certification method for quantum random number generators is integrated into the offered model of the quantum random number generator. The paper analyzes its security and efficiency. The paper offers to use the new random number generator in the crypto-schemes.

Vardaan Mongia, Abhishek Kumar, Shashi Prabhakar +2 more

Zhenguo Lu, Jundong Wu, Yu Zhang +4 more