Analysis of the effects of temperature increase on quantum random number generator

Abstract

Quantum random number generator (QRNG) relies on the intrinsic randomness of quantum mechanics to produce true random numbers which are important in many fields, such as cryptographic protocols, security applications and Monte Carlo simulations. Recently, QRNGs with semiconductor light source have attracted a lot of attention due to their operational simplicity and high generation rate. However, the output characteristics of the light source in QRNG may be affected by temperature variations, which impacts the performance of QRNG. In this work, we study the effects of temperature increase on the super-luminescent emitting diode (SLED)-based QRNG with a numerical method. The results show that the temperature increase in SLED leads to the decrease of min-entropy. When the temperature is higher than some threshold value, the output sequence cannot pass the NIST Statistical Test. Our analysis also shows that the QRNG can be more robust by improving the post-processing method and monitoring the temperature as well as the min-entropy and thus provides a reference for the standardization of QRNG.