High Throughput 512Mbps True Random Number Generation Using Uncertainty in Mobile Communication Bands

Abstract

This study presents a new approach to generating true random numbers by exploiting the random nature inherent in mobile communication networks. Taking advantage of fluctuations in total users, user activity rate, and channel fading, we receive a raw signal in the radio spectrum using bowtie microstrip antennas designed for GSM and 5G frequency bands in the 1.8 and 3.5 GHz bands that can be used in our system. The captured data is then processed using a field-programmable gate array to generate high-quality true random numbers. Bow tie antennas are designed to have an omnidirectional pattern and can receive the signal from all directions. We have evaluated this through numerical calculations and antenna pattern measurements in the laboratory, and the results and their comparison show a promising performance. Our random number generation system also includes a Zynq SoC module and a dual-channel AD9226 ADC, enabling real-time random number generation through the ethernet port at up to 512Mbps. To ensure the system’s performance, standard statistical tests, including the NIST, DIEHARDER, and ENT test suites, have been done, and all confirm the true randomness of the generated random numbers. Our findings show the effectiveness of the 3.5 GHz frequency band as a suitable source for generating random numbers, especially in mobile devices and the Internet of Things. Due to the high speed of generating random numbers, this optimal and efficient TRNG system can have broad applications in cryptography, statistical analysis, and computational simulations.

Keywords

• 5G
• antenna
• cryptography
• field programmable gate arrays
• GSM
• true random number generator