AI-BASED QUANTUM RANDOM NUMBER GENERATION FOR SECURE COMMUNICATIONS

Abstract

One of the solutions to the problem of secure communication is the Quantum Random Number Generators (QRNGs) which generate true randomness based on quantum mechanical principles. QRNGs are different from conventional pseudo random number generator in that they use quantum uncertainty to generate outputs which are very unpredictable. This work attempts to examine different QRNG architectures, study the possible integration of cryptographic frameworks with these architectures and highlight potential vulnerabilities in the integration of cryptographic frameworks with QRNG architectures. Post-processing techniques, i.e. Zero-Phase Component Analysis (ZPCA) and neural network based techniques are also studied to increase entropy and suppress environmental noise in QRNG systems. Key findings show that system jitter models based QRNG architectures are more scalable and less expensive for use in real world applications. Additionally, incorporating cryptographic protocols with the QRNGs endows them with the ability to resist quantum threats, and hence, provide better key distribution in the secure IoT communication frameworks. However, as security problems such as side channel attacks and hardware manipulation are still big issues. It also points out some of the emerging research trends including hybrid QRNG architectures, i.e., QRNG with classical PRNG to increase the entropy rate. As such, the proposed methodologies attempt to enhance QRNG scalability, security, and energy efficiency in cloud infrastructures, in mobile systems and in critical security environments. The advancement of these technologies will assume a crucial function in the advancement of quantum-safe frames for current advanced computerized foundation.