INTEGRATION OF PHYSICAL SENSORS IN THE GENERATION OF PSEUDO-RANDOM NUMBERS

Abstract

The paper presents an approach to building a pseudo-random number generator (PRNG) based on physical sensors, which is a promising direction for ensuring reliability and security in various fields such as cryptography, Internet of Things (IoT), gaming industry, and cyber-physical systems. Sensors of various types, including temperature, sound, light, gyroscopes, and magnetometers, show high reliability as sources of entropy due to the natural fluctuations and unpredictable characteristics they register. The physical phenomena measured by these sensors generate a significant level of unpredictability, which is critical for building reliable, predictability-resistant pseudo-random number generators. The work describes in detail the process of digital processing of signals from sensors, which includes several stages: noise filtering, signal normalization, quantization and binarization to obtain random bits. Noise filtering ensures the elimination of unwanted interference, normalization allows you to bring signals to a single scale, and quantization and binarization ensure the conversion of analog signals into a discrete form suitable for generating pseudorandom numbers. This approach guarantees the high quality of random bits, which is confirmed by the results of cryptographic and statistical tests that were conducted to assess the randomness and predictability of the received numbers. The research results showed that generators based on physical sensors can effectively provide a high degree of randomness sufficient for applications requiring high security. In particular, the approach can be useful in cryptography to generate encryption keys, in IoT systems to protect data privacy, in cyber-physical systems to ensure reliable operation, and in the gaming industry to create unpredictable game scenarios. The article also provides practical recommendations for the selection of sensors depending on the conditions of use and equipment requirements. For example, motion sensors (gyroscopes, accelerometers) can be effective for mobile devices, while for stationary systems it is appropriate to use temperature or sound sensors. Aspects of integrating sensor generators into various systems are also considered, including how they can be combined with existing infrastructures and hardware requirements. Thus, the considered approach to building a generator of pseudo-random numbers based on physical sensors is a promising direction that can significantly increase the reliability and security of modern systems. The proposed methods of signal processing and recommendations for the selection of sensors contribute to the effective implementation of generators in various fields, ensuring a high level of entropy and protection against possible attacks.