Skip to content

STRUCTURAL AND PARAMETRIC SYNTHESIS OF A USER BIOMETRIC AUTHENTICATION SYSTEM

2026 · RSUH/RGGU Bulletin. Series Information Science. Information Security. Mathematics · 0 citations

Abstract

Modern requirements for information systems that process restricted information define the need for developing new methods and approaches to user authentication that combine high reliability with usability. The article examines a current approach to creating a biometric user authentication system based on the neural network transformation of dynamic biometric parameters (analysis of password entry on the keyboard) into a unique identifier. A functional model of a highly reliable user authentication system has been constructed. Structural and parametric synthesis of the biometric user authentication system is carried out by integrating convolutional neural networks, recurrent layers, and an encoder, which together enhance the stability of dynamic biometric parameters. Unlike traditional data compression methods based on their statistics and entropy, an algorithm based on a trained neural network is proposed, which allows for identifying patterns in processed data and optimizing the compression process of unstable biometric data. The practical significance of the proposed solutions lies in integrating the functions of registration, verification, and user identification using a unified biometric system, which enhances information security efficiency while reducing allocated resources. Testing of the NPBC revealed a pattern showing that increasing the length of the entered password improves the stability of the identifier.

View source

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.