Microservice Architecture of an Adaptive Authentication System for Users of Financial Institutions

Abstract

The article addresses the problem of developing a microservice architecture for an adaptive authentication system of financial institutions’ users as a key direction for enhancing the security, scalability, and flexibility of modern information systems. The authors analyze current authentication methods, including single-factor, two-factor, and multi-factor approaches, and emphasize the particular importance of adaptive authentication, which provides dynamic adjustment of identity verification parameters depending on the assessed risk level of an operation. The study explores methods successfully applied in the financial sector, such as behavioral biometrics, contextual authentication, anomaly detection, and the use of machine learning algorithms. The proposed conceptual model implements a division of functionality into independent microservices, which facilitates flexibility, modularity, and easy integration with national and international identification standards (including BankID and Diia services). Special attention is given to the visualization of architectural dependencies, UML diagrams, and the use of modern client- and server-side software development tools such as Vue.js, Nest.js, and Spring Boot. The system provides multi-channel authentication (email, SMS, QR code), supports load balancing, and incorporates fault tolerance checking, which are critically significant for financial organizations. The results of the study may serve as a foundation for implementing scalable and reliable cybersecurity solutions, as well as a basis for further research in the field of adaptive authentication.