A GAIT-Based Three-Level Authentication System for Enhanced Cybersecurity
Vijaya Shetty S, P Sowmya, Shruthi Shetty J, Samparna Rautray · 2023
Although centralized digital organizations offer secure communication between people, services, and technology, the digital revolution poses significant threats. Data can be mined, profiled, and used without the user's knowledge or permission. Additionally, centralized solutions are wasteful, contain security holes, and are difficult for people to use. To ensure the privacy and security of dispersed digital identities, it is essential to authenticate and verify digital identities. Unfortunately, the existing body of knowledge lacks in-depth studies on privacy preservation and securing sensitive information in identity and access management systems, as well as concerns related to unified communications. Blockchain technology offers innovative solutions for digital identity management and verification, which are the most promising applications. Authenticating, distributing, and securing sensitive data requires safer techniques, and blockchain identification system development can help solve problems faced by centralized databases. Physical technology and human action are not enough to protect against cybercrime. Cybercriminals use online tools to commit crimes, but AI technology can help reduce incursions. In this study, password-based authentication, biometric authentication, and AI-powered anomaly detection have been employed for Continuous monitoring and real-time analysis to enhance defence against AI-driven attacks. The developed robust gait motion data acquisition system allows in measuring of human limb acceleration, angular velocity and muscle activity throughout time and helps in determining genuine, active, and passive imposters. The performance of the system is studied in terms of DET curve which shows how well the authentication system operates at different threshold points.