Brainwaves Signal Based Recognition Using Graph Convolutional Neural Network

Abstract

This research explores secure personal authentication technologies amid the dynamic landscape of information technology. Traditional methods face vulnerabilities such as theft and replication, leading to an increased focus on robust biometric solutions. However, even advanced biometrics carry security risks. To address these concerns, we propose a novel approach using brainwaves signal recognition technology, offering non-reproducible, non-forgeable, and tamper-resistant advantages. Our method employs Graph Convolutional Networks (GCN) for feature extraction and recognition classification of brainwaves signals, reducing computational complexity. While the proposed GCN-based classification shows promise, further refinement is needed. This research contributes insights and methodologies to enhance secure personal authentication in the evolving technological landscape.