Abstract—Biometric authentication systems have become essential for secure and reliable identity verification. Most of the existing biometric methods are unimodal systems. Unimodal methods typically focus on single method such as fingerprint, iris, face or Electrocardiogram (ECG). However, unimodal systems often suffer from limitations such as susceptibility to spoofing, noise sensitivity, and reduced accuracy. In this study, we propose a novel multimodal biometric authentication framework that integrates ECG and fingerprint data based on deep learning method. The proposed model used a Transformer-based Generative Adversarial Network (TGAN) for the dataset augmentation. An enhanced Visual Geometry Group (VGG-16) model is applied for deep feature extraction. To combine modality-specific representations, the model used a weighted feature fusion technique. To further improve classification performance, we employ a Multi-Support Vector Machine (Multi-SVM) classifier. The proposed model was evaluated using the Physikalisch-Technische Bundesanstalt (PTB-XL), Electrocardiogram Identification Database (ECG-ID), Fingerprint Verification Competition 2004 (FVC2004), and LivDet 2023 datasets. Results demonstrated that our proposed approach outperforms the baseline models. Specifically, the multimodal system achieved up to 98.4% accuracy with an Area Under the Curve (AUC) of 0.993, and an Equal Error Rate (EER) as low as 1.1%.
 
Keywords—biometric authentication, Convolutional Neural Network (CNN), deep learning, Fingerprint, Electrocardiogram (ECG), Support Vector Machine (SVM)


Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).