Secure Multimodal Biometric Authentication with mGAN - Driven Hyperparameter Optimization: Enhancing BiLSTM–FOA

DOI

10.2991/978-94-6239-616-6_74How to use a DOI?

Keywords

Multimodal Biometrics; BiLSTM; Falcon Optimization Algorithm; Generative Adversarial Networks; Hyperparameter Optimization; Deep Hashing; Key Entropy

Abstract

Multimodal biometric authentication benefits from combining complementary modalities such as face, iris, and fingerprint to increase recognition accuracy and robustness. Prior work introduced a BiLSTM - based feature extraction pipeline with Falcon Optimization Algorithm (FOA) for cryptographic key extraction, demonstrating strong authentication performance and key security. However, manual tuning of the network and FOA hyperparameters limits generalizability and may settle on suboptimal configurations. In this paper, a Modified Generative Adversarial Network (mGAN)–driven Hyperparameter Optimization Module (HOM) that autonomously and simultaneously optimizes BiLSTM architecture and FOA parameters is proposed. The mGAN generates parameter candidates which are evaluated by the BiLSTM–FOA pipeline; a discriminator learns to accept high-performing parameter sets, guiding the generator toward optimal regions of the mixed discrete–continuous search space. We present the theoretical formulation, the joint objective combining authentication performance and key security, and a reproducible evaluation protocol. Using the same datasets as earlier work (FIFD, CASIA-IrisV4, PolyU Fingerprint), we report hypothetical but plausible improvements in recognition and key metrics that demonstrate the potential of the proposed approach. We conclude with a discussion of theoretical implications, limitations, and future directions.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - S. Jebapriya
AU  - V. Ganaga Durga
PY  - 2026
DA  - 2026/03/31
TI  - Secure Multimodal Biometric Authentication with mGAN - Driven Hyperparameter Optimization: Enhancing BiLSTM–FOA
BT  - Proceedings of the International Conference on Artificial Intelligence and Secure Data Analytics (ICAISDA 2025)
PB  - Atlantis Press
SP  - 1009
EP  - 1022
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-616-6_74
DO  - 10.2991/978-94-6239-616-6_74
ID  - Jebapriya2026
ER  -