Aug 2026· PeerJ Computer Science· Vol 12, pp. e3916· 0 citations· 43 references
TL;DR
The evaluation confirms that the proposed model, incorporating differential privacy, provides a secure and scalable solution for managing sensitive citizen data while achieving reliable performance in privacy-aware biometric verification for real-world e-governance applications.
Abstract
Today, biometric authentication has become a central component of user security in social governance systems, where each government department demands access to user-specific data that varies across agencies. However, storing such data in centralized repositories increases serious privacy concerns, as unrestricted access by multiple entities maximizes the risk of data leakage. To address this, our research presents a novel biometric authentication system integrating robust privacy-preserving techniques, built on advanced deep learning architectures and differential privacy algorithms. A blockchain ledger integrated with a Merkle tree is used to securely store user identities, providing tamper-evident cryptographic validation of registered users. We further develop a novel hybrid model by integrating a pre-trained Vision Transformer (ViT) with a differential privacy-based machine learning enhanced training strategy, wherein the model is trained on noise-induced images to resist inference attacks. The system without differential privacy achieves 90.80% accuracy, 0.94 precision, 0.91 recall, and an F1-score of 0.90 in the standard configuration, while the differentially private model maintains 68.97% accuracy with
ε
= 6.2, ensuring a strong privacy—accuracy balance. The evaluation confirms that our proposed model, incorporating differential privacy, provides a secure and scalable solution for managing sensitive citizen data while achieving reliable performance in privacy-aware biometric verification for real-world e-governance applications.
The proposed framework for financial system fraud detection that is safe and protects privacy while resolving issues with data sharing, legal restrictions, and cybersecurity threats is appropriate for practical financial applications since it successfully improves fraud detection while guaranteeing Privacy Preservation, security, and openness.
Biometric authentication offers enhanced usability for digital payments, but traditional centralized architectures suffer from single points of failure. While blockchain integration promises decentralized trust, existing solutions often store biometric helper data (e.g., fuzzy commitments) directly on-chain. We demonstrate that such transparency, even in permissioned settings, exposes low-entropy biometric inputs to offline brute-force attacks and identity-linkage risks if the immutable ledger is accessed by malicious nodes. To resolve this transparency–privacy paradox, we propose a threshold-based decentralized authentication framework. Unlike prior works, our protocol keeps all sensitive Biometric-Enhanced Key Derivation (BEKD) tokens entirely off-chain, using the blockchain solely for freshness enforcement. We provide a game-based security analysis of brute-force resistance, unforgeability, and unlinkability. Our experimental results demonstrate that our scheme’s gas cost is acceptable, offering a robust solution for self-sovereign biometric identity.
Hui Cui, Haoze Cheng, James Boorman· Pragmatic Cybersecurity· 0 citations
The rapid growth of cybercrime, ransomware attacks, digital fraud, and large-scale cyber threats has
significantly increased the need for secure and collaborative cyber forensic investigations. Traditional machine learning
approaches often require organizations to share or centralize sensitive forensic datasets, creating challenges related to
privacy, confidentiality, data ownership, and security. To address these limitations, this project proposes a PrivacyPreserving Distributed Training Architecture for Cyber Forensics using Blockchain and Homomorphic Encryption. The
proposed framework integrates Federated Learning, Distributed Learning, CKKS-based Homomorphic Encryption,
Blockchain Technology, and a Secure Model Exchange Space to enable multiple agencies to collaboratively train
machine learning models without exposing their raw forensic data. Federated Learning allows organizations to train
models locally and securely aggregate encrypted model updates, while Distributed Learning enables encrypted dataset
partitions to be processed collaboratively by helper nodes without revealing the original data. CKKS Homomorphic
Encryption protects sensitive information during computation, and blockchain technology provides decentralized trust
through secure node authentication, transparent validation, immutable audit trails, and trusted model exchange among
participating agencies. The framework is implemented using Python, Flask, Scikit-learn, TenSEAL, Ganache, Solidity,
and Web3.py, providing a web-based platform for collaborative project management, encrypted training, blockchain
monitoring, secure model sharing, performance evaluation, and cyber forensic prediction. Experimental results
demonstrate that the proposed architecture successfully supports secure collaborative learning, encrypted computation,
blockchain-based validation, and trusted model sharing while maintaining effective prediction performance. By
integrating distributed learning, federated learning, homomorphic encryption, and blockchain into a unified framework,
the proposed system provides a scalable, secure, and privacy-preserving solution for next-generation cyber forensic
intelligence, enabling organizations to collaboratively strengthen cybersecurity without compromising the privacy,
confidentiality, or ownership of sensitive forensic data
R. Sridevi, P. Kumar· International Journal of Inn...· 0 citations
Cybersecurity in financial services remains a significant challenge, as remote-access platforms are frequent targets of credential theft, phishing, session hijacking, replay attacks, and other authentication-related threats. This paper presents a multi-factor, multi-layer authentication framework designed to enhance the security and auditability of financial-service access. The proposed framework integrates biometric verification, one-time password (OTP) validation, adaptive risk-based authentication, challenge–response verification, ECC-based digital signatures, and a ledger-supported audit layer within a unified authentication architecture. The framework operates across four security layers: biometric identity verification, OTP validation, dynamic contextual risk assessment, and adaptive challenge–response authentication. To strengthen integrity, confidentiality, and non-repudiation, authentication records are hashed, digitally signed using client-side ECC private keys, encrypted using the server’s public key, and recorded within a cryptographically linked audit ledger. Unlike purely conceptual approaches, the proposed framework has been fully implemented using Python (Flask), SQLite, and Docker, demonstrating practical deployability in real-world environments. Experimental evaluation includes functional validation, adversarial security testing, and operational performance assessment. The results demonstrate strong resistance against multiple authentication-related attack scenarios, including brute-force, replay, session hijacking, phishing, and man-in-the-middle attacks, while maintaining acceptable operational performance. The findings indicate that the proposed framework improves authentication assurance, traceability, and security resilience for modern financial-service environments through the coordinated integration of biometric, cryptographic, behavioral, and ledger-based security controls.
H. Hamidi, Far. Fathi, S. Mohammadi· Journal of King Saud Univers...· 1 citation
A block chain-based Privacy-preserving and Secure Federated Learning (BPS-FL) system that uses threshold homomorphic encryption to safeguard the local gradients of clients in order to successfully solve such privacy and security assault challenges is suggested.
Umema Samreen, Dr. I. Samuel, Peter James· American Journal of AI Cyber...· 0 citations
The proposed framework effectively integrates encryption, federated intrusion detection, explainable artificial intelligence, and blockchain security to enhance privacy, transparency, and reliability in IoMT healthcare networks.
P. Banupriya, K. Vanitha· Journal of Vibration Enginee...· 0 citations