MUMBAI, India, Feb. 13 -- Intellectual Property India has published a patent application (202541124069 A) filed by Stanley College Of Engineering & Technology For Women, Hyderabad, Telangana, on Dec. 9, 2025, for 'secure biometric authentication system using on-device federated learning.'
Inventor(s) include Dr. M. Swapna; Dr. Latha Yallanki; Dr. N. Nagasekhara Reddy; T. Sandeep; and Ryavadi Bucha Rao.
The application for the patent was published on Feb. 13, under issue no. 07/2026.
According to the abstract released by the Intellectual Property India: "A secure biometric authentication system using on-device federated learning is disclosed. The system employs a secure enclave that performs all biometric acquisition, feature extraction, and model inference locally, preventing raw biometric data from leaving the device. Encrypted, differentially private model updates are generated on each device and aggregated using a decentralized, tamper-resistant federated learning workflow to enhance global model accuracy without compromising user privacy. Integrated anomaly-detection, anti-spoofmg mechanisms, and adaptive local training provide resilience against poisoning attacks, synthetic biometric forgeries, and natural biometric variability. The invention delivers a privacy-preserving authentication framework suitable for mobile devices, loT nodes, accesscontrol systems, and distributed security applications."
Disclaimer: Curated by HT Syndication.
