MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202611057876 A) filed by Panipat Institute Of Engineering And Technology Piet on May 06, 2026, for Blockchain-Verified Resource-Aware Secure Federated Learning System For Industrial Internet Of Things Consortiums.
Inventors include Dr. Sakshi Patni; and Mr. Anurag Vashist.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: [065] The present invention relates to a blockchain-verified resource-aware secure federated learning system and method for Industrial Internet of Things consortiums. Multiple industrial entities train local machine learning models using private industrial datasets and generate model updates without exposing raw data. Each update is associated with metadata, cryptographically hashed, digitally signed, and submitted as a verifiable commitment to a permissioned blockchain. Smart-contract based validation checks participant identity, submission freshness, model version compliance, and consortium policy rules. The actual update payload is encrypted and maintained off-chain for efficient storage and controlled retrieval. Verified updates are statistically screened to detect abnormal, malicious, stale, or low-quality contributions before aggregation. A resource-aware scheduling engine selects participants and assigns workload according to compute capacity, memory, bandwidth, reliability, and trust indicators. Robust aggregation produces a global model, and a hash of the resulting model is recorded on-chain for auditability. The invention enables privacy-preserving, traceable, secure, and scalable collaborative learning for industrial organizations. Accompanied Drawing [FIGS. 1-2]
Disclaimer: Curated by HT Syndication.
