MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641079376 A) filed by Cmr Engineering College on June 27, 2026, for Federated Learning Based Privacy-Preserving Data Analytics System For Distributed Devices.
Inventors include Dr. G. Sumalatha; Mrs. Karnati Durga; Mr. K. Anil; Mrs. Suhasini T S; Mrs. Padakanti Bhargavi; Mr. P. Chander; and Mrs. M. Priyanka.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: Abstract A federated analytics system is disclosed for use in a distributed device environment in which data remains stored at local devices or local gateways. The system comprises a processor, memory, a communication interface, an input acquisition module, a preprocessing module, an analysis engine, and an output control module. The communication interface receives local training datasets, device-side feature updates, model gradients, validation statistics, and aggregation messages. The preprocessing module validates and normalizes received input records before the analysis engine performs privacy-preserving federated analytics using stored rules, profiles, policies, model parameters, or reference records. The output control module compares a generated decision value with a stored threshold or policy condition and generates global model updates, device training instructions, aggregation logs, and analytics results. The system stores an audit record linking source identity, validation state, decision value, and selected output action. The arrangement provides a controlled technical data path from input Fig. 1 Dated June 23, 2026 / Mungalpara Jigneshbhai Chhaganbhai IN/PA-2640 Agent for the Applicant
Disclaimer: Curated by HT Syndication.
