MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641075958 A) filed by Cvr College Of Engineering on June 19, 2026, for Federated Deep Learning System For Privacy-Preserving Smart City Intelligence..
Inventors include Srikanth Lakumarapu; and Sampurnima Pattem.
The application for the patent was published on June 26, 2026, under issue no. 26/2026.
Abstract: The current invention introduces a Federated Deep Learning System for Privacy-Preserving Smart City Intelligence that allows collaborative artificial intelligence model training across dispersed smart city infrastructures while maintaining data privacy. Data from several sources, including traffic sensors, monitoring systems, medical equipment, environmental monitoring units, and smart grids, is gathered by the system. Each node trains a local model and shares only encrypted model parameters instead of sending raw data to a central server. To produce an improved global deep learning model, a federated aggregation server securely integrates these changes. The improved model is given back to the nodes that are taking part so that they can keep learning and getting better. The suggested approach improves prediction accuracy, facilitates real-time decision-making, lowers communication overhead, improves data security, and allows scalable intelligent services for contemporary smart city surroundings.
Disclaimer: Curated by HT Syndication.
