MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202641099114 A) filed by Dr. Dinesh Singh Dhakar; Dr. Zahid Hussain Wani; Dr. Preeti Dhankar; Dr. Sudhir Mahajan; Rakesh Verma; and Dr. Nisha on August 17, 2026, for System For Artificial Intelligence-Based Network Capacity Forecasting Using Traffic Learning Nodes.
Inventors include Dr. Dinesh Singh Dhakar; Dr. Zahid Hussain Wani; Dr. Preeti Dhankar; Dr. Sudhir Mahajan; Rakesh Verma; and Dr. Nisha.
The application for the patent was published on August 28, 2026, under issue no. 35/2026.
Abstract: A system for artificial intelligence-based network capacity forecasting using traffic learning nodes is disclosed. The system comprises a plurality of traffic learning nodes distributed at distinct locations within a communication network, each node comprising a traffic data acquisition module for acquiring traffic parameters from an associated network hardware element, a local memory for storing a traffic history, and a local learning module for training a local forecasting model and generating a set of local model parameters. A central capacity forecasting server receives the local model parameters, without receiving the traffic history, aggregates the parameters to generate an aggregated forecasting model, and generates a network capacity forecast. A network resource control module adjusts allocation of a network resource when the forecast exceeds a predetermined threshold, thereby reducing data transmission volume and enabling automatic reconfiguration of network resources ahead of predicted congestion, as validated through working examples across multiple trial network deployments.
Disclaimer: Curated by HT Syndication.
