MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202611116342 A) filed by Chandigarh University on September 29, 2026, for Intelligent Edge-Computing System For Predictive Power-State Calibration In Mesh Networks.
Inventors include Nidhika Chauhan; and Srijan Tripathi.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: An intelligent edge-computing system for predictive power-state calibration in battery-powered mesh networks includes sensing circuitry, a local telemetry buffer, a federated learning engine, prediction outputs, a power-state scheduler, a microcontroller, a battery, an antenna and a multi-state radio. The sensing circuitry acquires environmental measurements, while the local telemetry buffer records sensor readings and network telemetry including packet activity, collisions, link quality and battery level. The federated learning engine trains a local prediction model and exchanges model updates with neighbouring nodes without exchanging raw sensor data. The model generates a data relevance probability and a network congestion probability. The power-state scheduler uses the prediction outputs to select a radio state among Active, Idle and Deep Sleep, thereby reducing unnecessary radio activity while maintaining transmission of relevant data under suitable network conditions. The distributed learning and hardware power-state control cooperate to provide adaptive, self-calibrating energy management for mesh IoT networks
Disclaimer: Curated by HT Syndication.
