MUMBAI, India, Aug. 12 -- Intellectual Property India has published a patent application (202611069533 A) filed by Chandigarh University on June 03, 2026, for Ai-Driven Privacy-Preserving Rf-Based Domestic Escalation Detection System.
Inventors include Shaman Sharma; Rishi Kumar; Nistha; Dr. Manoj Gaur; Dr. Neeru Bala; and Dr. Deepak Kumar.
The application for the patent was published on August 07, 2026, under issue no. 32/2026.
Abstract: The present invention relates to a privacy-preserving edge artificial intelligence framework for non-invasive domestic escalation detection using radio frequency (RF)- based biometric sensing and spatiotemporal behavioral intelligence. The system utilizes Wi-Fi Channel State Information (CSI) to detect physiological and behavioral variations caused by human movement without employing cameras, microphones, or wearable devices. Captured RF signals are processed through filtering, normalization, and feature extraction techniques to derive biometric indicators including respiration patterns, heart-rate fluctuations, and motion intensity. A hybrid artificial intelligence architecture comprising Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) networks, and Graph Neural Networks (GNN) analyzes spatial, temporal, and interaction-based behavioral patterns to predict escalation risks. A probabilistic escalation assessment mechanism generates risk levels and activates encrypted alerts upon threshold exceedance. Federated learning and edge computing ensure decentralized processing, privacy preservation, reduced latency, and secure real-time deployment across smart domestic environments.
Disclaimer: Curated by HT Syndication.
