MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641064006 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on May 21, for 'a system for detection and spatio-temporal mapping of invisible disturbances in a multi-sensor environment.'
Inventor(s) include Gopika S; Dhaya Lakshmi A; and Ramya G.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "The present invention relates to a system and computer-implemented method are disclosed for detecting and spatio-temporally mapping invisible disturbances in a multi-sensor environment. The system employs a multi-modal sensor array to acquire heterogeneous time-series data, followed by preprocessing to synchronize, normalize, and extract features. A self-supervised temporal learning model, such as a Temporal Convolutional Network (TCN), learns baseline inter-signal dynamics, while a causal inference module, utilizing techniques including Granger causality within a Directed Acyclic Graph (DAG) framework, generates directional causal influence matrices. Real-time causal relationships are compared with baseline behavior to compute a collective causal deviation score, and an adaptive threshold is dynamically learned from historical distributions to identify latent disturbances not observable in individual signals. Upon detection, the system generates a disturbance fingerprint vector capturing causal and temporal characteristics, and estimates the spatial origin and time of occurrence using propagation delay analysis across distributed sensors. A confidence scoring and feedback interface outputs the disturbance characteristics, enabling reliable downstream decision-making in complex multi-signal systems."
Disclaimer: Curated by HT Syndication.
