MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641086039 A) filed by Dr. A. Vijayalakshmi; Divya N; Dr. P. Sriraaj; Dr. C. P. Vijayalakshmi; Dr. T. N. Aruna; V. K. Janani Sivapriya; and Dr. K. Angayarkkani on July 14, 2026, for System For Deep Neural Signal Decoding In Wearable Electronics Using Multi-Band Sensor Fusion.
Inventors include Dr. A. Vijayalakshmi; Divya N; Dr. P. Sriraaj; Dr. C. P. Vijayalakshmi; Dr. T. N. Aruna; V. K. Janani Sivapriya; and Dr. K. Angayarkkani.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: [046] The present invention relates to a wearable electronic system for deep neural signal decoding using multi-band sensor fusion. The system comprises a wearable body, body-contact electrical sensor units, a multi-band analog front-end, an auxiliary sensor array, a synchronisation controller, an adaptive fusion processor, an embedded deep neural decoding engine, a confidence evaluation module, a power-management unit, and an output interface. The system acquires neural-indicative body signals, separates the acquired signals into multiple frequency bands, acquires auxiliary non-electrical context signals, aligns the signals using a common timing reference, evaluates artefacts and channel reliability, and generates a fused multi-band representation. The embedded deep neural decoding engine locally decodes user-intent, neuromuscular, attention, fatigue, gesture, or interaction-related states from the fused representation. The confidence evaluation module validates the decoded output, while the power-management unit controls sensing and inference operations to reduce power consumption. The invention improves wearable signal decoding accuracy, reduces artefact- induced false outputs, lowers latency, and enhances real-time human-machine interaction reliability. Accompanied Drawing [FIGS. 1- 2]
Disclaimer: Curated by HT Syndication.
