MUMBAI, India, Nov. 2 -- Intellectual Property India has published a patent application (202611116470 A) filed by Vivekananda Institute Of Professional Studies - Technical Campus on September 29, 2026, for System And Method For Analog-Domain Semiconductor Failure Prediction Using A Nonlinear Material Neural Network.
Inventors include Prof. Dr. Ravi Prakash Iyer; and Dr. Rajesh Yadav.
The application for the patent was published on October 30, 2026, under issue no. 44/2026.
Abstract: The invention relates to semiconductor health monitoring and predictive failure analysis for power, RF, and high-frequency semiconductor devices. A semiconductor device monitoring unit observes live physical observables including drain current variations, gate voltage fluctuations, thermal changes, electric field disturbances, switching noise patterns, and electromagnetic disturbance behavior, and supplies continuous-time analog signals to an analog sensor interface and a metamaterial-based signal conditioning layer for selective enhancement of degradation-related content. A nonlinear material analog neural processing core, including a lithium niobate waveguide array, performs analog-domain pattern recognition, feature extraction, and failure signature mapping without analog-to-digital conversion or conventional digital computation, followed by analog decision and prediction of degradation level or failure likelihood. An adaptive feedback and control module initiates protective action upon threshold exceedance. The invention provides lower latency, reduced conversion overhead, and direct analog inference for real-time semiconductor protection.
Disclaimer: Curated by HT Syndication.
