MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202641069313 A) filed by Dr. V. Kavitha; Dr. R. Pandiyarajan; Dr. R. Rajesh; Kuldeep Chouhan; Meriga Kiran Kumar; Dr. Sunil Babu Melingi; Krishna Katyal; Avula Geethanjali; Dr. Ravi Ray Chaudhari; Dr. Namrata Tripathi; Dr. Ayesha Heena; and Dr. Najimuddin M Maroof on June 02, 2026, for A Neuromorphic Computing System For Brain-Inspired Artificial Intelligence Using Event-Driven Neural Processing.
Inventors include Dr. V. Kavitha; Dr. R. Pandiyarajan; Dr. R. Rajesh; Kuldeep Chouhan; Meriga Kiran Kumar; Dr. Sunil Babu Melingi; Krishna Katyal; Avula Geethanjali; Dr. Ravi Ray Chaudhari; Dr. Namrata Tripathi; Dr. Ayesha Heena; and Dr. Najimuddin M Maroof.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: The present invention discloses a neuromorphic computing system for brain-inspired artificial intelligence using event-driven neural processing. The proposed system is designed to emulate the operational behavior of biological neural networks through asynchronous spike-based communication and adaptive neural computation mechanisms. The invention comprises sensory acquisition modules, event encoding units, spiking neural processing cores, adaptive synaptic learning modules, neural event routing frameworks, memory-efficient neural state management units, and low-power hardware acceleration components configured for intelligent real-time processing. The disclosed system converts multimodal sensory information including visual, audio, motion, biomedical, and environmental data into sparse neural spike representations for efficient event-driven computation. Spiking neural processing units selectively activate only when meaningful neural events occur, thereby reducing unnecessary processing operations, communication overhead, latency, and power consumption. Adaptive synaptic learning mechanisms dynamically modify neural connection strengths based on spike timing relationships, neural activity patterns, and contextual feedback conditions to support continuous learning and autonomous behavioral adaptation. The invention further incorporates distributed neural communication pathways and hierarchical cognitive processing layers capable of performing feature extraction, pattern recognition, anomaly detection, contextual analysis, and intelligent decision-making. Low-power neuromorphic hardware acceleration modules enable scalable deployment in embedded systems, edge computing platforms, robotics, healthcare monitoring systems, autonomous vehicles, industrial automation environments, and intelligent surveillance applications. The proposed neuromorphic computing framework overcomes the limitations of conventional artificial intelligence architectures associated with high energy consumption, centralized processing dependency, and limited scalability. The invention thereby provides an efficient, adaptive, scalable, and biologically inspired artificial intelligence platform suitable for next-generation cognitive computing applications.
Disclaimer: Curated by HT Syndication.
