MUMBAI, India, Oct. 12 -- Intellectual Property India has published a patent application (202621101589 A) filed by Dr. D. Y. Patil Institute Of Technology, Pimpri, Pune -; and Dnyaan Prasad Global University, School Of Technology And Research on August 22, 2026, for Ai Powered Cyber Security Based Intrusion Detection System For Embedded Systems.
Inventors include Mr. Pawan Gawade; Varad Raverkar; Rashi Dekate; and Uttkarsha Mane.
The application for the patent was published on October 09, 2026, under issue no. 41/2026.
Abstract: The present invention relates to an AI-Powered Cybersecurity Based Intrusion Detection System for Embedded Systems, particularly designed for real-time detection of anomalous and potentially malicious network activities in resource-constrained embedded and Internet of Things (IoT) environments. The proposed system integrates embedded hardware, network traffic monitoring, lightweight machine learning, and Tiny Machine Learning (TinyML) technologies to provide an autonomous and computationally efficient cybersecurity mechanism. The system comprises an embedded processing unit, preferably an ESP32 microcontroller, configured to capture and preprocess network traffic and extract relevant traffic characteristics for anomaly analysis. An Isolation Forest-based machine learning model is deployed on the embedded processing unit to identify deviations from normal network behavior and detect suspicious activities without depending exclusively on predefined attack signatures or continuous cloud connectivity. The trained lightweight model enables localized inference with reduced computational, memory, and energy requirements, thereby making the system suitable for low-power embedded platforms. Upon detection of anomalous network behavior exceeding a predefined or dynamically determined threshold, the system generates an immediate security alert through one or more visual and/or audible indicators, including LEDs and buzzers. The system may further record detected events and corresponding traffic parameters for subsequent analysis, monitoring, or system-level response. By performing network traffic processing and threat inference locally at the edge, the invention reduces detection latency, minimizes dependency on remote servers and cloud infrastructure, enhances data privacy, and improves operational reliability during intermittent or unavailable internet connectivity. The proposed system provides a scalable, low-cost, and energy-efficient cybersecurity solution for smart homes, industrial IoT networks, healthcare devices, smart infrastructure, and other embedded applications requiring continuous and autonomous intrusion monitoring and rapid anomaly detection.
Disclaimer: Curated by HT Syndication.
