MUMBAI, India, July 13 -- Intellectual Property India has published a patent application (202611052409 A) filed by Thapar Institute Of Engineering & Technology on April 24, 2026, for Xai-Enhanced Hybrid Deep Learning Framework For Iot Device Identification And Attack Detection.
Inventors include Dr. Aashima Sharma; Dr. Gurpal Singh Chhabra; Prabhav Jain; and Anshika Rathour.
The application for the patent was published on July 10, 2026, under issue no. 28/2026.
Abstract: The present invention relates to an artificial intelligence-based framework for Internet of Things (IoT) device identification and attack detection using network traffic analysis. The system utilizes machine learning and deep learning models, including Convolutional Neural Networks (CNN), for automated feature extraction, and machine learning-based classifiers such as Extreme Gradient Boosting (XGBoost) for classification of IoT devices and detection of network-based anomalies. Network traffic data comprising HTTPS attributes, TCP handshake metadata, user agent strings, and flow-based parameters is preprocessed through cleaning, normalization, and feature selection. The framework further incorporates an explainable artificial intelligence (XAI) technique, such as SHapley Additive exPlanations (SHAP), to provide interpretability by determining feature contributions. Feature contribution scores are utilized to refine feature selection and improve classification performance. The system is scalable, computationally efficient, and suitable for deployment on IoT gateways or edge devices for real-time or near real-time monitoring.
Disclaimer: Curated by HT Syndication.
