MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541124333 A) filed by Shaik Hasane Ahammad, Vijayawada, Andhra Pradesh, on Dec. 9, 2025, for 'adaptive dl architecture for iov/icv network anomaly monitoring.'
Inventor(s) include Bethi Saiteja; and V Sudha Rani.
The application for the patent was published on Jan. 2, under issue no. 01/2026.
According to the abstract released by the Intellectual Property India: "This work proposes a scalable deep learning platform for automated network monitoring and anomaly interpretation in Internet of Vehicles (IoVs) and Intelligent Connected Vehicles (ICVs). The system integrates heterogeneous data from onboard units, roadside infrastructure, and cloud services to enable continuous, real-time situational awareness. A hybrid deep learning model combining CNN, BiGRU, and Transformer layers performs hierarchical feature extraction and dynamic behavior modeling to detect anomalies with 99% accuracy and a 99% true positive rate (TPR). An adaptive learning module addresses concept drift, evolving cyber-attack vectors, and changes in vehicular mobility patterns. Detected anomalies are further analyzed through an explainable AI (XAI) layer using SHAP and LIME to generate interpretable threat reports, risk scores, and actionable insights. The platform supports edge-cloud scalability, ensuring low-latency inference and continuous global model updates. This unified architecture significantly enhances cybersecurity, reliability, and resilience for next-generation IoV and ICV ecosystems."
Disclaimer: Curated by HT Syndication.
