MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202641095189 A) filed by P. Sree Manvitha; Ramdas Kapila; and P. Neelakantan on August 06, 2026, for Deep Feature Extraction And Variational Quantum Classifier For Network Intrusion Detection.
Inventors include P. Sree Manvitha; Ramdas Kapila; and P. Neelakantan.
The application for the patent was published on August 14, 2026, under issue no. 33/2026.
Abstract: ABSTRACT DEEP FEATURE EXTRACTION AND VARIATIONAL QUANTUM CLASSIFIER FOR NETWORK INTRUSION DETECTION The main design of the present invention discloses a Deep Feature Extraction and Variation Quantum Classifier System for Network Intrusion Detection, which comprises a a data preprocessing unit, a deep feature extraction unit, a feature compression unit, a quantum feature encoding unit, a variational quantum classifier unit, an intrusion decision unit, an alert and reporting unit and a system infrastructure comprising computing, storage, and processing modules. The main purpose of the present invention is to enable accurate detection of malicious network activities in large-scale and high-dimensional network environments by combining classical deep learning-based feature compression with quantum-enhanced classification mechanisms. The system preprocesses network traffic, extracts representative features, generates compressed feature embeddings, encodes them into quantum states, classifies network traffic into normal and attack categories and continuously updates model parameters to improve intrusion detection performance and adaptability to evolving cyber threats.
Disclaimer: Curated by HT Syndication.
