MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202641116100 A) filed by Konerulakshmaiah Education Foundation on September 28, 2026, for An Intelligent Generative Adversarial Network-Based Ransomware Detection Model Incorporating An Enhanced Normalization Mechanism For Network Security.

Inventors include G Badrinath; and Dr. Alampallysreedevi.

The application for the patent was published on October 02, 2026, under issue no. 40/2026.

Abstract: ABSTRACT OF THE INVENTION: The present invention discloses an intelligent Generative Adversarial Network (GAN)-based model for detecting ransomware activity in computer networks, characterised by the incorporation of an enhanced normalization mechanism. Multi-dimensional feature vectors are extracted from network traffic flows and optional host behavioural indicators. These features are processed by an enhanced normalization module that combines adaptive instance normalization with learnable, traffic-context-conditioned scaling and shifting operations. The normalized features are supplied to a generator–discriminator pair. The generator synthesises realistic ransomware-like samples, while the discriminator performs simultaneous real/fake discrimination and benign-versus-ransomware classification. The enhanced normalization stabilises adversarial training across heterogeneous network traffic distributions, reduces mode collapse and improves generalisation to zero-day and polymorphic ransomware variants. During deployment only the discriminator (or a distilled compact network) is required. Incoming feature vectors are normalised and scored, producing a continuous risk value and a binary decision. Experimental evaluation on datasets containing multiple ransomware families and synthetic polymorphic samples demonstrates accuracy exceeding 98 %, an F1-score of approximately 97.6 % and a false-alarm rate below 2 %, substantially outperforming classical machine-learning, standard deep-neural-network and vanilla-GAN baselines. The invention is suitable for integration into network security appliances, next-generation firewalls, intrusion-detection systems and cloud-based threat-detection platforms, enabling timely identification and mitigation of ransomware-related network behaviour.

Disclaimer: Curated by HT Syndication.