MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202641096798 A) filed by Andhra University on August 11, 2026, for Enhanced Firefly Whale Optimization With Stacked Cnn And Rnn Layers Cyber Attacks Detection And Categorization.
Inventors include Katikam Mahesh; and Prof. Kunjam Nageswara Rao.
The application for the patent was published on August 14, 2026, under issue no. 33/2026.
Abstract: The present invention discloses an adaptive cyberattack detection and categorization system and method using an enhanced Firefly–Whale Optimization mechanism integrated with stacked Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) layers. The system receives network traffic data and performs preprocessing including normalization, encoding, transformation, and traffic representation. The enhanced Firefly–Whale Optimization mechanism searches the network-traffic feature space to identify an optimal subset of informative features and optionally optimizes selected parameters of the stacked CNN–RNN architecture. The stacked CNN processes the optimized feature representation to extract hierarchical and discriminative traffic characteristics, while the stacked RNN learns temporal and sequential dependencies among the extracted representations. A classification mechanism determines normal and malicious network activities and categorizes detected malicious traffic into corresponding cyberattack classes. The hybrid optimization mechanism combines Firefly-based attraction and movement operations with Whale-based exploration and exploitation operations to improve the search process. The invention reduces redundant feature processing, false detections, and computational overhead while providing an adaptive framework for accurate cyberattack detection and categorization in networked computing environments.
Disclaimer: Curated by HT Syndication.
