MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202641048630 A) filed by Valli Priyadharshini K; R V V N Bheema Rao; Dr. B. Narendra Kumar; Chandrakala V; Suresh S; and Dr. B. Gunasundari. on April 16, 2026, for Autonomous Ai System For Detecting And Mitigating Threats In Encrypted Network Traffic.
Inventors include Valli Priyadharshini K; R V V N Bheema Rao; Dr. B. Narendra Kumar; Chandrakala V; Suresh S; and Dr. B. Gunasundari.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: The rapid growth of enciypted network communications has improved data privacy but also created challenges in detecting cyber threats hidden within encrypted traffic. Traditional signature-based security mechanisms are ineffective when packet payloads are encrypted, making it necessary to analyze traffic behavior and metadata rather than content. This study proposes an autonomous Al-based system for detecting and mitigating threats in encrypted network traffic using a hybrid deep learning framework that combines Long Short-Term Memory (LSTM) networks and Random Forest classification. The system extracts statistical and flow-based features from encrypted traffic, such as packet size distribution, inter-arrival time, flow duration, and byte rate. These features are preprocessed and fed into an LSTM network to capture temporal patterns in network flows. The learned representations are then classified using a Random Forest model to identify benign or malicious traffic. Additionally, an autonomous mitigation module dynamically blocks suspicious IP addresses and isolates abnormal traffic flows in real time. The proposed model effectively detects malicious patterns in enciypted environments while maintaining a low false-positive rate, thereby improving the reliability and efficiency of network security monitoring.
Disclaimer: Curated by HT Syndication.
