Intellectual Property India Publishes Patent Application for A System And Method For Federated Intrusion Detection Using Secure Multi-Party Model Aggregation And Encrypted Traffic Feature Extraction Architecture

MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641056133 A) filed by Rmk Engineering College on May 03, 2026, for A System And Method For Federated Intrusion Detection Using Secure Multi-Party Model Aggregation And Encrypted Traffic Feature Extraction Architecture.

Inventors include Dr Shanthi M; Dr Sandra Johnson; Dr. P. Shobha Rani; Dr. S Srijayanthi; and Dr. Gladiss Merlin N. R.

The application for the patent was published on June 26, 2026, under issue no. 26/2026.

Abstract: The present invention relates to a system and method for federated intrusion detection using secure multi-party model aggregation and encrypted traffic feature extraction architecture. The invention addresses critical challenges in modern cybersecurity systems, particularly data privacy, secure collaboration, and efficient threat detection across distributed network environments. The proposed system utilizes a federated learning framework in which multiple client nodes independently capture and process network traffic data while ensuring that sensitive information is protected through encryption techniques. Each node performs local feature extraction and model training using machine learning algorithms, thereby eliminating the need to share raw data. The locally trained models generate encrypted updates that are transmitted to a central aggregation server, where secure multi-party computation techniques are applied to combine model parameters without exposing individual contributions. The aggregated global model is redistributed to client nodes, enabling continuous and collaborative learning. The invention enhances intrusion detection accuracy by leveraging diverse datasets while maintaining strict confidentiality and compliance with data protection regulations. Furthermore, the system provides real- time detection and alert mechanisms for identifying malicious activities, ensuring robust and scalable cybersecurity solutions for enterprise and cloud-based infrastructures. Keywords:Federated Learning, Intrusion Detection System, Secure Multi-Party Computation, Encrypted Traffic Feature Extraction, Cybersecurity

Disclaimer: Curated by HT Syndication.