MUMBAI, India, Aug. 10 -- Intellectual Property India has published a patent application (202411103979 A) filed by Lovely Professional University on December 28, 2024, for Decentralized Federated Learning For Blockchain Security And Privacy In Penetration Testing.
Inventor includes Muhammed Rafeeq War.
The application for the patent was published on July 31, 2026, under issue no. 31/2026.
Abstract: The invention relates to blockchain security and privacy through a decentralized federated learning system for penetration testing. It consists of a network of blockchain nodes, each hosting a local federated learning model trained on local data like transaction histories and smart contract details without sharing data itself. A global model aggregator collects and aggregates updates from each node to form a global model, which is redistributed. Each node includes a penetration testing module using the federated learning model to simulate attacks and identify vulnerabilities, such as double-spending and 51% attacks, with local data. A security collaboration interface allows multiple organizations to share insights and threat detection securely without exposing proprietary data. Privacy mechanisms within the federated learning framework and testing module ensure data privacy through encrypted updates and secure aggregation. The invention includes a method for deploying the federated learning framework across a blockchain network, facilitating secure collaboration among organizations.
Disclaimer: Curated by HT Syndication.
