Intellectual Property India Publishes Patent Application for 'Federated Deep Reinforcement Learning Driven Adaptive Ddos Defense For Smart Grid Communication Networks' Filed by Akula Vishalakshi; and Dr. T. Harikrishna

MUMBAI, India, Feb. 27 -- Intellectual Property India has published a patent application (202641017586 A) filed by Akula Vishalakshi; and Dr. T. Harikrishna, Hyderabad, Telangana, on Feb. 17, for 'federated deep reinforcement learning driven adaptive ddos defense for smart grid communication networks.'

Inventor(s) include Akula Vishalakshi; and Dr. T. Harikrishna.

The application for the patent was published on Feb. 27, under issue no. 09/2026.

According to the abstract released by the Intellectual Property India: "The present invention discloses a system and method for an adaptive, privacy-preserving Distributed Denial of Service (DDoS) defense in Smart Grid communication networks. The system employs a distributed architecture comprising multiple intelligent defense agents deployed at strategic edge nodes, such as substation gateways. Each agent integrates a local traffic analyzer with a Deep Reinforcement Learning (DRL) engine. The traffic analyzer continuously monitors network flows to generate a real-time state representation, which is fed into the DRL engine. The DRL engine, trained via a reward function that balances network utility and security, autonomously selects and executes optimal mitigation actions, such as dynamic rate-limiting or traffic blocking. Crucially, the agents operate within a federated learning framework. Instead of sharing raw, sensitive network data, each agent periodically uploads its learned DRL model weights to a central federated aggregation server. This server securely aggregates the updates from all agents to create a continuously improving global DRL model, which is then redistributed. This collaborative cycle enables the entire grid's defense system to collectively learn from localized attacks, rapidly adapt to new threats, and maintain resilience without central bottlenecks or compromising data privacy. The invention provides a robust, self-optimizing, and scalable immune system against evolving cyber threats to critical power infrastructure."

Disclaimer: Curated by HT Syndication.