In today’s business world, it is crucial to protect company data from cyber threats and attacks. However, traditional intrusion detection systems with centralized data collection raise privacy concerns as well as scalability problems in network growth. To address these issues, the project proposes a new system utilizing federated learning and blockchain technology for decentralized intrusion detection in software-defined networks (SDNs). The proposed project utilizes the collaborative power of federated learning, which is a sub-field of machine learning where a group of distributed computers collectively train intrusion detection models locally without compromising its own sensitive data. Additionally, the integration of blockchain maintains immutability and integrity of model updates ensuring security and trust in the system. By employing federated learning across distributed nodes, the proposed system continuously refines models for threat detection, including zero-day attacks.