Resilient Federated Learning Framework for 6G

Federated Learning enables collaborative model training while preserving data privacy, making it ideal for 5G, 6G, and Internet of Things environments. However, FL faces challenges such as communication costs, node failures, and scalability in dynamic networks. This paper proposes a Resilient Federated Learning Framework that enhances robustness by leveraging Zenoh for efficient communication. Results demonstrate improved training speed and reliability in heterogeneous networks, making FL more adaptable for real-world deployment.