Federated Learning (FL) is revolutionizing edge networks by enabling decentralized Machine Learning (ML) while preserving data privacy. This chapter explores the integration of Federated Learning in Edge Networks, highlighting its role in distributed intelligence, real-time decision-making, and adaptive learning at the edge. Unlike traditional centralized learning, FL allows models to be trained locally on edge devices—such as IoT sensors, mobile devices, and autonomous systems—without transferring raw data, ensuring privacy, security, and bandwidth efficiency. We discuss key architectural frameworks, communication protocols, and optimization techniques that enhance FL’s performance in edge environments. Challenges such as heterogeneous data distribution, resource constraints, security vulnerabilities, and model aggregation are examined, along with recent advancements in privacy-preserving mechanisms, including differential privacy and secure multiparty computation. Additionally, the chapter presents real-world applications of FL in smart cities, healthcare, autonomous systems, and industrial IoT, demonstrating its potential to drive intelligent, decentralized decision-making. By connecting FL with Edge Computing (EC) and AI-based analytics, this work provides insights into the future of privacy-centric, scalable, and efficient AI solutions at the edge. The discussion offers a roadmap for researchers and practitioners to leverage FL-driven intelligence in dynamic, resource-constrained edge networks.