Investigating Federated Learning Implementation Challenges in 6G Network

The rapid progress and deployment beyond 5G and moving towards a 6G network demand the stipulation for network intelligence by applying advanced ML-driven approaches. The modern state-of-the-art ML-driven approaches in existing 5G network have various concerns of communication and computing resource scarcity, computing overhead and privacy issues. Federated learning (FL) incorporated 6G network will play a significant role in overcoming such challenges. However, the FL-incorporated 6G network suffers due to various implementation challenges considering 6G and FL. In this paper, the author discusses the potential of FL for the requirements of 6G and investigates various federated learning implementation challenges, followed by trends and developments of FL-incorporated massive edge intelligence (EI) in future 6G in a comprehensive way.