Blockchain technology is characterized by decentralization and immutability, which enhances the credibility and reliability of systems. However, these features pose numerous challenges in the context of increasingly complex blockchain application platforms. Distributed Denial of Service (DDoS) attacks pose a significant threat to blockchain technology due to the inherent distributed architecture and operational characteristics of blockchain networks. However, the limited computing resources in blockchain environments, especially on low-performance devices, make it difficult to support complex DDoS detection mechanisms. At the same time, Federated Learning (FL), a distributed machine learning approach, enables collaborative detection while maintaining data privacy. However, the efficiency of FL is limited in resource-constrained environments. This makes it difficult to meet real-time detection requirements. This paper reviews the current state of research on DDoS Attack Detection mechanisms within blockchain environments, focusing on how to optimize FL methods under resource constraints to enhance detection efficiency. By examining the existing literature, the paper proposes a DDoS Attack Detection framework based on asynchronous adaptive FL and explores the challenges associated with its practical implementation and future research directions.