A Survey on Secure Blockchain-Based Federated Learning for Privacy-Preserving Data Sharing and Protection Against Unauthorized Access

The integration of blockchain with federated learning (FL) has emerged as a promising solution to several challenges in Machine Learning (ML), particularly for decentralized data processing across diverse domains. This survey explores the application of this integrated approach in health care, smart cities, smart agriculture, smart grids, and other applications. The blockchain ensures privacy and integrity of medical data in health care and smart cities, while challenges exist in scalability and latency. Meanwhile, the FL allows collaborative training of models without compromising the security of the data. Hence, the integrated approach is preferred for application in smart agriculture for the optimization of resources and in smart grids for the secure handling of energy data and optimization of grid operations. Further, the survey looks into the other emerging applications such as autonomous vehicles, Internet of Things (IoT) networks, and industrial automation. Despite all these potentials, there are critical challenges that persist, such as real-time processing, system scalability, and model synchronization. This survey brings to light detailed examinations of existing methods, draws up a list of merits and demerits associated with every application and has undertaken comparative surveys in an assessment of effectiveness as an innovative measure using the blockchain-enabled FL. Finally, future research directions are proposed to address the limitations and enhance the applicability of these technologies in emerging applications.