A Novel Framework for Data Trading Markets based on Blockchain-enabled Federated Learning

In recent years, with the continued technical advances and increasing prosperity of digital economy ecosystems, the value of data has been widely recognized, leading to a huge potential demand of establishing data markets for sharing and trading private data resources. The existing efforts of centralized data trading systems, however, cannot completely solve the longstanding issues of data security and privacy leakage, which motivates our work. In this paper, we designed a novel framework for data trading marketplace based on blockchain and federated learning, aiming at combining their technical advantages and solving the key issue of secured and decentralized data trading with privacy protection. Specially, we proposed a six-layered conceptual model for data markets, and each layer is focused on the key elements and technologies in digital assets, technical architecture, scheduling algorithms, incentive mechanisms, trading models and application scenarios, respectively. We also discuss the potential applications of our framework in areas including intelligent transportation and Internet of vehicles. Our work is targeted to offer useful references and guidance for designing a decentralized, secured and trusted data trading market, and also promoting the development of digital economy and industries.