Abstract
Blockchain has the characteristics of decentralization, tamper proof, and untraceability and has been used in various areas. Nowadays, users are increasingly concerned about their privacy. However, anonymity poses a huge challenge on anti-crime, such as anti-money laundering, counter-terrorism financing and tax evasion. In order to achieve a broader use of blockchain, a trade-off between privacy and regulation must be achieved. We study the contradiction between privacy-preserving payment mechanisms including controllable anonymity, and penetrating regulation on digital currency. We then propose a solution based on homomorphic encryption and federated learning. Finally, a digital currency prototype is implemented with corda framework to verify our strategies.