Asynchronous Federated Learning (AFL) is an efficient and secure distributed machine learning paradigm designed to tackle challenges such as data silos and client heterogeneity. However, practical implementations often suffer from sporadic client participation due to suboptimal network conditions or varying device quality, resulting in significant latency for model updates. Additionally, data privacy is also a security issue that cannot be overlooked in Federated Learning. Nevertheless, existing adaptive federated optimization algorithms fall short in effectively addressing the dual challenges of high client-side latency and robust privacy protection. To address these challenges, we propose DPAAFL, an adaptive asynchronous federated learning framework with per-client differential privacy. DPAAFL improves the accuracy–efficiency trade-off by adopting a dual-event triggering rule that admits only sufficiently informative updates with acceptable latency, and by mitigating the impact of staleness through delay-adaptive aggregation; meanwhile, a buffered asynchronous update scheme reduces the communication overhead by aggregating updates in batches. In addition, each client perturbs its clipped update with calibrated Gaussian noise before transmission, providing differential privacy protection against privacy inference attacks under the honest-but-curious threat model. We provide theoretical analyses for both convergence and privacy. Extensive experiments demonstrate that the proposed scheme not only reduces the communication overhead and enhances system robustness, but also significantly improves the training efficiency and ensures the accuracy of the model. In comparison with contemporary state-of-the-art algorithms, the proposed scheme improves model accuracy and training efficiency by up to 3.5% and 29.3%, respectively.

Abstract The low transaction capacity, high transaction cost and long-term privacy concerns of the current Ethereum platform are notorious. Developers are seeking alternative blockchain platforms to migrate their blockchain-based applications to reduce their applications’ use-cost and improve their applications’ user experience. The Hyperledger Fabric (HLF) platform with resiliency, flexibility, scalability and confidentiality is preferred for developers to migrate their Ethereum blockchain-based applications. However, it is laborious for developers to migrate blockchain-based applications from the Ethereum platform to the HLF platform. In this paper, we first propose a complete and secure migration solution to ease the migration process. The main idea of our solution is to design a toolbox to help developers automatically eliminate the adverse effects that the differences between Ethereum and HLF may bring to the migrated application. Developers with the toolbox can migrate the application with little time and minimal modification. It is theoretically proved that the migrated application with the toolbox is secure. Besides, a prototype of the toolbox is implemented. The extensive experiments demonstrate that the time for the migration process is acceptable, and the toolbox has little impact on the migrated application’s performance.