Blockchain and Federated Learning-enabled Distributed Secure and Privacy-preserving Computing Architecture for IoT Network

With the adoption of the 5G network, the exponential increase in the volume of data generated by the Internet of Things (IoT) devices, pushes the system to learn the model locally to support real-time applications. However, it also raises concerns about the security and privacy of local nodes and users. In addition, the approach such as collaborative learning where local nodes participate in the learning process of global model also raise critical concern regarding the cyber resilience of the network architecture. To address these issues, in this article, we identify the research gaps and propose a blockchain and federated learning-enabled distributed secure and privacy-preserving computing architecture for IoT network. The proposed model introduces the lightweight authentication and model training algorithms to build secure and robust system. The proposed model also addresses the reward and penalty issues of the collaborative learning with local nodes and propose a reward system scheme. We conduct the experimental analysis of the proposed model based on various parametric metrics to assess the effectiveness of the model. The experimental result shows that the proposed model is effective and capable of providing a cyber-resilience system.