With the development of artificial intelligence and Internet of Things (IoT), the era of industry 4.0 has come. According to the prediction of IBM, with the continuous popularization of 5G technology, the IoT technology will be more widely used in factories. In recent years, federated learning has become a hot topic for Industrial Internet of Things (IIoT) researchers. However, many devices in the IIoT currently have a problem of low computing power, so these devices cannot perform well facing the tasks of training and updating models in federated learning. In order to solve the above problems, we introduce edge computing into the IIot, so that the device can complete the federated learning operation. In order to ensure the security of data transmission, blockchain is introduced as the main algorithm of equipment authentication in the system. What's more, in order to increase the efficiency and versatility of training model in IIoT, we introduce transfer learning to improve the system performance. The experimental results show that our algorithm can achieve high security and high training accuracy.

The Internet of Things (IoT) and edge computing technologies have been rapidly developing in recent years, leading to the emergence of new challenges in privacy and security. Personal privacy and data leakage have become major concerns in IoT edge computing environments. Federated learning has been proposed as a solution to address these privacy issues, but the heterogeneity of devices in IoT edge computing environments poses a significant challenge to the implementation of federated learning. To overcome this challenge, this paper proposes a novel node selection strategy based on deep reinforcement learning to optimize federated learning in heterogeneous device IoT environments. Additionally, a metric model for IoT devices is proposed to evaluate the performance of different devices. The experimental results demonstrate that the proposed method can improve training accuracy by 30% in a heterogeneous device IoT environment.

No abstract

No abstract