Towards Edge-Based Deep Learning in Industrial Internet of Things
Fan Liang, Wei Yu, Xing Lu, David W. Griffith, Nada T. Golmie
As a typical application of the Internet of Things (IoT), the Industrial Internet of Things (I- IoT) connects all the related IoT sensing and actuating devices ubiquitously so that the monitoring and control of numerous industrial systems can be realized. Deep learning, as one viable way to conduct big data-driven modeling, can be integrated in I-IoT systems to aid the automation and intelligence of I-IoT systems. As deep learning requires large computation power, it is commonly deployed in cloud servers. Therefore, the data collected by IoT devices must be transmitted to the cloud for training, contributing to network congestion and affecting the IoT network performance as well as the supported applications. To address this issue, in this paper we propose an edge-based deep learning model, which utilizes edge computing to migrate the deep learning process from cloud servers to edge nodes, reducing data transmission demands in the I-IoT network and mitigating network congestion. Since edge nodes have limited computation compared to server hardware, we design a mechanism to optimize the deep learning model so that its requirements for computational power can be reduced. To evaluate our proposed solution, we design a testbed implemented in the Google cloud and deploy the proposed Convolutional Neural Network (CNN) model, utilizing a real-world I-IoT dataset to evaluate our approach. Our extensive experimental results demonstrate the effectiveness of our approach, which can not only reduce the network traffic overhead for I-IoT, but also maintain the classification accuracy in comparison with several baseline schemes.