Optimizing Tradeoff Between Learning Speed and Cost for Federated-Learning-Enabled Industrial IoT

A combination of Industrial Internet of Things (IIoT) and federated learning (FL) is deemed as a promising solution to realize Industry 4.0 and beyond. However, scheduling more IIoT devices engaged in FL contributes to accelerated learning speed, but resulting in increased learning cost in terms of energy consumption and model accuracy reduction. In this article, we investigate the tradeoff between learning speed and cost in a three-layer FL-enabled IIoT system. Particularly, a weighted learning utility function is designed by capturing such a tradeoff. We aim to maximize the weighted learning utility in an FL training round by jointly optimizing the edge association as well as the allocations of resource block, computation capacity, and transmit power of the IIoT device. The resulting problem is a nonconvex and mixed-integer optimization problem, and consequently, it is difficult to solve. We thereby decompose the original problem into three subproblems, and then propose an overall alternating optimization algorithm to solve the subproblems iteratively until convergence. Via experimental results, it is demonstrated that the proposed scheme significantly improves the system-wide learning utility as compared to other baseline schemes. It is also shown that the proposed scheme can achieve the optimized tradeoff between learning speed and learning cost.

I. Introduction

The Industrial Internet of Things (IIoT) has emerged as a promising technology that drives greater efficiency and productivity in industries, thus attracting huge attention from academia, manufacturers, and plant designers [1], [2], [3]. A variety of new IIoT applications are developed in manufacturing and industrial processes, from digital factory and smart factory to Industry 4.0 and beyond. As an example of smart factory, a large number of IIoT devices are connected and synchronized to monitor, collect, and transmit industrial data on the field for meeting industrial expectations of security and reliability. Particularly, the industrial data is aggregated and transmitted to either the edge servers (ESs) for time-critical IIoT applications, or the central cloud server (CS) for further processing and analyzing [3], [4]. However, due to the communication resource constraints and the data privacy concerns, it may no longer be suitable for all IIoT devices to upload their industrial data to the CS for training data models. As a distributed learning paradigm, federated learning (FL) has been developed to allow multiple IIoT devices to cooperatively learn a global model from the locally trained model parameters without aggregating the raw data to the CS [5], [6], [7], [8], [9], [10]. This helps reduce communication overheads while alleviating the privacy concerns of industrial data sharing in IIoT systems.