Abstract
The Internet of Things (IoT) is a vast collection of interconnected sensors, devices, and services that share data and information over the Internet with the objective of leveraging multiple information sources to optimize related systems. The technologies associated with the IoT have significantly improved the quality of many existing applications by reducing costs, improving functionality, increasing access to resources, and enhancing automation. The adoption of IoT by industries has led to the next industrial revolution: Industry 4.0. The rise of the Industrial IoT (IIoT) promises to enhance factory management, process optimization, worker safety, and more. However, the rollout of the IIoT is not without significant issues, and many of these act as major barriers that prevent fully achieving the vision of Industry 4.0. One major area of concern is the security and privacy of the massive datasets that are captured and stored, which may leak information about intellectual property, trade secrets, and other competitive knowledge. As a way forward toward solving security and privacy concerns, we aim in this paper to identify common input-output (I/O) design patterns that exist in applications of the IIoT. These design patterns enable constructing an abstract model representation of data flow semantics used by such applications, and therefore better understand how to secure the information related to IIoT operations. In this paper, we describe communication protocols and identify common I/O design patterns for IIoT applications with an emphasis on data flow in edge devices, which, in the industrial control system (ICS) setting, are most often involved in process control or monitoring.