Identification Techniques in the Internet of Things: Survey, Taxonomy and Research Frontier

Abstract

The widespread adoption of the Internet of Things (IoT) has led to a notable rise in the number of interconnected smart devices. Within this IoT environment, we perceive an immense diversity of IoT devices originating from numerous manufacturing entities. They differ in size, form, storage, and processing capability, as well as in their set of features, capabilities, and limitations. In this heterogeneous environment, one of the most arduous tasks is to guarantee security and privacy, especially for resource-constrained devices that possess limited power and low processing capability. Due to these limitations, these devices often lack robust security functions, rendering them prone to malevolent actors or unauthorized users who can access private information or alter device functions. In this context, one of the most critical security challenges in this domain is to accurately identify each device in an IoT network. The efficient management of the IoT ecosystem hinges on the ability to uniquely identify each connected device. This, in turn, simplifies the process of monitoring, updating, and troubleshooting these devices for administrators, contributing to the establishment of a trustworthy and secure IoT network. To cope with this challenge, surveying the existing approaches within the IoT identification literature is of paramount importance to provide the IoT research community with relevant insights into the latest achievements. Furthermore, we aim to provide a thorough and comprehensive study of the key literature related to IoT identification research by elaborating proposals that rely on various techniques covering the physical, network, and application layers of the IoT ecosystem. We conclude this survey by discussing the persistent challenges and unresolved issues that may impede the mainstream adoption of IoT, prohibiting several entities from fully leveraging the potential of this evolving technology.