A Review of Recent Developments on Secure Authentication Using RF Fingerprints Techniques

Abstract

The Internet of Things (IoT) concept is widely used today. As IoT becomes more widely adopted, the number of devices communicating wirelessly (using various communication standards) grows. Due to resource constraints, customized security measures are not possible on IoT devices. As a result, security is becoming increasingly important in IoT. It is proposed in this study to use the physical layer properties of wireless signals as an effective method of increasing IoT security. According to the literature, radio frequency (RF) fingerprinting (RFF) techniques are used as an additional layer of security for wireless devices. To prevent spoofing or spoofing attacks, unique fingerprints appear to be used to identify wireless devices for security purposes (due to manufacturing defects in the devices' analog components). To overcome the difficulties in RFF, different parts of the transmitted signals (transient/preamble/steady-state) are used.
 This review provides an overview of the most recent RFF technique developments. It discusses various solution methods as well as the challenges that researchers face when developing effective RFFs. It takes a step towards the discovery of the wireless world in this context by drawing attention to the existence of software-defined radios (SDR) for signal capture. It also demonstrates how and what features can be extracted from captured RF signals from various wireless communication devices. All of these approaches' methodologies, classification algorithms, and feature classification are explained. 
 In addition, this study discusses how deep learning, neural networks, and machine learning algorithms, in addition to traditional classifiers, can be used. Furthermore, the review gives researchers easy access to sample datasets in this field.

FeRHA: Fuzzy-Extractor-Based RF and Hardware Fingerprinting Two-Factor Authentication

“…It has been shown that every transmitter has a distinct RF fingerprint, and it is also demonstrated how unlikely it is for two transmitters to share the same RF features. Accordingly, such a unique RF fingerprint can be used to authenticate the identity of a particular wireless transmitter and guarantee the confidentiality of the messages that are sent [48,49].…”

Considerations, Advances, and Challenges Associated with the Use of Specific Emitter Identification in the Security of Internet of Things Deployments: A Survey

“…Our survey differs from those presented by the authors of [93][94][95][96] that use IoT as motivation or context for their surveys but do not use IoT as a lens through which to analyze the surveyed SEI papers, as we do herein. This survey aims to answer: "What technical gaps must be addressed for SEI to be a viable PLS solution for IoT deployments?"…”

RF Signal Feature Extraction in Integrated Sensing and Communication

“…The resulting fingerprint features (feature vector) for each burst consist of 24D features (2 types of data × 2 types of features × 4 statistical features). The fingerprint features from device c can be given by Equation (30).…”

“…Other features, such as constellation trajectory mapping features [22,23], fractal dimensional features [24,25], and multidimensional integrated features, have also been studied more [26][27][28]. For databased intelligent processing methods, deep models are used for automatic feature extraction and individual classification and recognition [29][30][31][32][33][34]. However, the models used are becoming increasingly complex and require very large data samples to train, which are not suitable for application to micromodels or microsystems.…”