A Comprehensive Survey on Radio Frequency (RF) Fingerprinting: Traditional Approaches, Deep Learning, and Open Challenges

Abstract

Fifth generation (5G) networks and beyond envisions massive Internet of Things (IoT) rollout to support disruptive applications such as extended reality (XR), augmented/virtual reality (AR/VR), industrial automation, autonomous driving, and smart everything which brings together massive and diverse IoT devices occupying the radio frequency (RF) spectrum. Along with spectrum crunch and throughput challenges, such a massive scale of wireless devices exposes unprecedented threat surfaces. RF fingerprinting is heralded as a candidate technology that can be combined with cryptographic and zero-trust security measures to ensure data privacy, confidentiality, and integrity in wireless networks. Motivated by the relevance of this subject in the future communication networks, in this work, we present a comprehensive survey of RF fingerprinting approaches ranging from a traditional view to the most recent deep learning (DL) based algorithms. Existing surveys have mostly focused on a constrained presentation of the wireless fingerprinting approaches, however, many aspects remain untold. In this work, however, we mitigate this by addressing every aspect - background on signal intelligence (SIGINT), applications, relevant DL algorithms, systematic literature review of RF fingerprinting techniques spanning the past two decades, discussion on datasets, and potential research avenues - necessary to elucidate this topic to the reader in an encyclopedic manner.

Transient‐based radio frequency fingerprinting with adaptive ensemble of transforms and convolutional neural network

“…In the DIFFSNR case, the classification accuracy of all the approaches is inferior to SAMESNR because the CNN is trained on data samples, which are increasingly different from the noisy samples for decreasing values of SNR in dB. This result for DIFFSNR is consistent with the findings from literature [4, 7, 8] (see Figure 3).…”

“…While older studies implemented RFF with the extraction of features (e.g. variance, entropy) [2, 5] or the envelope signal [6] and shallow machine learning algorithms, the application of deep learning (DL) algorithms to RFF have been quite successful [7, 8]. In particular, the study described in this letter applies a convolutional neural network (CNN), which was used with success in [4, 8].…”

“…This letter addresses a specific problem (highlighted in literature in [1, 4, 7]) of RFF using transients: the number of signal values related to the transient could be too limited to achieve a high identification accuracy. On the other side, this leads to the advantage that transforms applied to the transient part are also limited in space.…”

Secure Industrial IoT Systems via RF Fingerprinting Under Impaired Channels With Interference and Noise

“…Channel impairments and environmental circumstances affect accuracies of DL-based techniques while detecting RF fingerprints of transmitters [18], [19]. As it is infeasible to collect data in all environmental conditions, augmentationbased approaches for RF fingerprinting can be considered as an efficient tool to deal with channel impairments.…”

RF Fingerprinting Needs Attention: Multi-task Approach for Real-World WiFi and Bluetooth

“…T HE unprecedented adoption of Internet of Things (IoT) devices in everyday life especially WiFi and Bluetooth have exacerbated privacy and security concerns. Radio Frequency (RF) fingerprinting is envisioned as a physical layer security scheme that can be integrated with the existing upper layer security protocols to ensure robust wireless security [1], [2]. RF fingerprinting is a passive security scheme that can be implemented at a passive receiver radio whereby the wireless emitters can be distinguished from one another based on the minute imprint made in its emissions due to the imperfections in its RF circuitry.…”