Radio Frequency Fingerprint Identification for 5G Mobile Devices Using DCTF and Deep Learning

Radio frequency fingerprint (RFF) identification is a promising technique for identifying Internet of Things (IoT) devices. This paper presents a comprehensive survey on RFF identification, which covers various aspects ranging from related definitions to details of each stage in the identification process, namely signal preprocessing, RFF feature extraction, further processing, and RFF identification. Specifically, three main steps of preprocessing are summarized, including carrier frequency offset estimation, noise elimination, and channel cancellation. Besides, three kinds of RFFs are categorized, comprising I/Q signal-based, parameter-based, and transformation-based features. Meanwhile, feature fusion and feature dimension reduction are elaborated as two main further processing methods. Furthermore, a novel framework is established from the perspective of closed set and open set problems, and the related state-of-the-art methodologies are investigated, including approaches based on traditional machine learning, deep learning, and generative models. Additionally, we highlight the challenges faced by RFF identification and point out future research trends in this field.

With the rapid development of the cognitive radio networks, the number of terminal devices has exploded. Massive devices generate a large amount of privacy-sensitive data, typically WiFi signals. This paper proposes a method for Radio frequency (RF) fingerprinting identification of WiFi signals based on federated learning, which trains a cooperative model to complete RF fingerprinting identification without transmitting privacy-sensitive data. The experimental findings on a real-world dataset validate that the strategy described in this study increases the RF fingerprinting identification accuracy in a variety of size circumstances, and ensures that data privacy will not be compromised.

Aiming at the actual application scenario of radio frequency (RF) fingerprint identification, a low-cost, small-volume embedded system for specific identification of RF devices is designed and implemented, and the improved VGG-16 model is used to extract the RF fingerprint features of RF signals. The hardware of the system consists of two parts: the RF Signal Acquisition Module and the RF Fingerprint Identification Module. The RF Signal Acquisition Module uses Zedboard as the main control board, and the AD9361 chip is the RF signal receiving circuit; the RF Fingerprint Identification Module uses FPGA to accelerate the deep learning algorithm, and displays the identification results on the host computer. Using 6 wireless routers of the same model for testing, the system has an identification accuracy rate of 96% for wireless routers in the actual electromagnetic environment, and the real-time identification feedback time is about 6 seconds, which has a certain practical application value.

A frequency-domain cepstrum correlation and temporal transient filtered radio frequency fingerprints (RFF) blind extraction method is proposed for long term evolution uplink device authentication in this work. With the merits of the constant-modulus nature of the Zadoff-Chu (ZC) sequence and the cyclic prefix (CP) structure of the physical random access channel (PRACH) frame, the device RFF integrated in both the frequency-domain ZC sequence and the transient ramp-on/ramp-off sequences are extracted. For the frequency-domain RFF extraction, the cepstrum of the ZC sequence is calculated to enlarge the gap between the constant RFF and the random fading effects, and the correlation between two cepstrum fragments is then calculated to mitigate the random fading components. For the temporal RFF extraction, the ramp-on and ramp-off features induced by the device-specific power amplifier are extracted by exploiting the CP structure of the PRACH frame, where the CP part and the duplicate part are adaptively filtered and the final adapted weight of the least mean square filter is used as RFF. No prior PRACH parameter knowledge is required in RFF extraction and both the frequency-domain and the temporal RFF are combined as the input of the convolutional neural network to enhance the classification accuracy. With four-months sampling of PRACH signals emitted from universal software radio peripheral devices, the experimental results show that the authentication accuracy of the proposed combined RFF can reach as high as 99.9% in the line-of-sight (LOS) scenario and 99.6% in the non-LOS scenario.

In physical-layer security schemes, radio frequency fingerprint (RFF) identification of WiFi devices is susceptible to receiver differences, which can significantly degrade classification performance when a model is trained on one receiver but tested on another. Consequently, the receiver-agnostic physical-layer identification of WiFi devices remains an important research challenge. Recent studies have addressed this issue by either training on signals collected from multiple receivers to extract receiver-agnostic RFF features, or by employing calibration processes to enhance classification accuracy. However, these approaches introduce substantial computational overhead. In this paper, we propose a division-based receiver-agnostic RFF extraction method for WiFi systems, which removes the receivers' effects by dividing different preambles in the frequency domain. The proposed method requires only a single receiver for training and does not rely on additional calibration or stacking processes. First, for flat fading channel scenarios, the legacy short training field (L-STF) and legacy long training field (L-LTF) of the unknown device are divided by those of the reference device in the frequency domain. The receiver-dependent effects can be eliminated with the requirement of only a single receiver for training, and the higher-dimensional RFF features can be extracted. Second, for frequency-selective fading channel scenarios, the high-throughput long training field (HT-LTF) is divided by the L-LTF part in the frequency domain. Only a single receiver is required for training and the higher-dimensional RFF features that are both channel-invariant and receiver-agnostic are extracted. Finally, simulation and experimental results demonstrate that the proposed method effectively mitigate the impacts of channel variations and receiver differences. The classification results show that, even when training on a single receiver and testing on a different one, the proposed method achieves classification accuracy improvements of 15.5% and 28.45% over the state-of-the-art approach in flat fading and frequency-selective fading channel scenarios, respectively.

The Physical random access channel (PRACH) of the fifth generation of mobile communication (5G) with a new radio-access (NR) is composed of cyclic prefix (CP) and preamble sequence. While the CP of PRACH guarantees sufficient length for full sequence detection to PRACH detector, it also hampers finding out the PRACH start point since the CP itself causes a considerable pseudo-peak at PRACH detection. Ruling out this pseudo-peak using threshold contains risk of mis-judgement depending on fading environment or PRACH searching time offset. In this paper, we propose a comprehensive PRACH detection procedure to find out PRACH sequence and the start point of uplink correctly.

Given the current radio frequency(RF) fingerprint identification methods based on deep learning, there is a problem of poor recognition performance in the scene of RF fingerprint offset. This paper designs an Internet of Things (IoT) devices identification system based on software defined radio(SDR) and transfer learning (TL) technology. The main advantage of this system is to use the universal software radio peripheral (USRP) to collect the RF signals of IoT devices and identify the RF signals preprocessed by SDR based on transfer learning. This system can reduce data redundancy and model complexity, has the characteristics of lightweight and high efficiency and meet different experimental scenarios' needs. The experimental results show that compared to reducing the recognition accuracy based on the deep learning method in RF fingerprint offset, transfer learning dramatically improves the recognition performance and reduces the model training time. Simultaneously, the experimental process is in a natural wireless communication environment, and the actual application has high reliability.

Radio frequency fingerprint (RFF) identification aims to identify emitters by extracting the inherent physical-layer features. Despite the significant accuracy achieved in a supervised learning manner, few works have focused on unsupervised RFF identification. To this end, we propose an unsupervised training paradigm based on the curriculum learning framework. Specifically, the unlabeled samples are first sorted by the signal-to-noise (SNR) of signals. Then, the network is pretrained with contrastive learning on the high SNR subset. To further improve identification performance, the model is gradually optimized from the easy (high SNR) to the hard (low SNR) subsets with pseudo label learning. Experimental results indicate that the proposed scheme achieves comparable performance against schemes based on supervised learning.

To support a high density of mobile devices and its connections by means of rapid increase of mobile devices as well as small cells, various studies related to random access such as the random access load, better uplink resource utilization and interference among random access attempts have been developed for enhanced long-term evolution (LTE) system and 5G communication technology. In this paper, we discuss about the technical issues of random access resource configuration in time division duplex (TDD) mode and propose an improved method of it to make better uplink resource utilization on frequency domains and to minimize interference among random access attempts for physical random access channel (PRACH) resource allocation point of view, but it can also easily extend the LTE system. The proposed method is compared with LTE TDD PRACH configuration method through theoretical analysis. It shows that the proposed method provides better resource utilization for PRACH and physical uplink shared channel resources (PUSCH), allows flexible resource utilization in uplink scheduling, and is intuitive and effective in cell deployment with random access channel optimization in terms of random access resource configuration.

Adversarial attacks and active defense on deep learning based identification of GaN power amplifiers under physical perturbation

In this paper, we introduce the Preamble Priority-Aware (PPA) Packet Downlink Control Channel (PDCCH) resources allocation algorithm to provide Quality of Service (QoS) differentiation in the Random Access (RA) procedure of the LTE-Advanced technology. The PPA algorithm uses the preamble priority defined by the RA procedure and Radio Access Network (RAN) overload control schemes to make scheduling decisions. Results derived via simulation show that the proposed PDCCH algorithm significantly increases the chance of accessing the network as well as reducing random-access delays for user equipment employing prioritized preamble sequences. Thus, the proposed algorithm provides enhanced QoS support to prioritized users during intense RA attempts.

Radio frequency fingerprint identification (RFFI) based on physical-layer characteristics provides a reliable solution for secure authentication of Internet of Things (IoT) devices. Deep neural networks have demonstrated strong capability in improving RFFI performance; however, their high computational complexity and large parameter size pose significant challenges for deployment on resource-constrained edge devices. In RFFI tasks, existing pruning methods often lack effective performance recovery strategies, which leads to noticeable degradation in identification accuracy after pruning. To address this issue, this paper proposes a pruning method based on adversarial learning and polarization regularization. Polarization regularization is applied to learnable soft masks to effectively distinguish channels to be pruned from those to be retained. In addition, an adversarial learning-based performance recovery strategy is introduced to align the output feature distributions between the baseline network and the pruning network, thereby improving identification accuracy after pruning. Experimental results on multiple RFFI datasets demonstrate that the proposed method can effectively prune ResNet18 and VGG16, achieving substantial reductions in model complexity with only minor losses in identification performance.

Satellite communication system is expected to play a vital role for realizing various remote Internet-of-Things (IoT) applications in sixth-generation vision. Due to unique characteristics of satellite environment, one of the main challenges in this system is to accommodate massive random access (RA) requests of IoT devices while minimizing their energy consumptions. In this article, we focus on the reliable design and detection of RA preamble to effectively enhance the access efficiency in high-dynamic low-earth-orbit (LEO) scenarios. To avoid additional signaling overhead and detection process, a long preamble sequence is constructed by concatenating the conjugated and circularly shifted replicas of a single root Zadoff-Chu (ZC) sequence in RA procedure. Moreover, we propose a novel impulse-like timing metric based on length-alterable differential cross-correlation (LDCC), that is immune to carrier frequency offset (CFO) and capable of mitigating the impact of noise on timing estimation. Statistical analysis of the proposed metric reveals that increasing correlation length can obviously promote the output signal-to-noise power ratio, and the first-path detection threshold is independent of noise statistics. Simulation results in different LEO scenarios validate the robustness of the proposed method to severe channel distortion, and show that our method can achieve significant performance enhancement in terms of timing estimation accuracy, success probability of first access, and mean normalized access energy, compared with the existing RA methods.

Radio frequency (RF) fingerprinting technology has been developed as a unique method for maintaining security based on physical layer characteristics. In this paper, we propose the RF fingerprinting by extracting the parameter characteristics such as information dimension, constellation feature, and phase noise spectrum in the transmitted information when applied to the universal software radio peripheral (USRP) software-defined radio (SDR) platform. To achieve a great performance improvement of classification, not only the traditional support vector machine (SVM) classifier, but also the machine-based integrated classifier bagged tree and the adaptive weighting algorithm weighted k-nearest neighbor (KNN) are both discussed. It is demonstrated that the proposed method achieves good classification performance under different signal-to-noise ratios (SNR).

Radio frequency fingerprint recognition method based on prior information