Recent advancements in pretrained foundation models have shown considerable promise across various machine learning tasks. However, their application in the broader IoT industry remains limited, particularly in IoT physical-layer authentication. In this domain, the presence of domain shift between training and deployment environments, combined with privacy and data security concerns, renders traditional domain adaptation methods that rely on source domain data impractical. Motivated by these challenges, source-free unsupervised domain adaptation (SFUDA) presents a more feasible solution. In this paper, we propose a novel SFUDA framework that leverages a pretrained generative foundation model to augment target domain data without requiring access to source domain information. Additionally, we integrate an uncertainty-aware pseudo-labeling strategy along with consistency regularization to further enhance the adaptation process. Experimental results validate that our approach significantly outperforms conventional techniques, providing an effective and robust solution for IoT physical-layer authentication under realistic constraints.

Abstract— Radio Frequency Fingerprinting Identification (RFFI) serves as a pivotal technology in the Industrial Internet of Things (IIoT), witnessing significant strides over the last decade primarily due to advances in deep learning. However, most existing studies assume that training and test data are independent and identically distributed (i.i.d.), an assumption that often breaks down in real-world IIoT applications, leading to substantial performance degradation in cross-domain scenarios. To address this, we propose FATransformer, a novel unsupervised domain adaptation technique. The method is built on a robust theoretical foundation, ensuring stability and reliability across different domains. Specifically, FATransformer employs a Transformer-based architecture to efficiently process and align intermediate feature maps from both domains. Incorporating an attention-based module, it dynamically adjusts the weights of alignment at various layers, thereby improving the model’s flexibility to handle diverse real-world data. Extensive evaluations across multiple datasets underscore FATransformer’s superiority over existing methods.

Radio-frequency (RF) signal processing has seen significant advancements with the advent of deep learning, providing more accurate and efficient solutions for tasks such as signal classification and generation. However, most existing methods are heavily dependent on large labeled datasets, which are often scarce and costly to obtain in real-world RF environments. Furthermore, these approaches tend to be task-specific, limiting their ability to generalize across various RF applications. To address these challenges, this paper proposes RF-MAE, a self-supervised adaptive frequency masked autoencoder. RF-MAE leverages self-supervised learning (SSL) to capture intrinsic patterns from large-scale unlabeled RF data. Central to RF-MAE is a novel Adaptive Frequency Masked (AFM) strategy, which dynamically masks frequency components based on their energy distribution. Supported by a robust theoretical foundation, AFM ensures the model focuses on the most informative signal components, thereby enhancing generalization across RF tasks. By pretraining on unlabeled data and fine-tuning on specific tasks, RF-MAE significantly reduces the reliance on labeled datasets while improving adaptability across diverse RF signal processing tasks. Experimental results demonstrate that RF-MAE consistently outperforms traditional models, underscoring its potential to generalize across tasks and deliver superior performance in a wide range of RF signal applications.

Source-free unsupervised domain adaptation (SFUDA) aims to improve performance in unlabeled target domain data without accessing source domain data. This is crucial in scenarios with data-sharing restrictions due to privacy or compliance constraints. Existing SFUDA approaches often rely on pseudo-labeling techniques based on entropy or confidence metrics. These often overlook fine-grained data features, resulting in noisy pseudo-labels that degrade model performance. To overcome this limitation, we develop a new method called fine-grained pseudo-labeling and feature alignment (FGPLFA) to enhance SFUDA's performance. FGPLFA starts with a gradient-based metric that integrates insights from both model knowledge and data features, creating a more reliable sample metric. To enhance fine granularity, the fine-grained pseudo-labeling (FGPL) module was introduced. This module clusters data based on the magnitude and direction of gradients, allowing for dataset partitioning into subsets at the sample level. The subsets are pseudo-labeled with category-specificity and domain specificity, establishing a multilevel granularity structure that reduces noisy pseudo-labels. Subsequently, the mean-covariance adjustment feature alignment (MCAFA) method was introduced. Features from the subsets are aligned in a specified sequence, enhancing model adaptability in the target domain. Extensive experiments conducted across multiple datasets validate the superiority of FGPLFA.

Specific Emitter Identification (SEI) is crucial in the Internet of Things (IoT) to ensure the authentication and security of devices. With advancements in deep learning (DL), SEI for IoT devices has achieved remarkable progress. However, traditional DL-based SEI relies on a blanket assumption that emitter signals are transmitted in a constant channel environment and collected by a fixed receiver before identification. This assumption overlooks the dynamic characteristics of real-world IoT scenarios, where the channel environment and receiver are subject to change. Such variations can significantly impact SEI systems, potentially leading to a substantial decrease in identification accuracy. This challenge is known as the cross-domain SEI problem, where different receivers and channel environments are viewed as distinct domains. To mitigate this issue, we integrate unsupervised domain adaptation (UDA) into SEI. We propose an innovative UDA framework named domain-specific classifier network (DSCN) for cross-domain SEI. In our method, we initially use a weight-shared extractor for feature extraction. Unlike most existing UDA methods, we do not enforce the extractor to generate domain-invariant features for cross-domain identification. Instead, we design domain-specific classifiers to process features from different domains: source signal features are recognized by a source-specific classifier, while target signal features are recognized by a target-specific classifier. Experimental results demonstrate that the DSCN framework effectively mitigates identification accuracy degradation in cross-domain scenarios and outperforms existing UDA methods.

Radio Frequency Fingerprinting Identification (RFFI) has become a critical technology in the physical-layer security (PLS) field, with deep learning emerging as the dominant approach over the past decade. However, most deep learning-based models rely on the assumption that training and testing data follow an independent and identical distribution (i.i.d.), which often does not hold in real-world scenarios. This mismatch significantly degrades model performance in cross-domain settings, making cross-domain RFFI a challenging task. Traditional unsupervised domain adaptation (UDA) methods attempt to address this issue by jointly optimizing task loss and domain loss which is able to reduce the distribution gap between training and testing data. However, we observe that during training, the gradients of these two losses often conflict, hindering effective optimization and limiting cross-domain performance improvements. To address these challenges, we propose a novel framework, Gradient Collaborative Optimization with Dynamic Weighted Feature Alignment (GCODWFA). Specifically, GCODWFA introduces a novel Gradient Collaborative Optimization (GCO) loss, which explicitly adjusts the gradient interaction between task and domain losses by optimizing their angular relationship. Additionally, it incorporates a Dynamic Weighted Feature Alignment (DWFA) strategy, which dynamically adjusts the layer-specific weights for feature alignment based on the angular similarity of task and domain gradients. Extensive experiments conducted on multiple datasets demonstrate the superiority of GCODWFA over existing methods.

Driven by the pursuit of enhanced performance, deep learning has recently seen rapid developments in the scaling of network architectures and parameters. However, this advancement has led to extremely high computational costs, undesirable in real-time and resource-limited scenarios. To address these challenges, we propose SwiftNet, a cost-efficient deep learning framework. Our novelty lies in SwiftNet's innovative multidimensional early-exit strategy that integrates seamlessly with existing neural network architectures. The framework includes additional branch classifiers concatenated to the backbone network, allowing high-confidence samples to exit early, thereby, reducing computational load. Unlike traditional methods, SwiftNet dynamically assesses confidence levels, ensuring only low-confidence samples proceed to subsequent classifiers or the final layer, optimizing resource usage without compromising accuracy. We have validated SwiftNet on multiple neural network models and datasets, demonstrating its ability to significantly reduce the computational cost of models while maintaining neural network performance.

While the Internet of Things (IoT) technology is booming and offers huge opportunities for information exchange, it also faces unprecedented security challenges. As an important complement to the physical-layer security technologies for IoT, radio frequency fingerprinting (RFF) is of great interest due to its difficulty in counterfeiting. Recently, many machine learning (ML)-based RFF algorithms have emerged. In particular, deep learning (DL) has shown great benefits in automatically extracting complex and subtle features from raw data with high-classification accuracy. However, DL algorithms face the computational cost problem as the difficulty of the RFF task and the size of the deep neural network have increased dramatically. To address the above challenge, this article proposes a novel cost-effective early-exit neural network consisting of a complex-valued neural network (CVNN) backbone with multiple random forest branches, called hybrid CVNN-RF. Unlike conventional studies that use a single fixed DL model to process all radio frequency (RF) samples, our hybrid CVNN-RF considers differences in the recognition difficulty of RF samples and introduces an early-exit mechanism to dynamically process the samples. When processing “easy” samples that can be well classified with high confidence, the hybrid CVNN-RF can end early at the random forest branch to reduce computational cost. Conversely, subsequent network layers will be activated to ensure accuracy. To further improve the early-exit rate, an automated multidimensional early-exit strategy is proposed to achieve scheduling control from multiple dimensions within the network depth and classification category. Finally, our experiments on the public ADS-B data set show that the proposed algorithm can reduce the computational cost by 83% while improving the accuracy by 1.6% under a classification task with 100 categories.

A prevailing assumption in existing deep learning research posits that data across source and target domains adhere to the independent and identically distributed (i.i.d.) assumption. However, this assumption often proves inadequate in real-world scenarios, leading to significant performance degradation when models encounter data with divergent distributions. To address this challenge, a novel unsupervised domain adaptation (UDA) algorithm, decoupling feature alignment (DFA), is introduced. The approach begins with the establishment of a robust theoretical framework, serving as the foundation for the mean-covariance adjustment feature alignment (MCAFA) algorithm. Simultaneously, a data decoupling (DD) module is introduced, effectively segregating target domain data into two subsets: one that mirrors the source domain and another that diverges markedly. Furthermore, a multidimensional alignment module is employed, leveraging the MCAFA algorithm and the DD module to align target data with source data across various layers and categories. Comprehensive evaluations on multiple data sets underscore the superiority of DFA.

With the development of wireless communication technology and the increasingly complex electromagnetic environment, radio frequency fingerprinting (RFF) plays a vital role in improving the security of communication and information systems. Recently, many RFF algorithms based on machine learning have emerged. However, most of them focus on improving the accuracy of RFF identification but ignore the computational cost. This paper proposes a novel classifier composed of a convolutional neural network (CNN) backbone with two random forest branches, called hybrid CNN-RF. Under the scheduling of the proposed multi-dimensional early exit strategy, hybrid CNN-RF can end early when processing “easy” samples to reduce the computational cost, and activate the inference of the subsequent network layer when processing “hard” samples to ensure accuracy. Finally, our experiments show that the proposed algorithm can reduce the computational cost by a factor of 3.41 while improving the accuracy by 1.5%.

Recently, various wireless Internet-of-Things devices and unmanned aerial vehicles have been frequently used to facilitate daily life. On the other hand, they can also pose serious threats to public security if maliciously used. A countermeasure against these threats is to transmit jamming signals to break the communication links between attacker and wireless devices. However, this is not easy since modern multi-device (multi-user) wireless communication systems are smart in avoiding jamming. Moreover, a relay is widely used to improve the jamming resistance. To successfully jam those malicious devices, we consider a jamming and anti-jamming zero-sum game, in which the attacker tries to maximize the sum-rate of a relay-assisted wireless system and the jammer tries to minimize it. Finding the equilibrium is challenging due to its non-convexity and complex structure. We address this problem in two cases. Specifically, in the single-device (single-user) case, we show that putting the whole jamming power to either relay or device achieves the Nash equilibrium. In the multi-device case, an efficient method is developed, which transforms this game into a min-max optimization problem and decouples it into two sub-problems by the hybrid block successive approximation method. The resultant sub-problems are solved by the multi-block alternating direction method of multipliers and the block successive upper-bound minimization method of multipliers, respectively. Then, a stationary point of the original problem can be iteratively achieved. Numerical results demonstrate that the two proposed jamming methods outperform many traditional methods.

An automatic inductor design process is helpful to reduce the design cycle of radio frequency (RF) integrated circuit (IC). This paper proposed an efficient synthesis-analysis machine (SAM) for on-chip inductor synthesis and modeling, as well as an automatic dataset generation (ADG) topology for the generation of artificial neural networks (ANNs) training dataset. The SAM consists of a synthesis ANN, two analysis ANNs and a proposed self-inspection machine (SIM). For a given design request, the SAM synthesizes a layout first, followed by analyzing the performance of the layout automatically, and finally improves the layout confidence through self-inspection. Compared to the modeling functions of analysis ANN, the synthesis ANN behaves as an inverse model, of which the inputs are desired inductor performances while the outputs are geometrical parameters of the layout. However, multi-value problems might occur in obtaining geometries of inductors, suggesting with evidence from mathematic relationships between geometrical and electrical parameters of an inductor. A multi-valued training dataset will mislead the synthesis ANN, resulting in unqualified layouts. To deal with this issue, a solution space contraction technique (SSCM) is also proposed. Furthermore, the SIM suspends most of the failures by comparing the output of analysis ANN back to the input of synthesis ANN. An electromagnetic simulation tool is used for experiments, and the effectiveness of the proposed SAM is proven by 6,500 inductor samples.

Optimal jamming has important applications in both military and civil communications. There have been a brunch of works investigating the optimal jamming signal design when the signal modes of the opponent are known. In this work, we focus on the less studied hidden mode jamming problem. That is, the jammer has partially recorded the signal modes of the opponent, but there are some hidden modes not revealed to the jammer as of the appearance of these modes. As such, when the hidden modes appear, the jammer has to quickly adapt its jamming strategy to achieve effective jamming. However, it is challenging to do so due to incomplete knowledge of the intrinsic relation between the known and the hidden modes. In this work, a learning-based approach is proposed to attack this problem. Specifically, we custom-devise a jamming network (J-Net) to automatically learn the intrinsic relation among different modes and transfer the jamming strategy from the known modes to the hidden ones. Experimental results demonstrate that the J-Net attains much better jamming effect than pulsed Gaussian jamming and random jamming, and is comparable to the reinforcement learning-based approach, which assumes all the (known and hidden) modes available at the jammer.

Transmit beamforming and artificial noise-based methods have been widely employed to achieve physical-layer (PHY) security. However, these approaches may fail to provide satisfactory secure performance if the channels of legitimate user (LU) and eavesdropper (Eve) are highly correlated, which usually occurs in the case of close-located LU and Eve. The goal of this paper is to address the PHY security problem for proximal LU and Eve in millimeter-wave transmissions. To this end, we propose a novel frequency diverse array (FDA) beamforming approach, which intentionally introduces some frequency offsets across array antennas to decouple the highly correlated channels of LU and Eve. By exploiting this decoupling capability, the FDA beamforming can degrade Eve's reception and thus enhance PHY security. Leveraging FDA beamforming, we aim to maximize the secrecy rate by jointly optimizing the frequency offsets and the transmit beamformer. This secrecy rate maximization problem is difficult due to the tightly coupled variables. However, we show that it can be reformulated into a form only depending on the frequency offsets. Building upon this reformulation, we further employ the block successive upper-bound minimization method to iteratively obtain a solution with stationary convergence guarantee. Numerical results demonstrate that FDA beamforming can provide higher secrecy rate than conventional beamforming, especially for proximal LU and Eve.

Traditional interference suppression approaches will enjoy the additional benefits when angle as well as polarization domain information are involved by polarization sensitive array (PSA). However, the information from range domain and its collaboration with other domains are rarely explored. In this paper, we present the polarization sensitive frequency diverse array (PSFDA) which combines frequency diverse array (FDA) and PSA for robust angle-range-polarization beamforming. To improve the angle-range-polarization resolution, the sparse reconstruction respective of compressive sensing is further applied. Theoretical analysis and simulation results demonstrate that our proposed algorithm can provide the good range-polarization resolution not just the target direction.

Large time-bandwidth product waveform diversity design is a challenging topic in multiple-input multiple-output radar high-resolution imaging because existing methods usually can generate only two large time-bandwidth product waveforms. This paper proposes a new low peak-to-average ratio (PAR) orthogonal frequency division multiplexing chirp waveform diversity design through randomly subchirp modulation. This method can easily yield over two orthogonal large time-bandwidth product waveforms. More waveforms means that more degrees-of-freedom can be obtained for the system. The waveform performance is evaluated by the ambiguity function. It is shown that the designed waveform has the superiorities of a large time-bandwidth product which means high range resolution and low transmit power are allowed for the system, almost constant time-domain and frequency-domain modulus, low PAR and no range-Doppler coupling response in tracking moving targets.

The split‐symbol moment estimation (SSME) algorithm is a conventional method to estimate the signal‐to‐noise ratio (SNR) in communication systems for M ‐ary phase shift keying signals. However, the conventional SSME is based on the assumption of symmetric pulse‐shaping waveforms, and becomes no longer proper when the shaping pulse is asymmetric. In this study, the authors propose a modified SSME algorithm for asymmetric pulse‐shaped signals, based on an odd‐even symbol splitting approach. Simulation results show that the modified SSME outperforms the conventional SSME for both symmetric and asymmetric pulse‐shaped signals in terms of SNR estimation accuracy.

Along-track interferometry (ATI) and displaced phase center antenna (DPCA) are two representative synthetic aperture radar (SAR) ground moving target indication (GMTI) techniques. However, the former is a clutter-limited detector, whereas the latter is a noise-limited detector. In this letter, we propose a two-antenna SAR with waveform diversity for GMTI. The two antennas placed in the azimuth dimension simultaneously transmit two orthogonal waveforms, namely, the up- and down-chirp waveforms. In doing so, four independent transmit-receive channels are obtained for the receiver, and thus, an efficient GMTI is developed by cooperatively utilizing the ATI GMTI and DPCA GMTI processing techniques. This method first uses the DPCA technique to cancel clutter and then the ATI technique to suppress noise. Next, the fractional Fourier transform is employed to estimate the Doppler parameters and corresponding reference filters are designed to focus the moving targets. The proposed approach cooperatively utilizes the advantages of ATI and DPCA GMTI techniques and overcomes their disadvantages. It does not prerequire knowledge of the target's across-track velocity. The effectiveness is verified by simulation results.

Signal-to-noise ratio (SNR) is a crucial parameter for adaptive transmission in orthogonal frequency division multiplexing (OFDM) systems. For the OFDM-based multicast/broadcast over single frequency network (MBSFN) system, the downlink transmission channels is quite frequency selective, which is a great challenge for SNR estimation in user terminals with simple implementation. We develop a pilotassisted algorithm operated in transform domain to improve estimation performance. The proposed estimation algorithm is robust to frequency and time selectivity, and is suited to different pilot patterns. Simulations show that the proposed estimator is insensitive to different type of channels, and is superior to conventional methods.