Jamming devices disrupt signals from the global navigation satellite system (GNSS) and pose a significant threat by compromising the reliability of accurate positioning. Consequently, the detection and localization of these interference signals are essential to achieve situational awareness, mitigating their impact, and implementing effective counter-measures. Classical Angle of Arrival (AoA) methods exhibit reduced accuracy in multipath environments due to signal reflections and scattering, leading to localization errors. Additionally, AoA-based techniques demand substantial computational resources for array signal processing. In this paper, we propose a novel approach for detecting and classifying interference while estimating the distance, azimuth, and elevation of jamming sources. Our benchmark study evaluates 128 vision encoder and time-series models to identify the highest-performing methods for each task. We introduce an attention-based fusion framework that integrates in-phase and quadrature (IQ) samples with Fast Fourier Transform (FFT)-computed spectrograms while incorporating 22 AoA features to enhance localization accuracy. Furthermore, we present a novel dataset of moving jamming devices recorded in an indoor environment with dynamic multipath conditions and demonstrate superior performance compared to state-of-the-art methods.

One significant challenge in research is to collect a large amount of data and learn the underlying relationship between the input and the output variables. This paper outlines the process of collecting and validating a dataset designed to determine the angle of arrival (AoA) using Bluetooth low energy (BLE) technology. The data, collected in a laboratory setting, is intended to approximate real-world industrial scenarios. This paper discusses the data collection process, the structure of the dataset, and the methodology adopted for automating sample labeling for supervised learning. The collected samples and the process of generating ground truth (GT) labels were validated using the Texas Instruments (TI) phase difference of arrival (PDoA) implementation on the data, yielding a mean absolute error (MAE) at one of the heights without obstacles of 25.71 Indoor Positioning (IP), has the potential to enable several different technologies that could massively improve patient management in care homes, asset management and general automation in warehouses, automated inspection and maintenance, etc [1], [2]. IP assists in determining the device location by measuring location-dependent phenomena. These measurements could include the received signal strength (RSS) of wireless signals at the device's location, the angle of arrival (AoA) of wireless signals between the device and a base station, or the time of arrival (ToA) or time difference of arrival (TDoA) of wireless signals [3]. A wide range of wireless technologies are employable for RSS, AoA, ToA, and TDoA measurements. These include, but are not limited to, Bluetooth low energy (BLE) [4], [5], ultra-wideband radio (UWB) [6]-[8], wireless fidelity (WiFi) [9], [10] and millimeter (mm) wave radio [11]-[13]. UWB radio provides extremely accurate ToA and TDoA measurements due to the narrow width of its signals in the time domain, and more recently work has started on applying UWB to the task of AoA determination [14]. However, when considering a large-scale IP system, it is essential to consider the deployment cost of the technology.

5G cellular networks are particularly vulnerable against narrowband jammers that target specific control sub-channels in the radio signal. One mitigation approach is to detect such jamming attacks with an online observation system, based on machine learning. We propose to detect jamming at the physical layer with a pre-trained machine learning model that performs binary classification. Based on data from an experimental 5G network, we study the performance of different classification models. A convolutional neural network will be compared to support vector machines and k-nearest neighbors, where the last two methods are combined with principal component analysis. The obtained results show substantial differences in terms of classification accuracy and computation time.

5G mobile networks are vulnerable to jamming attacks that may jeopardize valuable applications such as industry automation. In this paper, we propose to analyze radio signals with a dedicated device to detect jamming attacks. We pursue a learning approach, with the detector being a CNN implementing a GLRT. To this end, the CNN is trained as a two-class classifier using two datasets: one of real legitimate signals and another generated artificially so that the resulting classifier implements the GLRT. The artificial dataset is generated mimicking different types of jamming signals. We evaluate the performance of this detector using experimental data obtained from a private 5G network and several jamming signals, showing the technique's effectiveness in detecting the attacks.

The Radio frequency (RF) fingerprinting technique makes highly secure device authentication possible for future networks by exploiting hardware imperfections introduced during manufacturing. Although this technique has received considerable attention over the past few years, RF fingerprinting still faces great challenges of channel-variation-induced data distribution drifts between the training phase and the test phase. To address this fundamental challenge and support model training and testing at the edge, we propose a federated RF fingerprinting algorithm with a novel strategy called model transfer and adaptation (MTA). The proposed algorithm introduces dense connectivity among convolutional layers into RF fingerprinting to enhance learning accuracy and reduce model complexity. Besides, we implement the proposed algorithm in the context of federated learning, making our algorithm communication efficient and privacy-preserved. To further conquer the data mismatch challenge, we transfer the learned model from one channel condition and adapt it to other channel conditions with only a limited amount of information, leading to highly accurate predictions under environmental drifts. Experimental results on real-world datasets demonstrate that the proposed algorithm is model-agnostic and also signal-irrelevant. Compared with state-of-the-art RF fingerprinting algorithms, our algorithm can improve prediction performance considerably with a performance gain of up to 15\%.

In this study, we propose a general-purpose synchronization method that allows a set of software-defined radios (SDRs) to transmit or receive any in-phase/quadrature data with precise timings while maintaining the baseband processing in the corresponding companion computers. The proposed method relies on the detection of a synchronization waveform in both receive and transmit directions and controlling the direct memory access blocks jointly with the processing system. By implementing this synchronization method on a set of low-cost SDRs, we demonstrate the performance of frequency-shift keying (FSK)-based majority vote (MV), i.e., an over-the-air computation scheme for federated edge learning, and introduce the corresponding procedures. Our experiment shows that the test accuracy can reach more than 95% for homogeneous and heterogeneous data distributions without using channel state information at the edge devices.

A novel cross-domain attentional multi-task architecture - xDom - for robust real-world wireless radio frequency (RF) fingerprinting is presented in this work. To the best of our knowledge, this is the first time such comprehensive attention mechanism is applied to solve RF fingerprinting problem. In this paper, we resort to real-world IoT WiFi and Bluetooth (BT) emissions (instead of synthetic waveform generation) in a rich multipath and unavoidable interference environment in an indoor experimental testbed. We show the impact of the time-frame of capture by including waveforms collected over a span of months and demonstrate the same time-frame and multiple time-frame fingerprinting evaluations. The effectiveness of resorting to a multi-task architecture is also experimentally proven by conducting single-task and multi-task model analyses. Finally, we demonstrate the significant gain in performance achieved with the proposed xDom architecture by benchmarking against a well-known state-of-the-art model for fingerprinting. Specifically, we report performance improvements by up to 59.3% and 4.91x under single-task WiFi and BT fingerprinting respectively, and up to 50.5% increase in fingerprinting accuracy under the multi-task setting.

Existing datasets used to train deep learning models for narrowband radio frequency (RF) signal classification lack enough diversity in signal types and channel impairments to sufficiently assess model performance in the real world. We introduce the Sig53 dataset consisting of 5 million synthetically-generated samples from 53 different signal classes and expertly chosen impairments. We also introduce TorchSig, a signals processing machine learning toolkit that can be used to generate this dataset. TorchSig incorporates data handling principles that are common to the vision domain, and it is meant to serve as an open-source foundation for future signals machine learning research. Initial experiments using the Sig53 dataset are conducted using state of the art (SoTA) convolutional neural networks (ConvNets) and Transformers. These experiments reveal Transformers outperform ConvNets without the need for additional regularization or a ConvNet teacher, which is contrary to results from the vision domain. Additional experiments demonstrate that TorchSig's domain-specific data augmentations facilitate model training, which ultimately benefits model performance. Finally, TorchSig supports on-the-fly synthetic data creation at training time, thus enabling massive scale training sessions with virtually unlimited datasets.

Physical-layer security is regaining traction in the research community, due to the performance boost introduced by deep learning classification algorithms. This is particularly true for sender authentication in wireless communications via radio fingerprinting. However, previous research efforts mainly focused on terrestrial wireless devices while, to the best of our knowledge, none of the previous work took into consideration satellite transmitters. The satellite scenario is generally challenging because, among others, satellite radio transducers feature non-standard electronics (usually aged and specifically designed for harsh conditions). Moreover, the fingerprinting task is specifically difficult for Low-Earth Orbit (LEO) satellites (like the ones we focus in this paper) since they orbit at about 800Km from the Earth, at a speed of around 25,000Km/h, thus making the receiver experiencing a down-link with unique attenuation and fading characteristics. In this paper, we propose PAST-AI, a methodology tailored to authenticate LEO satellites through fingerprinting of their IQ samples, using advanced AI solutions. Our methodology is tested on real data -- more than 100M I/Q samples -- collected from an extensive measurements campaign on the IRIDIUM LEO satellites constellation, lasting 589 hours. Results are striking: we prove that Convolutional Neural Networks (CNN) and autoencoders (if properly calibrated) can be successfully adopted to authenticate the satellite transducers, with an accuracy spanning between 0.8 and 1, depending on prior assumptions. The proposed methodology, the achieved results, and the provided insights, other than being interesting on their own, when associated to the dataset that we made publicly available, will also pave the way for future research in the area.