Radio Frequency Fingerprint (RFF) is a unique characteristic of radio signals impacted by hardware imperfections of the device’s radio front-end. This characteristic can be used as a security measure to identify individual devices. There are various methods to extract RFF features, ranging from statistical analysis to deep learning methods, which have seen increasing adoption among researchers, given their automatic feature extraction capabilities and performance. However, the black-box nature of deep learning models is a significant drawback, as it raises concerns about the model’s reliability, especially in critical applications such as security. In this paper, we use In-phase and Quadrature (I/Q) samples to create spectrograms and train an EfficientNet model for device classification. The model is trained on a Wi-Fi dataset of 20 devices, achieving 99% accuracy on the test set. More importantly, to understand the model’s decision-making process, we apply the Gradient-weighted Class Activation Map (Grad-CAM) to visualize its attention. Grad-CAM highlights specific spectrogram regions that are linked to the device’s fingerprint signature, including power leakage, Direct Current (DC) components, and energy patterns.

Radio Frequency Fingerprint Identification (RF-FID) is a novel approach that aims to differentiate devices based on their unique signal transmissions, rather than their given identities. This approach has promising applications in wireless security, spectrum management, and sensing. Current RFFID research uses deep learning due to its success in various domains, which is largely attributed to the availability of massive labeled datasets for training. However, unlike other domains, labeled data in RFFID is limited, and the labeling process is expensive. In this paper, we propose a label-efficient learning approach for RFFID based on Contrastive Predictive Coding (CPC), a pre-training method that learns to predict future samples given the past without labels. Afterward, the model is fine-tuned to identify the device. We evaluate our approach on a fingerprint dataset of 20 devices. Our results show that CPC learns effective representations of RF signals and outperforms fully supervised learning in both classification performance and label efficiency, requiring up to 10 times fewer labels while maintaining competitive accuracy. Finally, we evaluate CPC's robustness against noise and observe competitive performance after fine-tuning.

Recent advances in Vehicle-to-Everything (V2X) communication have significantly improved road safety. However, the requirements of future vehicular applications are rapidly surpassing the capabilities of current generation cellular networks. Future wireless networks are expected to use millimeter wave (mmWave) frequencies due to spectrum availability and promises of very high throughput and ultra-low latency. These frequencies suffer from high path loss and are sensitive to mobility, which poses new challenges for V2X. In this paper, we investigate vehicular communication at mmWave frequencies through extensive simulation. We design, implement, and simulate various vehicular scenarios to investigate the impact of mmWave frequencies and mobility on system performance, with a specific focus on teleoperated driving applications due to their high uplink throughput requirements. Our simulation results demonstrate the potential of mmWave networks for vehicular communication, while highlighting the need to address challenges such as limited uplink capacity and poor reliability.

The combination of 5G and Multi-access Edge Computing (MEC) technologies can bring significant benefits to vehicular networks, providing means for achieving enhanced Quality of Service (QoS), and Quality of Experience (QoE) of wide variety of vehicular applications. Although beneficial in terms of latency reduction, the edge of the architecture for communication networks produces enormous heterogeneity of network services and resources. This challenge becomes even more severe when different administration domains are taken into consideration. Thus, efficient network Management and Orchestration (MANO) of network resources and services are inevitable. As ETSI provided guidelines and standardization for NFV MANO components, the MEC platform can be used to host network services, while MANO systems are in charge of network service management and orchestration. In this paper, we focus on the specific impact that the Virtualized Infrastructure Manager (VIM) has on the performance of the whole MANO system, used for management and orchestration of MEC services and resources in vehicular networks by enabling the on-demand service instantiation, and service teardown. In our testbed-based evaluation, we measured the network service instantiation and termination delays when evaluating: a) OpenStack and Amazon Web Services (AWS) as VIMs for Open Source MANO (OSM), and b) OpenStack and Docker in case of Open Baton. Such performance analysis with a strong experimental component can serve as a baseline for researchers and industry towards exploiting the opportunities that existing MANO solutions provide.

Crosstalk cancellation, also known as vectoring, is a promising technique that significantly increases the capacity of digital subscriber line access networks. In this paper, the potential of vectoring to boost the capacity in a realistic copper access network is demonstrated. Deployment of vectoring becomes challenging when not all crosstalk can be eliminated due to the presence of uncontrolled lines or due to limited cancellation resources. Lines can be outside of the vectoring group because of technological constraints or regulation. As a result, the performance of vectoring degrades as compared to full cancellation. Based on representative simulations and lab measurements, different deployment guidelines are proposed along with a discussion on their practicality.

To gradually expand their networks, operators deploy new xDSL technologies from remote terminals (RT's) closer to the customer. However, since xDSL lines deployed from an RT can share a binder with lines deployed from the central office (CO), a near-far problem is introduced and crosstalk may cause a severe performance degradation for the CO-deployed lines. RT lines have to be careful about how to allocate transmit power to the transmission frequencies, such that the impact on the CO lines is acceptable. Noise impact (NI) methods have been proposed based on a target noise shape generated by the disturbers (RT lines). In this paper new methods based on the actual impact on the bit rate of the victims (CO lines) are proposed. These bit impact (BI) methods closely approximate the optimal power allocation.

In this paper we investigate the tracking performance of the downstream (DS) crosstalk (XT) channel estimation based on the reported signal to noise ratio (SNR), in particular for digital subscriber line (DSL) systems. Aiming to its simplicity, the SNR-assisted XT estimation, has been recognized in ITU as a backward compatible method that does not require any change in the very high speed digital subscriber line 2 (VDSL2) standard. This low complex algorithm can be used for XT channel estimation in dynamic spectrum management (DSM) techniques today. The algorithm as proposed, relies on sending perturbing signals on the victim lines (VLs) and reporting the SNRs by those lines to acquire the crosstalk channel from some disturber line (DL) to the VLs. We generalize this concept to include full startup, tracking and joining scenarios as well as the impact of different perturbation signal choices. Simulation results reveal that starting from no crosstalk precompensation, and updating a precoder matrix based on the DS crosstalk channel estimates, the far-end crosstalk (FEXT) free SNR can be reached in few iterations (36 SNR measurements for the four lines case).

In this paper, we present a new method to estimate the crosstalk channels in VDSL systems, allowing to implement well-known pre-compensation schemes. As opposed to previously presented methods, this estimation only requires minimal changes to the current standard and equipments. It is based on the concept of adding small, controlled, perturbations to the transmitted signal and observing the related changes to the SNR at the receiver. It is shown in this paper that, based on a limited amount of SNR measurements from the receivers, it is possible to accurately estimate the crosstalk channels, both in amplitude and phase.

A statistical model has been constructed that extends the phenomenological 99% cumulant model to a full cumulative distribution function for amplitude and phase of the far end crosstalk between twisted pair telephone wires. The offset of the crosstalk amplitude relative to the 99% level is described by a frequency independent beta distribution. Inter-binder crosstalk is modeled by an additional parameter that reflects the distance between the two binders. The multi-line behavior of the beta model is shown to be close to the currently agreed scaling factor without additional forcing.

Rapid acquisition of accurate crosstalk estimates is a core requirement for effective preceding in digital subscriber line (DSL) systems. It is shown that signal-to-noise ratio (SNR) reports provided by customer premises equipment (CPE) can be used to perform this task by "tuning" the precoder, i.e., iterating alternate steps of estimation and precoder adaptation. Such an approach has the advantage that it can be applied to legacy CPEs. Estimation algorithms are designed using techniques from stochastic control.