Evaluating cellular systems, from 5th generation (5G) New Radio (NR) and 5G-Advanced to 6th generation (6G), is challenging because the performance emerges from the tight coupling of propagation, beam management, scheduling, and higher-layer interactions. System-level simulation is therefore indispensable, yet the vast majority of studies rely on the statistical 3rd Generation Partnership Project (3GPP) channel models. These are well suited to capture average behavior across many statistical realizations, but cannot reproduce site-specific phenomena such as comer diffraction, street-canyon blockage, or deterministic line-of-sight conditions and angle-of- departure/arrival relationships that drive directional links.This paper extends 5G-LENA, an NR module for the system-level Network Simulator 3 (ns-3), with a trace-based channel model that processes the Multipath Components (MPCs) obtained from external ray-tracers (e.g., Sionna Ray Tracer (RT)) or measurement campaigns. Our module constructs frequency-domain channel matrices, and feeds them to the existing Physical (PHY)/Medium Access Control (MAC) stack without any further modifications. The result is a geometry-based channel model that remains fully compatible with the standard 3GPP implementation in 5G-LENA, while delivering site-specific geometric fidelity. This new module provides a key building block toward Digital Twin (DT) capabilities by offering realistic site-specific channel modeling, unlocking studies that require site awareness, including beam management, blockage mitigation, and environment-aware sensing. We demonstrate its capabilities for precise beam-steering validation and end-to-end metric analysis. In both cases, the trace-driven engine exposes performance inflections that the statistical model does not exhibit, confirming its value for high- fidelity system-level cellular networks research and as a step toward DT applications.

Autonomous driving is a major paradigm shift in transportation, with the potential to enhance safety, optimize traffic congestion, and reduce fuel consumption. Although autonomous vehicles rely on advanced sensors and on-board computing systems to navigate without human control, full awareness of the driving environment also requires a cooperative effort via Vehicle-to-Everything (V2X) communication. Specifically, vehicles send and receive sensor observations to/from other vehicles to extend perception beyond their own sensing range. However, transmitting large volumes of data can be challenging for current V2X communication technologies, so data compression represents a crucial solution to reduce the message size and link congestion. In this paper, we present a statistical characterization of automotive data, focusing on Light Detection and Ranging (LiDAR) sensors. Notably, we provide models for the size of both raw and compressed point clouds. The use of statistical traffic models offers several advantages compared to using real data, such as faster simulations, reduced storage requirements, and greater flexibility in the application design. Furthermore, statistical models can be used for understanding traffic patterns and analyzing statistics, which is crucial to design and optimize wireless networks. We validate our statistical models via a Kolmogorov-Smirnoff (KS) test implementing a Bootstrap Resampling scheme. Moreover, we show via ns-3 simulations that using statistical models yields results in terms of latency and throughput that are comparable to real data, which also demonstrates the accuracy of the models.

The rapid development of Virtual Reality (VR) technology has led the industry and research community to look at its major challenges with increased interest. The main challenge in ensuring a high Quality of Experience (QoE) for users is represented by cybersickness, a phenomenon similar to motion sickness experienced by many VR users, while at the same time, the high data rates needed by VR require the definition of traffic models for network optimization. These two problems are intertwined, but have never been studied jointly before due to the lack of suitable datasets. In this paper, we present Questset, the first dataset designed for this purpose. Questset contains over 40 hours of VR traces from 70 users playing commercially available video games, and includes both traffic data for network optimization, and movement and user experience data for cybersickness analysis. Therefore, Questset represents an enabler to jointly address the main VR challenges in the near future.

Digital twins are now a staple of wireless networks design and evolution. Creating an accurate digital copy of a real system offers numerous opportunities to study and analyze its performance and issues. It also allows designing and testing new solutions in a risk-free environment, and applying them back to the real system after validation. A candidate technology that will heavily rely on digital twins for design and deployment is 6G, which promises robust and ubiquitous networks for eXtended Reality (XR) and immersive communications solutions. In this paper, we present BostonTwin, a dataset that merges a high-fidelity 3D model of the city of Boston, MA, with the existing geospatial data on cellular base stations deployments, in a ray-tracing-ready format. Thus, BostonTwin enables not only the instantaneous rendering and programmatic access to the building models, but it also allows for an accurate representation of the electromagnetic propagation environment in the real-world city of Boston. The level of detail and accuracy of this characterization is crucial to designing 6G networks that can support the strict requirements of sensitive and high-bandwidth applications, such as XR and immersive communication.

Over the past few years, the concept of Virtual Reality (VR) has attracted increasing interest thanks to its extensive industrial and commercial applications. Currently, the 3D models of the virtual scenes are generally stored in the VR visor itself, which operates as a standalone device. However, applications that entail multi-party interactions will likely require the scene to be processed by an external server and then streamed to the visors. However, the stringent Quality of Service (QoS) constraints imposed by the VR's interactive nature require Network Slicing (NS) solutions, for which profiling the traffic generated by the VR application is crucial. To this end, we collected more than 4 hours of traces in a real setup and analyzed their temporal correlation, focusing on the CBR encoding mode, which should generate more predictable traffic streams. From the collected data, we then distilled two prediction models for future frame size, which can be instrumental in the design of dynamic resource allocation algorithms. Our results show that even the state-of-the-art H.264 CBR mode may have significant frame size fluctuations, impacting NS optimization. We then exploited the models to dynamically determine requirements in an NS scenario, providing the required QoS while minimizing resource usage.

Future wireless networks and sensing systems will benefit from access to large chunks of spectrum above 100 GHz, to achieve terabit-per-second data rates in 6th Generation (6G) cellular systems and improve accuracy and reach of Earth exploration and sensing and radio astronomy applications. These are extremely sensitive to interference from artificial signals, thus the spectrum above 100 GHz features several bands which are protected from active transmissions under current spectrum regulations. To provide more agile access to the spectrum for both services, active and passive users will have to coexist without harming passive sensing operations. In this paper, we provide the first, fundamental analysis of Radio Frequency Interference (RFI) that large-scale terrestrial deployments introduce in different satellite sensing systems now orbiting the Earth. We develop a geometry-based analysis and extend it into a data-driven model which accounts for realistic propagation, building obstruction, ground reflection, for network topology with up to 105nodes in more than 85 km2. We show that the presence of harmful RFI depends on several factors, including network load, density and topology, satellite orientation, and building density. The results and methodology provide the foundation for the development of coexistence solutions and spectrum policy towards 6G.

Accurate scene understanding from multiple sensors mounted on cars is a key requirement for autonomous driving systems. Nowadays, this task is mainly performed through data-hungry deep learning techniques that need very large amounts of data to be trained. Due to the high cost of performing segmentation labeling, many synthetic datasets have been proposed. However, most of them miss the multi-sensor nature of the data, and do not capture the significant changes introduced by the variation of daytime and weather conditions. To fill these gaps, we introduce SELMA, a novel synthetic dataset for semantic segmentation that contains more than 30K unique waypoints acquired from 24 different sensors including RGB, depth, semantic cameras and LiDARs, in 27 different weather and daytime conditions, for a total of more than 20M samples. SELMA is based on CARLA, an open-source simulator for generating synthetic data in autonomous driving scenarios, that we modified to increase the variability and the diversity in the scenes and class sets, and to align it with other benchmark datasets. As shown by the experimental evaluation, SELMA allows the efficient training of standard and multi-modal deep learning architectures, and achieves remarkable results on real-world data. SELMA is free and publicly available, thus supporting open science and research.

The worldwide commercialization of fifth generation (5G) wireless networks and the exciting possibilities offered by connected and autonomous vehicles (CAVs) are pushing toward the deployment of heterogeneous sensors for tracking dynamic objects in the automotive environment. Among them, Light Detection and Ranging (LiDAR) sensors are witnessing a surge in popularity as their application to vehicular networks seem particularly promising. LiDARs can indeed produce a three-dimensional (3D) mapping of the surrounding environment, which can be used for object detection, recognition, and topography. These data are encoded as a point cloud which, when transmitted, may pose significant challenges to the communication systems as it can easily congest the wireless channel. Along these lines, this paper investigates how to compress point clouds in a fast and efficient way. Both 2D- and a 3D-oriented approaches are considered, and the performance of the corresponding techniques is analyzed in terms of (de)compression time, efficiency, and quality of the decompressed frame compared to the original. We demonstrate that, thanks to the matrix form in which LiDAR frames are saved, compression methods that are typically applied for 2D images give equivalent results, if not better, than those specifically designed for 3D point clouds.

In a fully autonomous driving framework, where vehicles operate without human intervention, information sharing plays a fundamental role. In this context, new network solutions have to be designed to handle the large volumes of data generated by the rich sensor suite of the cars in a reliable and efficient way. Among all the possible sensors, Light Detection and Ranging (LiDAR) can produce an accurate 3D point cloud representation of the surrounding environment, which in turn generates high data rates. For this reason, efficient point cloud compression is paramount to alleviate the burden of data transmission over bandwidth-constrained channels and to facilitate real-time communications. In this paper, we propose a pipeline to efficiently compress LiDAR observations in an automotive scenario. First, we leverage the capabilities of RangeNet++, a Deep Neural Network (DNN) used to semantically infer point labels, to reduce the channel load by selecting the most valuable environmental data to be disseminated. Second, we compress the selected points using Draco, a 3D compression algorithm which is able to obtain compression up to the quantization error. Our experiments, validated on the Semantic KITTI dataset, demonstrate that it is possible to compress and send the information at the frame rate of the LiDAR, thus achieving real-time performance.

The millimeter wave (mmWave) band will provide multi-gigabits-per-second connectivity in the radio access of future wireless systems. The high propagation loss in this portion of the spectrum calls for the deployment of large antenna arrays to compensate for the loss through high directional gain, thus introducing the need for a spatial dimension in the channel model to accurately represent the performance of a mmWave network. In this perspective, ray tracing can characterize the channel in terms of Multi Path Components (MPCs) to provide a highly accurate model, at the price of extreme computational complexity (e.g., for processing detailed environment information about the propagation), which may limit the scalability of the simulations. In this paper, we present possible simplifications to improve the trade-off between accuracy and complexity in ray-tracing simulations at mmWaves by reducing the total number of MPCs. The effect of such simplifications is evaluated from a full-stack perspective through end-to-end simulations, testing different configuration parameters, propagation scenarios, and higher-layer protocol implementations. We then provide guidelines on the optimal degree of simplification, for which it is possible to reduce the complexity of simulations with a minimal reduction in accuracy for different deployment scenarios.

Complex phenomena are generally modeled with sophisticated simulators that, depending on their accuracy, can be very demanding in terms of computational resources and simulation time. Their time-consuming nature, together with a typically vast parameter space to be explored, make simulation- based optimization often infeasible. In this work, we present a method that enables the optimization of complex systems through Machine Learning (ML) techniques. We show how well-known learning algorithms are able to reliably emulate a complex simulator with a modest dataset obtained from it. The trained emulator is then able to yield values close to the simulated ones in virtually no time. Therefore, it is possible to perform a global numerical optimization over the vast multi-dimensional parameter space, in a fraction of the time that would be required by a simple brute-force search. As a testbed for the proposed methodology, we used a network simulator for next-generation mmWave cellular systems. After simulating several antenna configurations and collecting the resulting network-level statistics, we feed it into our framework. Results show that, even with few data points, extrapolating a continuous model makes it possible to estimate the global optimum configuration almost instantaneously. The very same tool can then be used to achieve any further optimization goal on the same input parameters in negligible time.