Predicting Networks Before They Happen: Experimentation on a Real-Time V2X Digital Twin

Roberto Pegurri, Habu Shintaro, Francesco Linsalata, Wang Kui, Tao Yu, Eugenio Moro, Maiya Igarashi, Antonio Capone, Kei Sakaguchi, Email: {name.surname}@polimi.it, {habu, kuiw, yutao, igarashi, sakaguchi}@mobile.ee.titech.ac.jp · 2026

Read Predicting Networks Before They Happen: Experimentation on a Real-Time V2X Digital Twin by Roberto Pegurri, Habu Shintaro, Francesco Linsalata, Wang Kui, Tao Yu, Eugenio Moro, Maiya Igarashi, Antonio Capone, Kei Sakaguchi, Email: {name.surname}@polimi.it, {habu, kuiw, yutao, igarashi, sakaguchi}@mobile.ee.titech.ac.jp (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-16559.txt.

Introduction

Roberto Pegurri and coauthors guess car net states ere they shift with a twin. Their 2026 preprint deals with car to all links, called V2X. Safe car apps need nets that act ere change hits. Past twins oft just ape the net as it now stands. A twin that can guess next states could warn cars in time.

The team tests guess twins on a live V2X twin set up. Tests check if next state guesses hold true in real runs. True guesses let nets shift load ere jams form. That shift is key for safe linked car fleets. The work moves twins from ape to guess tools.

Catalog record

CREATORRoberto Pegurri, Habu Shintaro, Francesco Linsalata, Wang Kui, Tao Yu, Eugenio Moro, Maiya Igarashi, Antonio Capone, Kei Sakaguchi, Email: {name.surname}@polimi.it, {habu, kuiw, yutao, igarashi, sakaguchi}@mobile.ee.titech.ac.jp
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE34 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-16559

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Predicting Networks Before They Happen: Experimentation on a Real-Time V2X Digital Twin Roberto Pegurri, Habu Shintaro, Francesco Linsalata, Wang Kui, Tao Yu, Eugenio Moro, Maiya Igarashi, Antonio Capone, Kei Sakaguchi, Email: {name.surname}@polimi.it, {habu, kuiw, yutao, igarashi, sakaguchi}@mobile.ee.titech.ac.jp Abstract Emerging safety-critical Vehicle-to-Everything (V2X) applications require networks to proactively adapt to rapid environmental changes rather than merely reacting to them. While Network Digital Twins (NDTs) offer a pathway to such predictive capabilities, existing solutions typically struggle to reconcile high-fidelity physical modeling with strict real-time constraints. This paper presents a novel, end-to-end real-time V2X Digital Twin framework that integrates live mobility tracking with deterministic channel simulation. By coupling the Tokyo Mobility Digital Twin—which provides live sensing and trajectory forecasting—with VaN3Twin—a full-stack simulator with ray tracing—we enable the prediction of network performance before physical events occur. We validate this approach through an experimental proof-of-concept deployed in Tokyo, Japan, featuring connected vehicles operating on 60 GHz links. Our results demonstrate the system’s ability to predict Received Signal Strength (RSSI) with a maximum average error of 1.01 dB and reliably forecast Line-of-Sight (LoS) transitions within a maximum average end-to-end system latency of 250 ms, depending on the ray tracing level of detail. Furthermore, we quantify the fundamental trade-offs between digital model fidelity, computational latency, and trajectory prediction horizons, proving that high-fidelity and predictive digital twins are feasible in real-world urban environments. I Introduction Connected and automated vehicles are pushing wireless networks to operate increasingly close to their physical limits, both in terms of reliability and reaction time. Safety critical Vehicle-to-Everything (V2X) applications such as collision avoidance, cooperative perception, and automated intersection management require the network to reason not only on the current state of the environment, but also on how positions and channels will evolve in the immediate future, from tens of milliseconds to a few seconds ahead [ 4 ] . This is precisely the promise of Network Digital Twins (NDTs) : a software replica of the physical system, continuously synchronized with live data, that can be queried to evaluate “what–if” situations before they affect real users [ 6 ] . Despite significant progress, most existing NDTs operate in an essentially offline fashion. High–fidelity ray tracing, realistic mobility engines, and multi-RAT protocol stacks are routinely used to obtain accurate performance evaluations, but without strict timing constraints, simulation runs may take seconds, minutes, or even hours to complete. This is acceptable for design-time analysis, but incompatible with a reactive or predictive Digital Twin (DT) , where results must be delivered within the tight latency budget [ 20 ] imposed by V2X applications, as an example. If the twin answers too late, vehicles will have moved, the channel will have changed, and any recommended action may already be obsolete or even unsafe. Two recent platforms partially address this gap. The Tokyo Mobility DT —from here on referred to as Mobility DT—developed by the Institute of Science Tokyo, Japan [ 18 ] provides a detailed 3D reconstruction of the Ookayama Campus area in Tokyo, Japan supporting real or replayed trajectories and enabling realistic urban-scale mobility scenarios. Conversely, VaN3Twin [ 9 ] extends the ms-van3t/ns-3 framework [ 13 ] by integrating Sionna RT [ 5 ] in-the-loop, thereby enabling deterministic ray-traced channel prediction and the implementation of Intelligent Transportation Systems (ITS) protocols over full-stack realizations of multiple Radio Access Technologies (RATs) . The resulting framework has been validated against on-field measurements, demonstrating close agreement between predicted and observed channel behavior [ 9 ] . Each platform solves a different part of the problem: the Mobility DT offers realism in geometry and mobility, while VaN3Twin offers realism in the complete communication modeling. Neither, however, provides an end-to-end and real-time pipeline capable of ingesting live mobility, predicting near-future trajectories, computing near-future channels, and request action within a strict deadline. I-A State of the art General-purpose network simulators such as ns-3 and OMNeT+ form the foundation of most V2X simulation frameworks. Among them, ms-van3t is one of the most comprehensive: it supports IEEE 802.11p, LTE-V2X, NR-V2X, and the full ETSI ITS-G5 facilities layer, and integrates with SUMO and CARLA for mobility simulation [ 13 ] . Several recent works interpret such simulators as components of broader vehicular NDTs . Examples include blockage-aware mmWave DTs for multi-hop topologies [ 14 ] and multi-modal DTs combining mobility, sensing, and coarse communication abstractions [ 1 ] . These platforms move beyond offline simulation but still rely on simplified PHY abstractions to maintain responsiveness. Traditional wireless propagation models in ns-3 and OMNeT++ are stochastic (e.g., 3GPP TR 36.885/38.901) or based on analytical families such as WINNER, able to capture large-scale trends but not the site-specific multipath structure. This limitation becomes critical in DT settings, where small changes in geometry or mobility strongly influence short-term channel evolution. To overcome this, several works integrate external ray-based channel simulators into network simulators. MATLAB-based approaches [ 15 ] emphasize PHY accuracy and hardware emulation, but lack full V2X stacks and scale poorly. Other works import offline ray-traced maps into ns-3 [ 3 ] , enabling bit-level simulation but preventing synchronized updates. OPAL–OMNeT++ integration [ 16 ] achieves realistic geometry but targets non-V2X technologies. VaN3Twin, however, advances this direction by coupling ms-van3t with Sionna RT into a comprehensive V2X NDT framework, enabling high-fidelity, in-the-loop ray-traced propagation and realistic modeling of multi-RAT coexistence over shared spectrum portions. However, despite its channel realism, VaN3Twin has primarily been used offline or in emulation mode, without enforcing a strict real-time prediction deadline. GPU-accelerated ray tracing software such as Sionna RT now enable repeated evaluations with competitive runtime [ 21 ] . This has motivated new in-the-loop architectures, including ns3sionna [ 22 ] for Wi-Fi and our prior ns-3 + Sionna RT integration [ 10 ] for multi-RAT 6G studies. DT-CoVeSS [ 17 ] similarly uses Sionna RT for high-fidelity analysis of C-V2X security. These works demonstrate that in-the-loop ray tracing integration is feasible, but none targets city-scale mobility, full V2X communication stacks, and strict prediction deadlines simultaneously. Vehicular DTs outside ns-3 focus increasingly on near real-time operation. RAVEN [ 2 ] combines live data and contextual information to predict mmWave conditions and support interference management. Vehicle-to-Cloud ADAS [ 19 ] demonstrates cloud-assisted actuation using DT logic. Large-scale mobility DTs such as TuST [ 11 ] and indoor crowd models [ 12 ] show that realistic spatio-temporal digital replicas can be built from real data. However, these systems often rely on simplified communication models or separate mobility and communication loops. What is missing is a DT that combines: (i) realistic, city-scale mobility; (ii) a full-stack, ray-traced V2X simulator; and (iii) an explicit timing budget ensuring that predictions are delivered before the corresponding physical time occurs. Paper Contributions To address the gap identified above, we integrate VaN3Twin within the Tokyo Mobility DT ecosystem using Sionna RT as a shared ray tracing core, orchestrating data exchange and scheduling to enable short-term future channel prediction at latencies compatible with real-time V2X operation. Predictions that arrive too late are discarded to ensure consistency with the evolving physical system. This integration exposes the key challenges of real-time NDT operation, including the computational cost of ray-traced propagation, the synchronization of mobility and geometry between platforms, and the scheduling required to deliver actionable

Read the complete text