Enabling Site-Specific Cellular Network Simulation Through Ray-Tracing-Driven ns-3
Tanguy Ropitault 1 2 , Matteo Bordin 4 , Paolo Testolina 4 , Michele Polese 4 , Pedram Johari 4 , Nada Golmie 3 , Tommaso Melodia 4 · 2025
Read Enabling Site-Specific Cellular Network Simulation Through Ray-Tracing-Driven ns-3 by Tanguy Ropitault 1 2 , Matteo Bordin 4 , Paolo Testolina 4 , Michele Polese 4 , Pedram Johari 4 , Nada Golmie 3 , Tommaso Melodia 4 (2025) 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-2508-04004.txt.
Introduction
Tanguy Ropitault and co-authors give a cellular network simulator a memory of real streets in a 2025 paper. Testing 5G and future 6G systems needs simulation, but standard channel models deal in averages: they cannot know that this corner diffracts, that canyon blocks, or that rooftop sees the tower directly. Site-specific truth lives in ray tracing, which follows each signal path through actual geometry.
Their module plugs ray-traced multipath data, or field measurements, into the ns-3 simulator's established 5G stack. It builds the frequency-domain channel from those paths and hands it to the untouched physical and access layers. The result keeps full standards compatibility while seeing the actual site. The authors frame it as a building block for network digital twins, enabling beam management and blockage studies that averages wash out.
Catalog record
| CREATOR | Tanguy Ropitault 1 2 , Matteo Bordin 4 , Paolo Testolina 4 , Michele Polese 4 , Pedram Johari 4 , Nada Golmie 3 , Tommaso Melodia 4 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-04004
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Enabling Site-Specific Cellular Network Simulation Through Ray-Tracing-Driven ns-3 Thanks: This work was partially supported by the National Telecommunications and Information Administration (NTIA)’s Public Wireless Supply Chain Innovation Fund (PWSCIF) under Award No. 25-60-IF011, and by the National Science Foundation (NSF) under Award No. CNS 2112471. Tanguy Ropitault 1 2 , Matteo Bordin 4 , Paolo Testolina 4 , Michele Polese 4 , Pedram Johari 4 , Nada Golmie 3 , Tommaso Melodia 4 Abstract 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 corner 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. {tanguy.ropitault,nada.golmie}@nist.gov; {bordin.m, p.testolina, m.polese, p.johari, t.melodia}@northeastern.edu † † footnotetext: The mention of commercial products, their sources, or their use in connection with material reported herein is not to be construed as either an actual or implied endorsement of such products by the Department of Commerce. † † footnotetext: U.S. Government work, not subject to U.S. Copyright I Introduction System–level simulation is an indispensable tool for the design, evaluation, and optimization of 5th generation (5G) New Radio (NR) networks. Among the available simulators, the Network Simulator 3 (ns-3) 5G-LENA module [ 1 ] has established itself as a widely adopted open-source reference platform due to its compliance with 3rd Generation Partnership Project (3GPP) procedures and its support for configurable numerologies, bandwidth parts, and beamforming strategies. In its default configuration, however, 5G-LENA relies on the statistical channel model of 3GPP Technical Report (TR) 38.901 [ 2 , 3 ] (in the remainder of the paper, we refer to 3GPP TR 38.901 as TR 38.901 for brevity), in which large-scale parameters (e.g., path-loss, shadowing) and small-scale clusters (powers, delays, angles) are generated according to scenario-dependent probability distributions. While this approach is well-suited for extensive Monte-Carlo studies and standardization activities, it cannot capture site-specific propagation features, e.g., deterministic blockage or dominant paths that drive the performance of highly directional links. In particular, statistical clusters blur the geometric relationship between the transmitter and the receiver, masking the effects of Line-of-Sight (LoS) occlusion and limiting the realism of beam-management or sensing algorithms that depend on accurate Angle of Departure (AoD) and Angle of Arrival (AoA) information. To close this gap, we introduce a 5G-LENA extension that processes Multipath Component (MPC) traces, allowing ns-3 to compute a site-specific, physically realistic channel matrix through its native pipeline. The proposed trace-based channel model can operate with traces from ray-tracers (e.g., Sionna Ray Tracer (RT) [ 4 ] , Remcom Wireless Insite, etc.) or measurement campaigns. The new module replaces the stochastic cluster synthesis of TR 38.901 with the delays, powers, phases, and angular information contained in the traces, while preserving 5G-LENA’s existing beamforming and scheduling block. Consequently, users can perform site-specific evaluations of link adaptation, beam management and design procedures under the same Medium Access Control (MAC) and Physical (PHY) implementations already available in the simulator. The main contributions of this work are threefold. First, we design and release 1 1 1 https://github.com/usnistgov/siolena SioLENA, an ns-3 module that builds frequency-domain channel matrices from ray traces, modeling location-specific multipath and Doppler effects. Second, we maintain full compatibility with 5G-LENA release 4 without modifying its core scheduler, physical layer abstraction, or antenna models, thereby enabling a high-fidelity alternative to the native TR 38.901 channel model. Third, we validate the module in two realistic three dimensional (3D) outdoor scenarios, showing that the proposed channel model reproduces the expected angular behavior of LoS links and captures the pronounced end-to-end metrics degradation that occurs in Non-Line-of-Sight (NLoS) , effects that the statistical model largely smooths out. Our ns-3 extension, by integrating detailed propagation traces with a 3GPP -compliant network stack, enables reproducible research in advanced beamforming, positioning, and sensing in complex environments, and supports Artificial Intelligence (AI) / Machine Learning (ML) data collection for wireless networks. Because realistic propagation modeling is crucial for Digital Twin (DT) use cases, integrating it with ns‑3’s system‑level simulator, which offers 5G core capabilities, marks a substantial stride toward practical DT realization. The remainder of the paper is organized as follows. Section II reviews the state of the art and related works, highlighting our contributions. In Section III , we offer a brief overview of the 5G-LENA NR module and describe our trace‐based channel model extension. Section IV presents the beamforming validation results in the Place de l’Étoile scenario, analyzes end‐to‐end metrics in a Boston street canyon scenario, and discusses the implications of trace‐based versus statistical modeling. Finally, Section V concludes the paper and outlines directions for future work. II Related Works Stochastic channel models [ 2 , 5 , 3 ] are extremely effective in capturing general channel properties, but fail to reliably capture the propagation characteristics of specific environments. In this context, RTs enable physically consistent, site-specific full-stack simulations [ 6 ] , at the cost of a higher computational complexity. This shift toward realistic modeling aligns with the DT paradigm, where virtual representations aim to accurately mirror the physical network’s behavior and dynamics. Among early efforts, [ 7 ] integrated a Quasi-Deterministic (Q-D) channel model [ 8 ] into the IEEE 802.11ad/ay ns-3 implementation, providing for the first time trace-based channel generation capability to ns-3 . The Q-D methodology, specifically designed to accurately model IEEE 802.11ad/ay channels, generates MPCs based on the propagation conditions in a specific 3D environment, combining a deterministic ray-tracing component (large-scale fading) with a stochastic process (small-scale fading) whose distribution was extracted from an extensive measurement campaign [ 8 ] . The authors of [ 6 ] added similar capabilities to the 5G millimeter wave (mmWave) module. More recently, RTs that exploit the parallelization capabilities of Graphics Processing Units (GPUs) , e.g., Sionna RT [ 4 ] , dramatically reduced the channel computation time. VaN3Twin framework [ 9 ] embeds Sionna RT into the ms-van3t vehicular simulator via a modular UDP client-server