3D Ray Tracing for Site Specific Radio Propagation

30 concepts26 papers1 proposal

Traditional wireless network planning relies heavily on statistical channel models. These models treat the physical environment as a stochastic abstraction, predicting signal degradation based on distance and generalized terrain categories. While this approach works well for low-frequency signals that propagate widely, it fails at higher frequencies—such as 5G millimeter-wave (mmWave) and upcoming 6G bands. At these shorter wavelengths, radio waves behave more like light: they are easily blocked by concrete walls, reflected by glass windows, and scattered by foliage.

To reliably deploy next-generation wireless systems, engineers must move away from statistical guesswork and toward deterministic, physics-based modeling. This is achieved through site-specific 3D ray tracing. By simulating the explicit geometric trajectories of radio waves within a high-fidelity digital replica of an environment, ray tracing captures the exact physical interactions that dictate signal strength, phase, and direction.


Reconstructing the 3D Environment

A ray-tracing simulation is only as accurate as the digital map it traverses. Instead of relying on flat 2D layouts, modern propagation modeling begins by reconstructing detailed 3D scenes that capture both geometric structures and material properties.

[Physical World] ---> [LiDAR / Point Clouds] ---> [Semantic 3D Model] ---> [Ray-Tracing Engine]
                                                    (Walls, Windows)         (Reflections, UTD)

For indoor environments, capturing fine-grained structural details is critical because internal partitions dictate signal boundaries. Cui et al. demonstrate that mobile laser scanning point clouds can be used to automatically reconstruct 3D indoor models. By explicitly representing structural elements like walls, doors, and windows, these models provide the necessary geometric foundation for simulating indoor 5G signal behavior. To achieve high fidelity, these environments must also incorporate semantic and material awareness; Zhang et al. emphasize that precise geometric models of objects combined with their specific material properties are essential for accurate, fine-grained radio propagation modeling.

Outdoors, the complexity scales to include terrain, buildings, and vegetation. To optimize radio frequency coverage for a campus-scale private 5G network, Ferguson et al. combine building models from open-source mapping data with high-resolution foliage details extracted from LiDAR data.

However, tracing millions of rays through highly detailed environments is computationally expensive. To balance accuracy and performance, researchers often employ geometric simplifications. Liu et al. represent complex urban structures as volumetric geometric primitives to efficiently calculate non-line-of-sight (NLoS) blockages for unmanned aerial vehicles (UAVs). Similarly, Zhu et al. demonstrate that stripping away minor environmental details from digital maps can reduce simulation time by 50% while maintaining nearly identical statistical channel properties.


The Physics of Ray Propagation

Once a 3D digital environment is established, a ray-tracing engine simulates how electromagnetic waves interact with physical surfaces. This requires modeling three primary wave phenomena: reflection, diffraction, and refraction.

Reflection and Polarization

When a radio wave strikes a flat surface like a building facade, it undergoes specular reflection. The power and polarization of the reflected wave depend on the angle of incidence and the material's electrical permittivity. Lim et al. utilize 3D ray-tracing simulations to capture these site-specific interactions, verifying how material reflections cause polarization leakage and spatial correlation in indoor MIMO systems.

In highly dynamic scenarios, such as mmWave beam management, calculating full multipath reflections in real time is prohibitive. To solve this, Massoli et al. employ a simplified ray-tracing mechanism that focuses exclusively on first-order reflections within a vision-reconstructed 3D environment, successfully predicting the angular power profile of the radio channel with minimal overhead.

Diffraction around Obstacles

High-frequency signals do not stop completely when they hit a corner; instead, they bend around edges, creating a shadow zone. To model this behavior, ray-tracing frameworks integrate the Uniform Theory of Diffraction (UTD). Zhu et al. apply UTD within their 3D ray-tracing framework to simulate how mmWave signals bend around building edges in non-stationary UAV environments. This physics-based approach is equally valuable for localization; Aly et al. augment ray tracing with UTD to predict signal behavior around physical obstructions, even modeling human bodies as metallic cylinders to simulate device-free localization maps.

Atmospheric Refraction

At macroscopic scales—such as satellite communications or Global Navigation Satellite System (GNSS) meteorology—the primary propagation challenge is not building blockages, but the atmosphere itself. As radio waves pass through layers of varying atmospheric density, they bend. Feng et al. use a specialized ray-tracing algorithm to precisely calculate this curved path through the troposphere, accounting for atmospheric refraction to determine precise propagation delays.


Synchronizing the Network Digital Twin

The ultimate application of site-specific ray tracing is the creation of a Network Digital Twin (NDT)—a real-time, virtual mirror of the physical radio environment. As Bariah et al. highlight, high-fidelity virtual replicas achieved through ray tracing and multi-modal sensing allow for near-perfect alignment between physical signals and their digital representations, enabling zero-touch network automation.

+----------------+                  +-------------------+
| Physical World | --Sensors/GPS--> | 3D Digital Twin   |
| (Base Station) |                  | (Ray-Tracing Engine)|
+----------------+                  +-------------------+
        ^                                     |
        |                                  Predicts
        |                                     v
        +---------Optimize Beam/Path----------+

These digital twins allow autonomous systems to predict and adapt to radio environments before moving through them:

  • Mobile Robotics: Roda-Sánchez et al. integrate ray-tracing simulations directly into an NDT framework for mobile robots, allowing vehicles exploring unknown spaces to predict radio quality and maintain reliable, energy-efficient connections.
  • UAV Communications: In low-altitude drone networks, real-world channel measurements are difficult to collect. Han et al. use ray tracing to simulate air-to-air propagation, generating detailed multipath data (delay, power, and angles of arrival) to train generative channel models. To optimize flight paths, Sun et al. leverage realistic ray-tracing tools with actual city map data to validate energy-efficient UAV trajectory training.
  • Edge Computing and Infrastructure: To evaluate high-bandwidth mobile edge systems, Yuan et al. implement a full-stack simulation that couples network modeling with site-specific ray tracing to account for physical blockages in dense urban environments. For network deployment, Ferguson et al. demonstrate that GPU-accelerated ray tracing can predict urban signal coverage prior to physical hardware installation, dramatically reducing deployment risks.

By anchoring network simulations in 3D spatial coordinates, engines can apply standardized 3D pathloss models to simulate advanced features like beamforming and non-line-of-sight propagation. As Sthankiya et al. note, this spatial awareness provides a safe, high-speed digital twin environment to train optimization models without risking live network outages. Furthermore, these physics-grounded baselines can even enhance security; Wang et al. incorporate Rician fading and obstacle-based attenuation models via ray tracing to simulate realistic urban communication environments, providing a "ground truth" proxy to detect falsified location reports and "ghost" vehicles in autonomous driving applications.

History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_ray_tracing_simulation_d78a70f5,
  title = {ray tracing simulation},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/ray-tracing-simulation/rdr_d78a70f5436741cd8bd65406be3527bd},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}

This exact version (version 1, frozen text):

@misc{lacuna_ray_tracing_simulation_d78a70f5_v1,
  title = {ray tracing simulation},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/ray-tracing-simulation/rdr_d78a70f5436741cd8bd65406be3527bd/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake