3D Modeling for Ray Tracing Systems
(2026) EITM02 20251Department of Electrical and Information Technology
- Abstract
- Modern wireless communication systems are increasingly operating in high-frequency bands to support applications requiring low latency, high bandwidth, and high data rates. Achieving reliable performance in these bands requires accurate channel characterization, particularly in complex indoor settings such as industrial environments, where numerous objects influence radio signal propagation through various propagation effects, such as reflections and scattering. Consequently, a detailed knowledge of the surrounding environment is essential for accurate channel modeling. Previous work has relied solely on LiDAR to capture environmental geometry. However, LiDAR has limitations in certain challenging sensing scenarios, motivating the... (More)
- Modern wireless communication systems are increasingly operating in high-frequency bands to support applications requiring low latency, high bandwidth, and high data rates. Achieving reliable performance in these bands requires accurate channel characterization, particularly in complex indoor settings such as industrial environments, where numerous objects influence radio signal propagation through various propagation effects, such as reflections and scattering. Consequently, a detailed knowledge of the surrounding environment is essential for accurate channel modeling. Previous work has relied solely on LiDAR to capture environmental geometry. However, LiDAR has limitations in certain challenging sensing scenarios, motivating the integration of complementary sensor modalities to enhance the fullness and reliability of captured geometric information.
This thesis proposes a multi-sensor framework that fuses LiDAR and stereo camera data to reconstruct detailed indoor 3D environment models for ray tracing-based channel analysis. The results demonstrate that the proposed approach outperforms LiDAR-only methods by providing more accurate environment reconstructions, thereby theoretically enabling more reliable wireless channel characterization. (Less) - Popular Abstract
- Wireless communication systems perform better when they have prior knowledge of their surrounding environment. This information can be used to predict how signals travel from the transmitter to the receiver. Ray tracing is a software-based technique that predicts how such signals propagate within a specific scene by simulating their behavior in a digital replica of the real-world environment. The accuracy of the signal predictions depends on how well the replica matches the physical environment.
Ray tracing in an indoor environment is particularly challenging as the environment contains many objects and various materials that affect how signals travel. Signals can bounce off surfaces, be blocked, or take multiple paths to reach their... (More) - Wireless communication systems perform better when they have prior knowledge of their surrounding environment. This information can be used to predict how signals travel from the transmitter to the receiver. Ray tracing is a software-based technique that predicts how such signals propagate within a specific scene by simulating their behavior in a digital replica of the real-world environment. The accuracy of the signal predictions depends on how well the replica matches the physical environment.
Ray tracing in an indoor environment is particularly challenging as the environment contains many objects and various materials that affect how signals travel. Signals can bounce off surfaces, be blocked, or take multiple paths to reach their destination. In recent years, the importance of indoor wireless communications has increased, necessitating the ability to use ray tracing in indoor environments. Consequently, having an exact and detailed representation of the indoor scene is crucial for understanding how wireless signals behave and for designing systems that ensure reliable signal transmission.
This thesis aimed to further develop a state-of-the-art system used for the construction of digital maps. The proposed method is based on a software using only a single camera for visual input. Integrating a second camera into the system enables depth estimation, similarly to how the human brain perceives depth using information from both eyes. By fusing the information from the additional camera with the rest of the system, the proposed method achieved higher accuracy in both qualitative (visual) and quantitative (geometric) evaluations.
Using the implemented system for the creation of digital ray tracing maps would likely provide more accurate propagation predictions compared to the state-of-the-art system. This can improve the design and implementation of indoor wireless networks, resulting in a more reliable and stable communication link. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/student-papers/record/9252074
- author
- Gustafsson, Otto LU and Ahmed, Hayam LU
- supervisor
- organization
- course
- EITM02 20251
- year
- 2026
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- Ray Tracing, SLAM, Channel Modeling, Computer Vision, Sensor Fusion
- report number
- LU/LTH-EIT 2026-1196
- language
- English
- id
- 9252074
- date added to LUP
- 2026-10-01 10:32:42
- date last changed
- 2026-10-01 10:32:42
@misc{9252074,
abstract = {{Modern wireless communication systems are increasingly operating in high-frequency bands to support applications requiring low latency, high bandwidth, and high data rates. Achieving reliable performance in these bands requires accurate channel characterization, particularly in complex indoor settings such as industrial environments, where numerous objects influence radio signal propagation through various propagation effects, such as reflections and scattering. Consequently, a detailed knowledge of the surrounding environment is essential for accurate channel modeling. Previous work has relied solely on LiDAR to capture environmental geometry. However, LiDAR has limitations in certain challenging sensing scenarios, motivating the integration of complementary sensor modalities to enhance the fullness and reliability of captured geometric information.
This thesis proposes a multi-sensor framework that fuses LiDAR and stereo camera data to reconstruct detailed indoor 3D environment models for ray tracing-based channel analysis. The results demonstrate that the proposed approach outperforms LiDAR-only methods by providing more accurate environment reconstructions, thereby theoretically enabling more reliable wireless channel characterization.}},
author = {{Gustafsson, Otto and Ahmed, Hayam}},
language = {{eng}},
note = {{Student Paper}},
title = {{3D Modeling for Ray Tracing Systems}},
year = {{2026}},
}