Low-Complexity Deterministic Channel Modeling for Outdoor Radio Propagation
Iftikhar, Muhammad Junaid (2025)
2025
Sähkötekniikan DI-ohjelma - Master's Programme in Electrical Engineering
Informaatioteknologian ja viestinnän tiedekunta - Faculty of Information Technology and Communication Sciences
Hyväksymispäivämäärä
2025-12-02
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-2025113011085
https://urn.fi/URN:NBN:fi:tuni-2025113011085
Tiivistelmä
Propagation models are fundamental to the advancement of wireless communication, as they enable us to predict and optimize how electromagnetic (radio) waves travel through different environments. Traditional propagation models, such as empirical ones, estimate average signal behavior based on measurements from many locations. Though they are simple and fast, they do not adapt well to new or unique environments. On the other hand, deterministic models can achieve high accuracy by simulating the actual physical environment. However, they are often complex and computationally heavy. These limitations motivated us to develop an efficient and flexible framework that balances realism, automation, and computational performance for studying wireless propagation.
This project presents a practical approach to create realistic three-dimensional (3D) models and use them to study how radio signals interact to our surrounding areas. The main goal is to build accurate 3D models from our real-world environments and employ them for wireless communication simulations while balancing computation complexity and time.
The 3D models were initially designed using Blender, an open-source 3D design software, along with the Blosm add-on, allowing the import of real-world map data from OpenStreetMap (OSM) and Google 3D Tiles, making it possible to automatically create real-world scenes as per our required areas that include buildings, roads, terrain, and vegetation based on real geographic coordinates. After the creation of the models, we exported individual objects in the map separately as .OBJ files and also converted Blender’s 3D coordinates into latitude and longitude by applying customized Python scripts. This step was crucial to keep the models organized and correctly linked to their real-world locations and to ensure that every building stayed georeferenced and spatially accurate. This automated process saved time and reduced the chances of human error when handling many buildings.
Next, we converted the .OBJ files into normals (surface orientation) and point clouds, which are small, evenly spaced (in our case 0.05 meters apart) surface data using PyWavefront and Trimesh (Python libraries). This process helps us to analyze how electromagnetic waves reflect and scatter off surfaces.
Finally, the Channel Impulse Response (CIR) was simulated within this 3D environment by using building geometry, surface details, and physical parameters to estimate how radio signals from a transmitter reflect, travel, and arrive at a receiver. Both direct line-of-sight (LoS) and reflected paths were analyzed to build a complete picture of signal behavior in our outdoor surrounding.
In summary, this work demonstrates an automated, scalable, efficient and cost-effective framework that integrates open-source tools for realistic 3D models and radio wave simulation. The proposed method overcomes the limitations of traditional empirical and deterministic models—providing a flexible foundation for future wireless communication research and planning.
This project presents a practical approach to create realistic three-dimensional (3D) models and use them to study how radio signals interact to our surrounding areas. The main goal is to build accurate 3D models from our real-world environments and employ them for wireless communication simulations while balancing computation complexity and time.
The 3D models were initially designed using Blender, an open-source 3D design software, along with the Blosm add-on, allowing the import of real-world map data from OpenStreetMap (OSM) and Google 3D Tiles, making it possible to automatically create real-world scenes as per our required areas that include buildings, roads, terrain, and vegetation based on real geographic coordinates. After the creation of the models, we exported individual objects in the map separately as .OBJ files and also converted Blender’s 3D coordinates into latitude and longitude by applying customized Python scripts. This step was crucial to keep the models organized and correctly linked to their real-world locations and to ensure that every building stayed georeferenced and spatially accurate. This automated process saved time and reduced the chances of human error when handling many buildings.
Next, we converted the .OBJ files into normals (surface orientation) and point clouds, which are small, evenly spaced (in our case 0.05 meters apart) surface data using PyWavefront and Trimesh (Python libraries). This process helps us to analyze how electromagnetic waves reflect and scatter off surfaces.
Finally, the Channel Impulse Response (CIR) was simulated within this 3D environment by using building geometry, surface details, and physical parameters to estimate how radio signals from a transmitter reflect, travel, and arrive at a receiver. Both direct line-of-sight (LoS) and reflected paths were analyzed to build a complete picture of signal behavior in our outdoor surrounding.
In summary, this work demonstrates an automated, scalable, efficient and cost-effective framework that integrates open-source tools for realistic 3D models and radio wave simulation. The proposed method overcomes the limitations of traditional empirical and deterministic models—providing a flexible foundation for future wireless communication research and planning.