Wireless Propagation with Advanced Ray Tracing Simulations

Abstract

We conducted a study to investigate the propagation characteristics of wireless signals at different frequencies (2.4 GHz, 30 GHz, and 100 GHz) at Northeastern University. By leveraging the power of NVIDIA's Sionna library and creating a detailed 3D model of the university campus, we performed advanced ray tracing simulations to analyze channel impulse response, delay spread, path loss, and bit error rates(BERs). Our findings provide valuable insights for network planning, frequency selection, and the development of future wireless technologies.

Effective modeling and simulation of wireless communication systems are essential for researchers and network operators to understand signal propagation in complex environments. Sionna, an open-source library developed by NVIDIA, offers a high-level API for rapid modeling of end-to-end communication systems. In this project, we harnessed Sionna's capabilities to explore wireless propagation at three key frequencies on the Northeastern University campus.

Ray Tracing in Wireless Research

Indoor Research Applications

Ray tracing has been used to study the propagation of wireless signals in complex, small spaces such as offices, factories, and shopping malls. These simulations help predict coverage gaps, optimize access point placement, and mitigate interference between devices. For instance, modeling the propagation of 2.4 GHz and 5 GHz signals in indoor environments helps optimize access point locations and antenna configurations to maximize coverage and minimize dead spots.

Outdoor Research Applications

In outdoor environments, ray tracing has been applied to scenarios from cellular networks to satellite links. By incorporating detailed models of buildings, terrain, and vegetation, these simulations provide accurate predictions of signal propagation in urban and rural settings. This helps in the planning and optimization of cellular networks by simulating signal propagation from base stations to mobile devices, identifying coverage gaps, and evaluating network upgrades and expansions.

Scenario Development

3D Environment Modeling

Sionna provides advanced capabilities for modeling the physical environment to enable realistic ray tracing simulations. We created a detailed 3D model of the Northeastern University campus using Blender, OpenStreetMap (OSM), and Mitsuba. The model included buildings, streets, which were then rendered in Mitsuba for advanced lighting and material effects, and integrated into Sionna for simulation.

Simulation Parameters

We defined relevant frequency bands (2.4 GHz for Wi-Fi, 30 GHz for 5G, and 100 GHz for 6G) and the physical environment of the Northeastern University campus. Geographic data from OSM was used to create a realistic representation of the campus, including buildings and natural terrain. The detailed 3D model served as the foundation for ray tracing simulations to evaluate wireless communication performance.

Simulations and Analysis

2.4 GHz Frequency Band

The simulations at 2.4 GHz revealed several key findings. The channel impulse response showed that scattering effects contributed to a more pronounced delay spread and a higher number of multipath components. The delay spread was significantly lower than at higher frequencies, attributed to the longer wavelength and better penetration capabilities of 2.4 GHz signals. The path loss was the lowest among the frequencies studied, resulting in the best BER performance.

30 GHz Frequency Band

The simulations at 30 GHz showed a more pronounced delay spread and a higher number of multipath components compared to 2.4 GHz. The delay spread was higher due to the shorter wavelength and higher susceptibility to obstacles. The path loss was higher, resulting in a slight degradation in BER performance.

100 GHz Frequency Band

At 100 GHz, the simulations revealed the most challenging propagation conditions. The delay spread was the highest, and the path loss was extreme, leading to severe signal loss and the worst BER performance among the frequencies studied. These findings indicate the need for advanced techniques to maintain reliable communication at higher frequencies.

Conclusion

In this project, we leveraged the advanced ray tracing capabilities of NVIDIA’s Sionna library to study the propagation characteristics of wireless signals at different frequencies on the Northeastern University campus. The results demonstrated that the 2.4 GHz band offers the best overall performance, while the 100 GHz band poses significant challenges. The insights gained from this study can inform network planning, frequency selection, and the development of advanced communication technologies. Sionna proves to be a powerful tool for wireless research, enabling realistic simulations and integration with machine learning workflows. Future work could extend this study to include more frequency bands, different environments, and various network configurations.