Purdue University Graduate School
2021_Thesis_mmWave_Yaguang_V3_4_Sejda_Good_288dpi.pdf (10.42 MB)
Download file

Improved Site-Specific Millimeter-Wave Channel Modeling and Simulation for Suburban and Rural Environments

Download (10.42 MB)
posted on 2021-07-28, 20:13 authored by Yaguang ZhangYaguang Zhang
Millimeter-wave (mmWave) bands have become the most promising candidate for enlarging the usable radio spectrum in future wireless networks such as 5G. Since frequent and location-specific blockages are expected for mmWaves, the challenge is understanding the propagation characteristics of mmWave signals and accordingly predicting the channel state information. This research direction has garnered great attention worldwide from industry, academia, and government. However, the majority of current research on mmWave communications has focused on urban areas with high population densities, with very few measurement campaigns in suburban and rural environments. These environments are extremely important for future wireless applications in areas including residential welfare, digital agriculture, and transportation. To fill in this research gap, we developed broadband mmWave channel sounding systems and carried out intensive measurement campaigns at 28 GHz, covering clear line-of-sight as well as non-line-of-sight scenarios over buildings and foliage clutters, to fully characterize the mmWave propagation in suburban and rural environments.

Moreover, the accuracy provided by traditional statistical models is insufficient for next-generation wireless networks with higher-frequency carriers, because they are unable to predict abrupt channel changes caused by site-specific blockages. To overcome this issue, we explored the possibility of utilizing site-specific geographic features such as buildings and trees in improving mmWave propagation models. A new channel modeling methodology highlighting site-specific parameter evaluation based on easily obtainable data sources (e.g., LiDAR) was proposed for accurate, fast, and automated channel state predictions. Accordingly, an overall root mean square error (RMSE) improvement of 11.79 dB was achieved in a one-building blockage scenario and a regional RMSE improvement of over 20 dB was observed in a coniferous forest. This approach also enables channel simulations for large-scale system performance evaluation, demonstrating a powerful and promising approach for planning and tuning future wide-area wireless networks. The simulation results showed that network densification alone is not enough for closing the digital gap, especially with mmWaves because of the impractical number of required towers. They also backed up supplementary solutions including private data relays, e.g., via drones and portable towers.


An Open Source Framework and Community for Sharing Data and Algorithms, Food and Agriculture Research under Award 534662

EARS: Collaborative Research: Real-time Control of Dense, Mobile, Millimeter Wave Networks Using a Programmable Architecture

Directorate for Computer & Information Science & Engineering

Find out more...

SpecEES: Energy-efficient Spectrum and Infrastructure Co-use for Sensing and Communications in Dense Networks

Directorate for Computer & Information Science & Engineering

Find out more...

NSF Engineering Research Center for the Internet of Things for Precision Agriculture (IoT4Ag)

Directorate for Engineering

Find out more...

CIF: Small: Transcoding: A New Approach For Multi-hop Communications

Directorate for Computer & Information Science & Engineering

Find out more...


Degree Type

  • Doctor of Philosophy


  • Electrical and Computer Engineering

Campus location

  • West Lafayette

Advisor/Supervisor/Committee Chair

James V. Krogmeier

Additional Committee Member 2

Christopher R. Anderson

Additional Committee Member 3

Nicolò Michelusi

Additional Committee Member 4

David J. Love

Additional Committee Member 5

Dennis R. Buckmaster