Physics-Based Machine Learning Propagation Models for Wireless Access Point Placement

点击翻译按钮获取中文摘要

The continuous expansion of wireless communication technologies creates a pressing need for intelligent planning of a plethora of existing and emerging wireless services, in a timely and cost-effective manner. Wireless propagation modeling, which is the prediction of the signal levels generated by a wireless communication system in a given environment, is an essential element of such an intelligent planning process. These models can be deduced by numerical algorithms such as ray-tracing, which are based on the physics of electromagnetic wave propagation. However, physics-based methods require a high level of relevant expertise and significant computational resources. While trading accuracy for efficiency could have been an acceptable solution a few years ago, the emerging, increasingly complex landscape of wireless services requires a proportional advancement in the state of the art in propagation modeling: high-fidelity, simple to derive, computationally efficient models. The main goal of this project is to leverage advances in scientific machine learning to accelerate the application of sophisticated, physics-based propagation analysis methods to wireless system design, facilitating their broad adoption by wireless engineering teams. Therefore, our research can overcome the traditional dichotomy between accuracy and efficiency of propagation models, leading to more cost-effective, fast and robust design for current, emerging and future generations of wireless communication systems. The project is based on a partnership between the applicant's research group at the University of Toronto, a group with a long track record in propagation modeling for wireless communication systems currently focusing on machine learning applications to this area, with iBwave, a Canadian company that is dedicated to improving wireless network design in large venues such as stadiums, malls, and multi-floor office spaces.