Map-Based Path Loss Prediction in Multiple Cities Using Convolutional Neural Networks
This work addresses path loss prediction for radio spectrum planning, offering a domain-specific improvement over traditional methods.
The paper tackled the problem of predicting path loss for radio deployments by using convolutional neural networks to extract features directly from 2-D obstruction height maps, achieving low prediction error across various environments without relying on derived metrics.
Radio deployments and spectrum planning benefit from path loss predictions. Obstructions along a communications link are often considered implicitly or through derived metrics such as representative clutter height or total obstruction depth. In this paper, we propose a path-specific path loss prediction method that uses convolutional neural networks to automatically perform feature extraction from 2-D obstruction height maps. Our methods result in low prediction error in a variety of environments without requiring derived metrics.