Investigating Map-Based Path Loss Models: A Study of Feature Representations in Convolutional Neural Networks
This work addresses path loss prediction for radio frequency spectrum management, but it is incremental as it builds on prior map-based models.
The paper tackled the problem of improving path loss prediction for efficient radio spectrum use by investigating different input representations for convolutional neural networks, finding that representing scalar features as image channels yielded the strongest generalization.
Path loss prediction is a beneficial tool for efficient use of the radio frequency spectrum. Building on prior research on high-resolution map-based path loss models, this paper studies convolutional neural network input representations in more detail. We investigate different methods of representing scalar features in convolutional neural networks. Specifically, we compare using frequency and distance as input channels to convolutional layers or as scalar inputs to regression layers. We assess model performance using three different feature configurations and find that representing scalar features as image channels results in the strongest generalization.