Improving Path Loss Prediction Using Environmental Feature Extraction from Satellite Images: Hand-Crafted vs. Convolutional Neural Network
Usman S. Sani's paper published in MDPI Applied Sciences
School of Digital Science PhD student in Computer Science Usman S. Sani has his recent work “Improving Path Loss Prediction Using Environmental Feature Extraction from Satellite Images: Hand-Crafted vs. Convolutional Neural Network“ published in Applied Science journal.
In this work, he and his team developed a path loss model based on the Extreme Gradient Boosting (XGBoost) algorithm, whose inputs are numeric (non-image) features that influence path loss and features extracted from images composed of four tiled satellite images of points along the transmitter to receiver path.
Path loss prediction is important in the design of cell networks with parameters such as cell radius, antenna heights, and the number of cell sites that can be set.
The developed model can predict path loss for multiple frequencies, antenna heights, and environments such that it can be incorporated into Radio Planning Tools. CNN-extracted and hand-crafted features, as well as their combinations were applied to the images in order to determine the best input features, which, when combined with non-image features, will result in the best XGBoost model.
The best model was obtained using image features extracted from CNN and GLCM combined with the non-image features. A result of RMSE improvement of 9.4272% against a model with non-image features only without satellite images. The XGBoost model performed better than Random Forest (RF), Extreme Learning Trees (ET), Gradient Boosting, and K Nearest Neighbor (KNN) based on the combination of CNN, GLCM, and non-image features.
The variations in features’ values with output path loss values were presented using SHAP summary plots. The interactions observed between some features based on their dependence plots from the computed SHAP values, when further explored, could serve as the basis for the development of an explainable/glass box path loss model.
—
Universiti Brunei Darussalam
Jalan Tungku Link, BE1410
Negara Brunei Darussalam
office.sds@ubd.edu.bn