Comparing Machine Learning Methods for Air-to-Ground Path Loss Prediction

Abstract

Machine Learning-based models gain increasingly momentum regarding the problem of path loss prediction. The work at hand deploys four machine learning algorithms (k Nearest Neighbors - kNN, Support Vector Regression - SVR, Random Forest - RF and AdaBoost), in order to simulate the radio coverage provided from a flying base station in the greek city of Tripolis. Their comparison shows that tree-based ensemble models (RF and AdaBoost) can be used as fast and reliable alternatives to the Ray Tracing technique.