Comparison of neural network models for path loss prediction

Abstract

This work presents the results of the studies concerning the applications of the feedforward neural networks to the prediction of propagation path loss in urban and suburban environment. First, neural network models are designed in order to predict the path loss. Further investigations are made on an error correction model, based on the combination between a theoretical model and a neural network. The performances of the neural models are compared to the measured path loss values from the measurements conducted in the city of Kavala and in Oia village on Santorini Island, Greece, based on the absolute mean square error, standard deviation and root mean square error between predicted and measured values. Also, the neural networks models are compared to each other and to the COST 231-Walfisch-Ikegami.