An accurate neural network approach in modeling an UWB channel in an underground mine

Modeling an ultra-wideband (UWB) channel is an important and challenging task in wireless communications. Modeling a channel in an underground mine environment presents additional challenges and difficulties. Many researchers and techniques have treated this subject. In this paper we will present a new approach in modeling the channel in an underground mine by using artificial neural networks (ANN) of type RBF (Radial basis function) focusing on the change of the path loss attenuation as a function of distance and frequency. Results presented show the accuracy of this method.