Neural networks model of an UWB channel path loss in a mine environment

this paper aims to present an UWB propagation channel modeling with neural networks in a mine environment, focusing on the change in the path loss attenuation as a function of distance and frequency. A trained neural network can be used for high-level design, providing fast and accurate answers to the task it has learned. Neural networks are effective alternatives methods to conventional methods such as statistical and stochastic modeling methods, which could be computationally expensive,or to analytical methods which could be difficult to obtain for new environments, or to empirical modeling solutions which range and accuracy may be limited.