A Neural Network based Model for Signal Coverage Propagation Loss Prediction in Urban Radio Communication Environment

For proper planning and optimisation of radio network coverage quality at the mobile station terminals, prediction of propagation path loss with reasonable accuracy is important. This research work proposes the application of a hybrid neural modelling technique to predict of signal coverage propagation losses in typical urban environment. The modelling technique is based on combining a conventional Log-distance model and a neural network. The hybrid model employs the adaptive neural learning techniques of multilayer Levenberg Marquardt backpropagation algorithm to outfit for the errors obtained by applying only conventional model in urban microcellular environment. By using the mean absolute error, root mean square error and standard deviation performance evaluation metrics, the hybrid – based algorithm provides more accurate prediction results with measured values compared to the conventional approach. The computationally effective prediction technique of the hybrid based neural network model can be used for tuning and enhancing conventional prediction methods.