Optimal Artificial Neural Network design for propagation path-loss prediction using adaptive evolutionary algorithms

In this paper we present an alternative procedure for the prediction of propagation path loss in urban environments, which is based on Artificial Neural Networks (ANN). The size of a neural network must be defined before it can be trained for any application. We apply different adaptive Differential Evolution (DE) algorithms, in order to design an optimal ANN for path loss propagation prediction. We present two different ANN design cases with two and three hidden layers respectively. The general performance of the both ANN shows their effectiveness to yield results with satisfactory accuracy in short time. The received results are compared to the respective ones yielded by the Ray-Tracing model and exhibit satisfactory accuracy.