Non-linear modeling of parabolic reflector: Neural network approach

In this paper, we present artifical neural network (ANN) models for the calculation of the focal region fields of a parabolic reflector and the estimation of the direction of arrival. We choose a feedforward NN architecture (specifically multilayer perceptrons (MLPs)) for their robust and massively interconnected structure. Simulated results achieved with these models show that the neural network approach is a promising alternative to the conventional methods for focal region calculations of the parabolic reflectors. This paper also represents the results obtained from a NN model trained for the estimation of the direction of arrival for the array-fed parabolic reflector.