ELECTRICAL ENGINEERING, cilt.86, sa.3, ss.157-163, 2004 (SCI-Expanded)
A new method based on artificial neural networks for calculating the narrow aperture dimension of the pyramidal horn is presented. The Levenberg-Marquardt algorithm is used to train the networks. The narrow aperture dimension calculated using artificial neural networks is used in the optimum gain pyramidal horn design. The computed gains of the designed pyramidal horns are in very good agreement with the desired gains.