An Automatically Converging Scheme Based on the Neural Network and Its Application in Antennas

In this paper, an automatically converging scheme based on the neural network and its application in antennas are presented. The proposed scheme can automatically converge toward the optimum of the system output without final solutions of system output as learning samples. Initially, the input and output of a complicated system are modeled by a neural network. By suitably extending the neural network, the gradient information of the system output is obtained and the scheme then automatically converges toward the optimum of the system output. Our scheme can be applied not only to optimization, but also to analyses of antenna problems.