Universal approximation using radial-basis-function networks
Neural Computation
Approximation and radial-basis-function networks
Neural Computation
Static and Dynamic Neural Networks: From Fundamentals to Advanced Theory
Static and Dynamic Neural Networks: From Fundamentals to Advanced Theory
Networks and the Best Approximation Property
Networks and the Best Approximation Property
| Hi-index | 0.00 |
This paper considers antenna array synthesis for regular antenna array using neural network. Because of the best approximation property, radial basis function neural network is used. Assuming that large number of neurons is available, following the concept of human brain, the choice is made for neural network with exact solution. For given values of the radiation pattern we estimate the phase difference of the excitations between the neighbouring antenna elements. We investigate the required angle step of the training data and sampling for different number of antenna elements and for different element distances. This analyze is necessary starting point for future investigation of neural network solution for irregular antenna array synthesis.