| Popis: |
In this paper, the artificial neural network is used to find the number of equivalent magnetic dipoles needed by the reconstruction of interference sources. First, we obtain the radiation field generated by the unknown source, and then obtain its plane wave spectrum through the two-dimensional Fourier transform. The amplitude pattern of the plane wave spectrum is used to find the number of magnetic dipoles, where the artificial neural network is trained for this purpose. In this way, we can obtain more accurate information of the number of magnetic dipoles, which is help for the reconstruction of the interference source during the electromagnetic interference analysis. |
