Retrieving Tensor Permeability of Ferrite by Transmission Spectrum of Ferrite Rod Array

We proposed a method for retrieving tensor permeability of ferrites by means of transmission spectra of ferrite rod array, implemented by a fully connected neural network and a particle swarm optimization algorithm. The neural network can accurately and quickly predicate the corresponding the transmission $S_{21}$ by the given tensor permeability at a specific frequency, greatly reducing the time spent on numerical simulations. By combining this neural network with the particle swarm algorithm, we were able to retrieve the tensor permeability parameters from the transmission spectrum, which were verified by simulations and experiments.