FDTD near field data reduction using principal component analysis

A time domain spherical multipole near-to-far-field algorithm running together with a principal component analysis (PCA) signal processing technique is introduced in this paper. Such approach is employed to reduce the near fields obtained by a finite-difference time-domain (FDTD) code that will be used as input to a spherical multipole near-to-far-field transformation algorithm. The technique is applied to obtain the radiation pattern of a half-wavelength dipole, thus validating the code, and also for a configuration of two array dipole. Preliminary results show that when we compare the original signal with the reconstructed using a threshold value of 0.1 the required near-field data is reduced by an average value of 83% for the half-wavelength dipole and also shown that even if the technique is more sensitive in the presence of lobes it can significantly reduce the near-field components.