Least squares filtering algorithm for reactive near field probe correction
Abstract
The probe correction technique applied to reactive near field characterization is based on a deconvolution process. However, the classical deconvolution based on an inverse Fourier transform has a restrictive limitation. It is based on the use of noiseless measurement data. Consequently, measurement noise makes the result obtained by the classical deconvolution based technique inefficient and requires an extremely low noise measurement facility. In this paper, a method to improve the probe correction stability when using corrupted measurement data is presented. The proposed constrained least squares filtering algorithm (CLSF) uses an inverse filtering approach that takes into account the statistical characteristics of the measurement noise. Computations data with electromagnetic software of two different structures validate this method and illustrate its reliability