De-embedding method for a sensing area characterization of planar microstrip sensors without evaluating error networks.
Sci Rep
; 14(1): 10062, 2024 May 02.
Article
em En
| MEDLINE
| ID: mdl-38698116
ABSTRACT
A de-embedding method for determining all scattering (S-) parameters (e.g., characterization) of a sensing area of planar microstrip sensors (two-port network or line) is proposed using measurements of S-parameters with no calibration. The method requires only (partially known) non-reflecting line and reflecting line standards to accomplish such a characterization. It utilizes uncalibrated S-parameter measurements of a reflecting line, direct and reversed configurations of a non-reflecting line, and direct and reversed configurations of the sensing area. As different from previous similar studies, it performs such a characterization without any sign ambiguity. The method is first validated by extracting the S-parameters of a bianisotropic metamaterial slab, as for a two-port network (line), constructed by split-ring-resonators (SRRs) from waveguide measurements. Then, it is applied for determining the S-parameters of a sensing area of a microstrip sensor involving double SRRs next to a microstrip line. The root-mean-square-error (RMSE) analysis was utilized to analyze the accuracy of our method in comparison with other techniques in the literature. It has been observed from such an analysis that our proposed de-embedding technique has the lowest RMSE values for the extracted S-parameters of the sensing area of the designed sensor in comparison with those of the compared other de-embedding techniques in the literature, and have similar RMSE values in reference to those of the thru-reflect-line calibration technique. For example, while RMSE values of real and imaginary parts of the forward reflection S-parameter of this sensing area are, respectively, around 0.0271 and 0.0279 for our de-embedding method, those of one of the compared de-embedding techniques approach as high as 0.0318 and 0.0324.
Texto completo:
1
Base de dados:
MEDLINE
Idioma:
En
Ano de publicação:
2024
Tipo de documento:
Article