sensitivity of the sensor when moving to objects with a greater dielectric constant. Also
on all the graphs it is seen that with the increase of the gap between the object and the
probe tip h
Z
the signal of the resonant frequency shift falls, as well as the quality shift,
which indicates the obvious losses in the sensitivity of the measurement. From these
data, we can make our conclusions on that the range of applicability of SMM. In
addition, it can be seen that all characteristics have a sharply nonlinear character, which
does not allow us to directly obtain information on the quantitative distribution of
parameters for the object under study, this information can only be of a qualitative
nature. Therefore, it is necessary to carry out preliminary processing to obtain a real
picture of the distribution of parameters. To obtain reliable data due to which it would
be possible to judge the magnitudes of the electrophysical parameters of an object, it is
necessary to solve the inverse problem.
One of the methods for solving the inverse problem is considered in [23], where
the method of inverse convolution, generally accepted in the reconstruction method, is
used. However, the author himself acknowledges the incorrectness of this approach in
connection with the fact that the scan result can be written in the form of a two-
dimensional convolution only as the first approximation of the perturbation method.
Recommendations for determining the core of integration by the experimental method
are given, but this approach does not eliminate the problem, but only gives an optimal
approximation to the used method. In addition, it is proposed to use parameters of the
aperture microwave sensor, such as the frequency, the size of the aperture, etc., as a
parameter that additionally varies to obtain an array of two-dimensional scanning
results, without analyzing the sensitivity and stability of the corresponding
reconstruction algorithms.
Broadly a similar problem was solved in work [24] in the development of the
method of probing by microwave radiation of a multilayer medium. An iterative
algorithm for solving the inverse problem was used, based on the method of successive
iterations of the unperturbed field. However, the specific nature of this technique
requires initial data on the distribution of the electric field on the surface of the
controlled medium, depending on the frequency of radiation or the height of the
location of the radiator. In fact, this method assumes an independent emitter and
receiver of the electromagnetic field, and the dimensions of the receiver are negligible.
Scanning is performed only by moving the receiver in a given plane above the surface
of the medium. Thus, this technique is primarily aimed at solving the problem of
microwave location of multilayer media in the near-field approximation, and cannot be
used for technical reasons for microwave diagnostics of semiconductors. However,
some mathematical techniques used by the authors can also be applied to our problem.
Most of the volume distribution reconstruction algorithms based on scanning data
are based, firstly, on solving direct problem of sensor interaction with a homogeneous
- 1552 -