Knowledge base: Warsaw University of Technology

Settings and your account

Back

The application of GPR in soil permittivity estimation

Mariusz Zych

Abstract

In this thesis, the novel methods of soil permittivity estimation, based on GPR image, is presented. Ground penetrating radars are one of the main fields of advancement of modern radiolocation. Because of its broad capabilities, GPR is applied to detect explosives such as mines. What is more GPR is used for archaeological discoveries, to detect damage of bulwarks or roads. In order to determine the depth, at which analyzed object is located, the knowledge of soil permittivity value is required. Majority of commercially available ground penetrating radars uses non-coherent Hough transform, which improve image resolution in movement direction. However, much better results can be obtained using synthetic aperture technique SAR. Regardless of the chosen method, its proper operation is dependent on correctly chosen electrical parameters of tested soil. Only in that case, the optimal GPR image resolution is received. This dissertation presents three methods of proper soil permittivity estimation. The first one is based on SAR image contrast optimization. In order to reduce computational complexity of the algorithm, least square approximation of the image contrast function has been applied. The approximation uses few data points. The method is intended to estimate permittivity value of homogeneous soil. Another solution to the problem uses autofocus technique of GPR image. Iterative, autofocus methods are generally applicable in estimation of motion parameters of the radar platform. Proposed, novel algorithm allows to correctly estimate permittivity value of tested soil. The third algorithm presented in the paper shows the permittivity estimation method of non-homogeneous soil. For that purpose, adaptive lattice filters are used. Each of the lattice filter stage corresponds to individual layer of tested soil. The method estimates values of permittivity and the depth of each layers. Theoretical results presented in the dissertation have been verified using computer simulations. In order to do that, mathematical environment Malab has been used. What is more, the author has tested data collected by commercial simulator of electromagnetic field - CST Studio Suite. Obtained simulation results confirm correctness of theoretical analysis showed in the paper. The paper includes the results of actual measurements carried by GPR demonstrator produced by Telecommunication Research Institute. The author shows measurement results of GPR made of devices belonging to Institute of Electronic Systems, Faculty of Electronics and Information Technologies. Presented results clearly convince about the effectiveness of proposed algorithms of soil permittivity estimation and need for further development in this radar field as well.
Record ID
WUTe6d71d06654545da9883eb6df8542277
Diploma type
Doctor of Philosophy
Author
Title in Polish
Zastosowanie radarów GPR do estymacji przenikalności elektrycznej gruntu
Title in English
The application of GPR in soil permittivity estimation
Language
(pl) Polish
Certifying Unit
Faculty of Electronics and Information Technology (FEIT)
Discipline
electronics / (technology domain) / (technological sciences)
Status
Finished
Start date
01-10-2010
Defense Date
25-11-2014
Title date
16-12-2014
Supervisor
Internal reviewers
External reviewers
Mateusz Pasternak Mateusz Pasternak,, External affiliation of publication: Wojskowa Akademia Techniczna im. Jarosława Dąbrowskiego
Pages
156
Keywords in English
GPR, SAR, CST Studio Suite, soil permittivity estimation, adaptive filtration, gradient adaptive lattice filter
Abstract in English
In this thesis, the novel methods of soil permittivity estimation, based on GPR image, is presented. Ground penetrating radars are one of the main fields of advancement of modern radiolocation. Because of its broad capabilities, GPR is applied to detect explosives such as mines. What is more GPR is used for archaeological discoveries, to detect damage of bulwarks or roads. In order to determine the depth, at which analyzed object is located, the knowledge of soil permittivity value is required. Majority of commercially available ground penetrating radars uses non-coherent Hough transform, which improve image resolution in movement direction. However, much better results can be obtained using synthetic aperture technique SAR. Regardless of the chosen method, its proper operation is dependent on correctly chosen electrical parameters of tested soil. Only in that case, the optimal GPR image resolution is received. This dissertation presents three methods of proper soil permittivity estimation. The first one is based on SAR image contrast optimization. In order to reduce computational complexity of the algorithm, least square approximation of the image contrast function has been applied. The approximation uses few data points. The method is intended to estimate permittivity value of homogeneous soil. Another solution to the problem uses autofocus technique of GPR image. Iterative, autofocus methods are generally applicable in estimation of motion parameters of the radar platform. Proposed, novel algorithm allows to correctly estimate permittivity value of tested soil. The third algorithm presented in the paper shows the permittivity estimation method of non-homogeneous soil. For that purpose, adaptive lattice filters are used. Each of the lattice filter stage corresponds to individual layer of tested soil. The method estimates values of permittivity and the depth of each layers. Theoretical results presented in the dissertation have been verified using computer simulations. In order to do that, mathematical environment Malab has been used. What is more, the author has tested data collected by commercial simulator of electromagnetic field - CST Studio Suite. Obtained simulation results confirm correctness of theoretical analysis showed in the paper. The paper includes the results of actual measurements carried by GPR demonstrator produced by Telecommunication Research Institute. The author shows measurement results of GPR made of devices belonging to Institute of Electronic Systems, Faculty of Electronics and Information Technologies. Presented results clearly convince about the effectiveness of proposed algorithms of soil permittivity estimation and need for further development in this radar field as well.
PKT classification
423100
KBN classification
32
EU classification
8030
Thesis file
  • File: 1
    Praca doktorska Mariusza Zycha
Request a WCAG compliant version

Uniform Resource Identifier
https://repo.pw.edu.pl/info/phd/WUTe6d71d06654545da9883eb6df8542277/
URN
urn:pw-repo:WUTe6d71d06654545da9883eb6df8542277

Confirmation
Are you sure?
Report incorrect data on this page