Parallel Implementation of the GPR Techniques for Detecting and Mapping Ancient Buildings by Using CUDA


Mumcu M. C. , Bayar S.

European Journal of Science and Technology, pp.352-359, 2020 (Refereed Journals of Other Institutions)

  • Publication Type: Article / Article
  • Publication Date: 2020
  • Doi Number: 10.31590/ejosat.780115
  • Title of Journal : European Journal of Science and Technology
  • Page Numbers: pp.352-359

Abstract

Ground-penetrating radar (GPR) is an ultra-wideband electromagnetic sensor used for the detection of objects which may be hidden behind a wall or inserted within the wall. The GPR method works on the principle of recording the reflection of electromagnetic waves sent to the underground at high speed from the interfaces by an antenna located in the horizontal direction, again by the receiver in the horizontal direction. Embedded structures are detected using collected data, computer programs, and various filters. Search for the presence of designated targets hidden between the walls, such as air pockets is help to archaeologists. In this work the Lorentz model was used for the distribution of the soil. The perfectly matched layer (PML) used for absorbing boundary conditions to simulate an open space and its expanded to match dispersive media. The finite-difference time-domain (FDTD) method is used to decompose partial differential equations for time cascading of the electromagnetic fields. FDTD calculation works very slowly. General-purpose programming can be done on the graphics processing unit (GPGPU) to solve this problem. In this work, the 3-D FDTD method was applied to the GPU by using CUDA and it was 10 times faster.