Premium
EM target location 1
Author(s) -
Gelius L.J.
Publication year - 1996
Publication title -
geophysical prospecting
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.735
H-Index - 79
eISSN - 1365-2478
pISSN - 0016-8025
DOI - 10.1111/j.1365-2478.1996.tb00160.x
Subject(s) - radar , generalization , range (aeronautics) , point (geometry) , data acquisition , estimator , algorithm , radar imaging , plane (geometry) , object (grammar) , synthetic aperture radar , diffraction , computer science , geometry , geology , computer vision , optics , mathematics , artificial intelligence , physics , mathematical analysis , statistics , engineering , telecommunications , aerospace engineering , operating system
We consider the problem of computing the most probable location of a target based on radar measurements of the subsurface. Our algorithm makes use of the maximum likelihood estimator (MLE), which represents a correlation between the measured data and synthetic data generated for the object of interest at different locations. Previous studies assume a plane‐wave acquisition geometry and target object(s) embedded in a uniform background. In this paper, a generalization of the MLE method is presented which is valid for discrete point sources (and receivers) and a 2D model (i.e. a 2.5D acquisition geometry). Within this formulation the treatment of a non‐uniform background model is also possible. We concentrate on geotechnical ground investigations and assume that the characteristic dimensions of the target object are in the range 1–2λ, (λ being the wavelength). The potential of the method is demonstrated employing cross‐hole radar data acquired in a controlled field experiment. The MLE result is also compared with the image obtained employing a full reconstruction method such as diffraction tomography.