The detection of buried objects, particularly hazardous waste containers an
d unexploded ordnance (UXO), has gained significant interest in the Unites
States in the late 1990s. The desire to remediate the thousands of sites wo
rldwide has become an increasing concern and the application of radar to th
is problem has received renewed attention. The US Department of Energy's Sp
ecial Technologies Laboratory (STL), operated by Bechtel Nevada, has develo
ped several frequency-modulated, continuous-wave (FM-CW) ground penetrating
radar (GPR) units. To meet technical requirements for higher-resolution da
ta, STL and the University of California, Santa Barbara (UCSB) is investiga
ting advanced GPR hardware, signal processing, and synthetic-aperture imagi
ng with the development of an innovative system. The goal is to design and
fabricate a lightweight, battery-operated unit that does not require surfac
e contact, can be operated by a novice user, and can achieve improved resol
ution. The latter is accomplished by using synthetic-aperture imaging, whic
h forms the subsurface images by fully utilizing the data sequences collect
ively along a scan path. We also present the backward propagation algorithm
as the basic structure of the multiple-frequency tomographic imaging techn
ique, and the conventional fast Fourier transform (FFT) method which can be
described as a degenerated case of the model where the computation procedu
re is approximated under the narrow-beam assumption. (C) 2000 Elsevier Scie
nce B.V. All rights reserved.