UNIVERSITY
OF TRENTO
DIPARTIMENTO DI INGEGNERIA E SCIENZA DELL’INFORMAZIONE
38050 Povo – Trento (Italy), Via Sommarive 14
http://www.disi.unitn.it
SVM-
BASED
C
LASSIFICATION
A
PPROACH FOR
S
YNTHETIC
-I
MPULSE
M
ICROWAVE
I
MAGING
–
DATA FORMAT
F. Viani, M. Benedetti, M. Donelli, L. Lizzi, A. Cresp, I. Aliferis,
C. Pichot, and A. Massa
August 2008
InformationandCommuni ation Te hnology Dept. Universityof Trento
ViaSommarive14,38050Trento,ITALY Phone+390461882057Fax +390461882093
E-mail: andrea.massaing.unitn.it
Contra t No. DIT-PRJ-08-005
Te hni al Report N. 2
SVM-based Classi ation Approa h for Syntheti -Impulse Mi rowave Imaging
Data Format F.Viani
(1)
,M.Benedetti(1)
,M.Donelli(1)
,L.Lizzi(1)
, A.Cresp(2)
,I.Aliferis(2)
,C.Pi hot(2)
, andA.Massa(1)
(1)
DepartmentofInformationEngineering andComputerS ien e ELEDIAResear hGroup
UniversityofTrento,ViaSommarive14,38050Trento-Italy Tel.+39 0461882057,Fax. +390461882093
E-mail: andrea.massaing.unitn.it
(2)
Laboratoired'Ele tronique,AntennesetTélé ommuni ations(LEAT) UniversitédeNi e-SophiaAntipolis
250rueAlbertEinstein,06560Valbonne
Tél: +33(0)492942802,Fax: +33(0)492942812
1. Introdu tion 3
2. A quired Data Format 3
The design of an ultra-wideband (UWB) surfa e penetrating radar requires a good understanding of the
medium properties and of the ele tromagneti propagation in ground-penetratingappli ations, whi h an
be transposed in the knowledge of the environment ontribution added up to the transmitted signal. In
ordertoremovethe ontributionoftheenvironment,thedata olle tedwiththeSIMISradararedierential
measurements [1℄. The overall on ept of this te hnique is to determine areferen e measurement, of the
environmentonly,that anberemovedbysubtra tiontoenhan etheintera tionsoftheobje tsand al ulate
thes atteredeld omingfromthetargetsonly. Themeasurement anbeperformedeitherinthefrequen y
orthetimedomainwhi hmostlywillprovidethesameinformation. Inthis ongurationthemeasurements
are made in the frequen y domain and onsequently time domain signals are obtained by an inverse fast
fouriertransform[2℄.
2 A quired Data Format
Inthefrequen ydomain, ea hfrequen y omponentissentandmeasuredoveranarrowbandwidthwitha
onstantfrequen ystepof
15 M Hz
fromf
min
= 2 GHz
tof
max
= 8.0 GHz
(foratotalnumberoffrequen iesF
tot
= 401
). Forea hstep,theadoptedve tornetworkanalyzer(VNAHP8720B) olle tsinatextleready tobepostpro essedtherealandimaginarypartsofalltheree tionS
mm
(f ), m = 1, ..., M
andtransmissionS
nm
(f ), n = 1, ..., N ; m = 1, ..., M ; n 6= m
parameters, whereN
andM
are the number of re eivers and transmitters, respe tively. Also a Rohde & S hwarzVNA is available to olle t the samedata, the maindieren es onsist in aslightly widerfrequen y bandwidth (
f
min
= 1.5 GHz
andf
max
= 8.0 GHz
)and a highernumberoftotalfrequen ypointsF
tot
= 2001
. Thewholefrequen ysweepofameasurementisnally storedinathree-dimensionalstru tureasshowninFig. 1. Everygrayplanerepresentsallthe ombinationsbetweenthe
m
-th transmitter andn
-th re eiver(in this ongurationM
= N
)for thedata apture of all theS
-parameters, while the pink pattern representsthe frequen y sweep of oneS
-parameter required by the inverse fouriertransform pro edure to obtaintheUWB time-domain signal. As it anbeseen,on thediagonal of ea h gray plane are stored all the ree tion parameters
S
mm
(f ); m = 1, ..., M
. Alternatively, whenanextraantennaisadoptedastransmitterandthearrayelementsa tasre eivers,onlythetransmissionparameters
S
n
1
(f ); n = 1, ..., N
are measuredandstored asin Tab. I.In thisparti ular onguration,the transmitterindex isxedtom
= 1
whilethere eiverindexn
= 1, ..., N
s ansallthearrayelements.[1℄ A. Cresp,I.Aliferis, M. J.Yedlin, C. Pi hot, and J.-Y. Dauvigna , Investigation oftime-reversal
pro- essing for surfa e-penetrating radar dete tion in a multiple-target onguration, submitted to 2008
EuropeanRadarConferen e (EuRAD08),30-31O tober,Amsterdam,2008.
[2℄ V. Chatelee, A. Dubois, I. Aliferis, J.-Y. Dauvigna , C. Pi hot, M. J. Yedlin, Real data mi rowave
imaging and time reversal, IEEE Antennas Propagat. So . Int. Symp., Homolulu, Hawaii, USA, pp.
RX
N
f
min
RX
1
T X
1
S
11
(f
min
)
f
max
S
22
(f
min
)
S
33
(f
min
)
S
M M
(f
min
)
S
N 1
(f
min
)
T X
M
f
n
= 1
n
= 2
· · ·
n
= N
f
min
ℜ
(S
11
(f
min
)) , ℑ (S
11
(f
min
))
ℜ
(S
21
(f
min
)) , ℑ (S
21
(f
min
))
· · ·
ℜ
(S
N
1
(f
min
)) , ℑ (S
N
1
(f
min
))
. . . . . . . . . . . .