Utilizing the Vector Network Analyzer

To Work as a GPR for Detection of Hidden Objects

By
Omar S. Elshebani
R & D center
oss5_1_73@yahoo.com

K. S. Belgath
S. I. Elkhetali
R & D center
The Higher Institute of Electronics
khbelgath@hotmail.com
elkhetali@yahoo.com

Abstract:
Detection and subsequent imaging and classification of hidden objects can be
done by the radar which transmits pulses and receives the reflections from the
objects or targets. The radar that directs its energy to detect targets hidden under
the surface of the ground or behind an obstacle such as a wall is called ground
penetrating radar (GPR). There are many types of GPRs that work according to
different theories. These are the pulse type, impulse, frequency modulated
continues wave (FMCW) and synthetic pulse or stepped frequency ground
penetrating radars. However, detection and classification of hidden targets is a
very difficult task because they depend on the nature of the target and the
environment surrounding it.
An experiment would involve variation the parameters of the target and those of
the radar. The former is easy but the latter is difficult because one needs a radar,
which is not available, and even that may not be adequate enough because a
commercial radars parameters may not be adjustable to allow for
experimentation.
In this paper we utilize general-purpose laboratory equipment, namely, a vector
network analyzer to operate it as a GPR. It has a wide operating frequency range
and other controllable parameters. Vector network analyzers measure the
magnitude and phase characteristics, these can be utilized to measure target
parameters in the frequency domain and then they may be displayed in the time
domain by application of the Inverse Fourier Transformation.
In this work many types of metallic and dielectric objects are placed within
different types of soil and behind walls made-up of bricks and attempts are made
to detect the targets.

PDF created with pdfFactory trial version www.pdffactory.com

1. Introduction
Electromagnetic energy is highly attenuated when propagating in the soil. The
higher the frequency the greater is the attenuation. Nevertheless it is possible to
employ radar to detect underground scattering objects if the distances are small.
Radars for probing the ground beneath the surface are called Ground-Probing
Radars '' GPR '' they have been used in a variety of applications including
detection of buried gas pipes, water pipes, landmines, measurement of
permeability and conductivity of coal, mapping of shallow geologic features and
applications in groundwater.
Ground-Probing Radars are three types:
1. Impulse radar ( VHF and UHF )
2. FM-CW radar ( Microwave frequencies )
3. Stepped-frequency radar ( VHF and UHF )
The impulse radar radiates a short pulse covering a wide frequency band. The
FMCW radar sweeps the frequency between two limits and measures the
distance depending on the difference in the frequency of the reflected signal and
the frequency being transmitted. On the other hand the stepped-frequency radar
transmits sequential pulses , increasing the frequency by a constant value step
each subsequent pulse to cover the desired bandwidth. The range is obtained by
measuring the phase of each step and then performing the inverse Fourier
transform.
The basic principle in object or target detection is that the radar pulse reflects
when strikes a boundary where there is an abrupt change in dielectric constant
the range is determined by measuring the time delay between the transmitted
and the received signal using the knowledge of the propagation velocity.
2R
=
(1)
c
where is the delayed time, R is the range of the target and c is the propagation
velocity.
The range resolution of closely spaced objects depends only on the signal
bandwidth and is independent of the transmitted waveform, range and
frequency. The equation that determines the limit of range resolution is:
c .
c
=
(2)
2
2. B
where B is the system bandwidth. Also because the stepped-frequency transmits
pulses in the frequency step f Hz, objects that extend beyond 1/f will cause
ambiguous range responses. Mathematically, the unambiguous range can be
derived as follows: The phase of a received signal due to a scatterer at range R is
= 4. .R = 4. . f .R
(3)

c
The signal is totally described by its phase and amplitude or by its real and
imaginary components:
R =

PDF created with pdfFactory trial version www.pdffactory.com

Re = A cos(4. . f .R / c)

Im = A sin(4. . f .R / c)

(4)
The maximum alias-free range corresponding to a given frequency step size is
obtained by solving (3) for a phase difference of 2 radians:
= 2. = 4. .RU .f / c since RU = c / 2.f
(5)
RU is the unambiguous range objects exceeding this value cause a folding of the
true range into the unambiguous extent.
2. The Vector network Analyzer as a substitute for GPR
The vector network analyzer (VNA) can transmit signals over a wide range of
frequencies to cover a specific bandwidth. The frequency step size can be
changed through the available control features of the vector network analyzer.
In this research we are planning to verify the ability of the VNA to work as a
GPR by connecting two antennas to the VNA ports to detect the position of the
buried objects.
In the simplest case the transmitted pulses will be generated with a specified
frequency band which is proportional to the desired resolution. The transmitted
pulse is expressed by

x (t ) =

cos( 2. . f .t )

(6)

Where the quantity ( 2. . f .t ) express the temporal variations due to the

transmitted signal, the received pulses are shifted with a phase angle relative
to the transmitted pulses
= 2. .l =

4.
.R

(7)

Where R is the distance to boundary where there is an abrupt change in

dielectric constant in the propagation medium, thus the received pulses are
expressed by

y (t ) = cos( 2 . . f .t )

(8)

Worth mentioning that the degree of desired target discrimination in this paper is
detection and location of targets also we will attempt to estimate the dielectric
constant of the propagation media through the knowledge of traveled distance
and time delay of the received signal.
3. Measurements and Results:
The vector network analyzer and the two horn antennas shown in figure (1) are
used to measure the phase of the transmission coefficient ( S21 ) of different
objects placed at certain distance from the antennas.
The following arrangements are considered in the measurements:
1. To have the required range resolution which 7.5 cm, a bandwidth of 2GHz
or higher is used.

PDF created with pdfFactory trial version www.pdffactory.com

2. To satisfy the minimum required bandwidth of 2 GHz and a frequency step

size of 40 MHz the number of points used by the VNA is 51 points, which
permits an unambiguous range of 375 cm.
3. Using the gating property of the VNA in some results to cancel the
directly leaked signals between the antennas.
4. The frequency range is from 4 GHz to 6 GHz.

Fig (1) Equipment setup: two horn antennas connecting to

the VNA ports
Using Mathcad software to calculate the Inverse Fourier Transform and
providing the phase of the transmission coefficients ( S21 ) to obtain the signal in
the time domain. This is done for different materials such as bricks, sand and
clay with reflectors and without. From time domain results of reflections the
range of the target is estimated.
The following experimental cases give details of the materials used and their
arrangements. The figures below show the time domain signals after the
application of the IFT in nano seconds.
Case 1: The two antennas are placed face to face without any targets, the results
in figure (2-a) and figure (2-b) show the insertion phase and the time delay due
to the presence of the antennas. From figure (2-b) the time delay is ~1.25 nsec.
This equivalent to antenna length of 18.75 cm.

Fig (2) the insertion phase response and the time domain signal
case 1

PDF created with pdfFactory trial version www.pdffactory.com

Case 2: The metallic reflector is placed behind a wall of single bricks having a
width of about 15 cm which is at a distance of 32 cm from the antennas. The
separation distance between the two antennas is 17 cm. Figure (3) shows the
leakage signal at 1.56 nsec, reflections from reflector placed behind the bricks
being at 5.6 nsec. The reflection at 3.1 nsec is from the front side of the wall.
60
45
y( t )

30
15
0
0

1.25

2.5

3.75

6.25

7.5

8.75

10

t
time ( ns )

Fig (3) time domain signal, case 1

Case 3: The target is double-brick target, it is made of two adjacent bricks hence
the width is ~30 cm. Its distance from the antennas is 30 cm. The antennas are
placed very close to each other. Figure (4) shows the reflections from the bricks
at ~3.4nsec emanating from the front face of the wall and at 5.5 nsec from the
gap between the two bricks. The reflection from the back of the wall does not
appear. The approximate space distance of the blocks from the antennas is
~32.25 cm. There is no metallic reflector.

y( t )

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10
t
time ( ns )

Fig (4) time domain signal, case 3

Case 4: A metallic reflector is placed at a distance of 16 cm behind a single
brick wall situated a distance of 30 cm from the antennas. The leakage between
the antennas has been gated out. Figure (5) shows the reflection at 3.25 nsec is
from the face of the wall and the reflection at 7 nsec is from the metallic
reflector.

PDF created with pdfFactory trial version www.pdffactory.com

40
30
y( t )

20
10
0
0

0.5

1.5

2.5

3.5

4.5

5.5

6.5

7.5

8.5

9.5

10

t
time ( ns )

Fig (5) time domain signal, case 4

Case 5: A metallic reflector is placed at the end of the sand container of depth
18 cm, the height of the antennas is 20 cm from the upper surface of the sand.
Figure (6) shows the reflections at 1.3 nsec being the leakage and reflection
from the surface of the sand is at 2.6 nsec. The approximate space distance from
the antennas~20 cm. The next reflection is at 4.3 nsec is from the metallic
reflector which shows that the signal traveled a distance of ~25.75 cm to the
position of the reflector.

y( t )

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5

7 7.5 8 8.5 9 9.5 10

t
time ( ns )

Fig (6) time domain signal, case 5

Case 6: A metallic reflector is placed a depth of 9 cm from the sand surface, the
height of the antennas is 24 cm from the upper surface of the sand. Figure (7)
shows the reflections from the upper surface of the sand at 2.8 nsec
corresponding to approximately 24 cm, and the other reflection is at 3.75nsec
which is from the metallic reflector. In this case the bandwidth has been
increased to 3 GHz.

y( t )

0.5

1.5

2.5

3.5

4.5

5.5

6.5

7.5

t
time ( ns )

Fig (7) time domain signa, case 6

PDF created with pdfFactory trial version www.pdffactory.com

8.5

9.5 10

Case 7: A metallic reflector is placed at a depth of 7 cm in clay material, the

antennas are at a height of 17 cm from the upper surface of the clay material, the
result is shown in figure (8). Another experiment is done using a reflector at a
depth of 10 cm from the surface of clay shown in figure (9) and the antennas are
at a height of 15 cm from the clay surface. Even though the bandwidth was
increased to 3.8 GHz but there were no apparent clearly distinguishable
reflections from the metallic reflector.

y( t )

10

t
time (ns)

Fig (8) time domain signal, case 7

y( t )

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10
t
time ( ns )

Fig (9) time domain signal, case 7

4. Conclusions and future work
From this work we can draw the following conclusions:
The vector network analyzer can be used as a stepped frequency ground
probing radar
The gating property of this network analyzer is a very convenient feature
because it removed the directly leaked signal between the antennas which is
big problem in detection of small and close targets to the antennas such as
landmines. They can easily masked by the leakage signal.
The target buried in sand can be detected without difficulty.
Clay has high attenuation property, especially when wet. We expect that it
has a high moisture content since it was collect few days after the rain. In
our case we were not able to estimate the moisture content of the clay nor
the other materials.

PDF created with pdfFactory trial version www.pdffactory.com

Further work can be done by including more natural and industrial materials
with controlled moisture content. It is not too difficult to estimate the relative
permittivity and conductivity of the materials from the experiments.
5. References
1. HP vector network analyzer manual
2. Garrara W G, R S Goodman and R M Majewski, Spotlight synthetic
Aperture Radar, Artech House, 1995.
3. Wahner D R, High resolution Radar, Artech House, 1994.
4. Young J D, R Caldecott and L Peters jr, Underground Radar Research at
Ohio State University, IEEE AP-S News Letter, Aug. 1979.
5. Skolnik M I, Radar Applications, IEEE Proceedings 1988, part 6, remote
sensing of the environment.

PDF created with pdfFactory trial version www.pdffactory.com