Progress In Electromagnetics Research, PIER 69, 341359, 2007

FDTD ANALYSIS OF DIPOLE ANTENNA AS EMI

SENSOR

M. Ali and S. Sanyal

Dept. of Electronics & Electrical Communication Engineering
Indian Institute of Technology
Kharagpur-721 302, India

AbstractElectromagnetic Interference (EMI) is becoming a crucial

issue in the era of modern electronic systems. For EMI measurement, it
is required to place a sensor to receive the radiation from the equipment
in a suitable test environment. The performance of the sensor depends
on its Antenna Factor, which is the ratio of the incident electric eld on
the antenna surface to the received voltage at the load end across 50
resistance. FDTD is one of the ecient numerical techniques to solve
radiation and scattering problem in any environments. To the best of
the knowledge of the authors no literature is available where FDTD is
used to evaluate the Antenna Factor. Here, in this work we applied
FDTD to predict the performance of dipole antenna when it is used as
a EMI sensor. The results presented here for free space environment
are compared with published results.

1. INTRODUCTION

All the electronic devices must conform to the standards of

electromagnetic emission set by dierent bodies in dierent countries.
The frequency range of conducted emission standards extend from
450 KHz to 30 MHz and for that radiated emissions begins at 30 MHz
and extends to 40 GHz [1]. Compliance of the devices conforming
to the standards (limits) of interference in this range is veried by
measuring the radiated electric elds in an anechoic chamber or at
an open test range after putting the measurement antenna at a
specied distance from the device under test. Wire antennas are
widely used as transmitting antenna and as sensor for electromagnetic
interference (EMI) measurements. The term wire refers to metallic,
highly conducting wire or wire-like structures.
342 Ali and Sanyal

FDTD is one of the ecient numerical techniques to solve

Maxwells equations in any environments. To the best of authors
knowledge no appreciable work is available on this area. In this
work our attention is concentrated on the characterization of the wire
antennas as EMI sensors in terms of the Antenna Factor using FDTD
methods.
For EMI measurement it is required to determine the eld strength
at the point of measurement using a sensor. To use the sensor for this
purpose, calibration data is required relating the electric eld at the
aperture of the receiving antenna to the voltage at the 50 matched
detector. The most common performance descriptor of EMI sensors
is the Antenna Factor. The ratio of the incident electric eld on the
surface of the sensor to the received voltage at the antenna terminal
when terminated with a 50 load is known as the Antenna Factor [1].
In this work the Finite Dierence Time Domain (FDTD)
technique has been used to evaluate the current distribution on the
dipole antenna surface when it is illuminated by the plane wave and
subsequently calculate the voltage and hence Antenna Factor of the
dipole antenna. For the validation of the theory, computed Antenna
Factor of a dipole was compared with the published results. First case
a simple dipole antenna was taken and compared the FDTD computed
result with MoM results, published in result [2]. Secondly the Antenna
Factor of a ANRITSU MP651A dipole obtained using FDTD technique
was compared with the data available from the instruction manual.
Finally the Antenna Factor of a broadband dipole (i.e., dipole top and
bottom loaded with circular disc) obtained using FDTD technique was
compared with the data available in the literatures [3, 4].

1.1. FDTD Formulation of the Problem

The Finite Dierence Time Domain method(FDTD) has been widely
accepted as a reliable computational tool in numerical electromagnetic.
The explicit nature of the time-stepping algorithm to solve Maxwells
equations conveniently enables the visualization of the electromagnetic
elds inside the medium under investigation. This feature is a great
benet compared to frequency-domain methods, like the Method of
Moments (MOM) or the Finite-Element Method (FEM).
The original FDTD paradigm was described by Y ee cell, named, of
course, after Kane Yee [5]. Let E and H eld are assumed interleaved
around a cell whose origin is at the location i, j, k. Every E eld is
located 1/2 cell width from the origin in the direction of its orientation;
every H eld is oset 1/2 cell in each direction except that of its
orientation.
Progress In Electromagnetics Research, PIER 69, 2007 343

Starting with Maxwells equations [6]:



D 1

= H
(1a)
t 0 0
 

=  () E

D r (1b)
H
1

= E
(1c)
t 0 0

 
where, r () is the complex relative dielectric constant. D

and E

are
the normalized value of the corresponding eld components given by,

 0

=
D D = fn D
(2a)
0

 0

E = E = fn E
(2b)
0

fn is the normalized factor given by,


0
fn = (3)
0

The notation is dropped, but it will always be assumed that we are

referring to the normalized value. The electric eld components are

Ex = gax Dx (4a)
Ey = gay Dy (4b)
Ez = gaz Dz (4c)

where,
1
gax = (5)
t
r +
0
Expression of gay & gaz is same as for gax [6] in Eqn. (5).
Where,
0 is the free space permittivity.
0 is the free space permeability.
r is the relative permittivity of the medium.
is the conductivity of the medium.
344 Ali and Sanyal

t is the time increment.

All of the information regarding the medium is contained in
Eqn. (4). For free space, gax = gay = gaz = 1; for PEC gax =
gay = gaz = 0 and for lossy material, gax, gay, & gaz are calculated
according to Eqn. (5).
In the FDTD method we solve the Eqns. (1a) and (1c)
simultaneously using central dierence approximations for both the
temporal and spatial derivatives.
To ensure the accuracy of the computed results, the spatial incre-
ment must be small compared to the wavelength (usually /10)
or minimum dimension of the scatterer. This amounts to having 10 or
more cells per wavelength. To ensure the stability of the nite dier-
ence scheme of Eqn. (1a) and (1c) the time increment t must satisfy
the following stability condition [7, 8]
 1/2
1 1 1
umax  t + + (6)
x 2 y 2 z 2
where umax is the maximum wave phase velocity within the model.
Since we are using a cubic cell with x = y = z, the time step t
is determined by Courant Condition [9, 6] and details of Eqn. (6) is
given by
x
t (7)
n c0
where, c0 is the speed of light in free space and n is the number of space
dimensions. For practical reasons, it is best to choose the ratio of the
time increment to spatial increment as large as possible yet satisfying
Eqn. (7). Though it is not necessarily the best formula; however, we
will use it for simplicity.
1 t x/2 c0 1
= c0 = (8)
0 0 x x 2

A 15  thick unsplit Perfectly Matched Layer (PML) [6, 10, 11]

is used as Absorbing Boundary Condition (ABC) on all six sides of the
FDTD lattice. This PML is spaced 4 cells from the closest surface
of the source or scatterer.
The excitation source is set up by assigning a desired time function
to specic electric or magnetic eld components in the FDTD space
lattice. In this study, the time function Gaussian pulse is taken as a
hard source.
Progress In Electromagnetics Research, PIER 69, 2007 345

Hard source is set up by forcing a desired time function to the

Ez components in the abcd plane. For Gaussian input pulse, the Ez
component in the abcd plane at time step t is given by
 2
tt0
0.5
Ezti,j,k = Ae t
(9)

where, t is the standared deviation and relates the line width at half-
height by the relationship

t1/2 = 8 ln(2) t = 2.35482 t (10)
FDTD is a time domain technique so to get frequency domain
data we have to use Fourier transformations [6]. We iterate the FDTD
progrm until the pulse has died out, and take the Fourier transform of
the eld components in the slab. Fourier transform of the E-eld E(t)
at a frequency f1 is
tT
E (f1 ) = E (t) ej2f1 t dt. (11)
o

Notice that the lower limit of the integral is 0 because the FDTD
program assumes all causal function. The upper limit of the integral
is tT , the time at which the FDTD interation is halted. Rewriting
Eqn. (11) in the nite dierence form,

T
E (f1 ) = E (n t) ej2f1 (nt) (12)
n=0

where T is the number of iterations and t is the time step, so

tT = T t.

1.2. Total/Scattered Field Formulation

To calculate the far eld antenna factor we have to illuminate the
antenna by a plane wave. So rst we have to simulate a perfect plane
wave within the problem space. In order to simulate a plane wave in a
FDTD program, the problem space will be divided up into region, the
total eld and scattered eld [6]. Details of the method given in [6], is
used in this work.

1.3. Modeling of Dipole Antenna

A simple dipole antenna is illustrated in Fig. 1, consists of two metal
arms. A dipole antenna functions by having current run through
346 Ali and Sanyal

gaz=0 to specify metal

r
l/2

H field resulting from the

current in the dipole
y
The simulating E value is
specified at the gap.

l/2

Figure 1. A dipole antenna. The FDTD program species the arms

of the dipole by setting gaz = 0. The source is specied by setting the
Ez eld to a value at the gap.

the arms, which results in radiation. FDTD simulates a dipole in

the following way: The metal of the arms is specied by setting the
gaz parameters of Eqn. (4) to zero in the cells corresponding to the
metal accept the place where source is placed. This insures that the
corresponding Ez eld at this point remains zero, as it would if that
point were inside metal. The source is specied by setting the Ez eld
in the gap to a certain value.
For the FDTD simulation, dipole is fed at the center(x = is 
x, y = js  y, z = ks  z) gap of length z with a Gaussian pulse of
A=1.0 V maximum amplitude as in Eqn. (9). So, electric eld in the
gap of the dipole is
V (n  t)
Ez |nia ,ja ,ka +1/2 = (13)
z
The current in the wire at the feed point is obtained by applying
Amperes law [9] to the surface S with the bounding contour C on
the wire at (is , js , ks + 3/2):



E
H dL =

J dS + 0

dS (14)
C S S t
Progress In Electromagnetics Research, PIER 69, 2007 347

This gives the current:

 
n+1/2 n+1/2
I|n+1/2 = x Hx |ia ,ja 1/2,ka +3/2 Hx |ia ,ja +1/2,ka +3/2 fn
 
n+1/2 n+1/2
+ y Hy |ia +1/2,ja ,ka +3/2 Hy |ia 1/2,ja ,ka +3/2 fn (15)

Where, fn is the normalized factor, given in the Eqn. (3). From the
Lees algorithm, we can write
 
n+1/2 n+1/2
I|n+1/2 = fn  x Hx |ia ,ja 1,ka +1 Hx |ia ,ja ,ka +1
 
n+1/2 n+1/2
+fn  y Hy |ia ,ja ,ka +1 Hy |ia 1,ja ,ka +1 (16)

1.4. Antenna Factor

1.4.1. Introduction
To carry out eld strength measurements, one typically connects an
antenna to a spectrum analyzer or some other type of receiver. The

Antenna Factor (AF ) is the parameter that is used to convert the
voltage or power reading of the receiver to the eld strength incident

on the antenna [13, 14]. In terms of an equation, the (AF ) is dened
[15] as
 
Incident electric f ield
i
E

AF = (17)
Received voltage (V )

where E
i is the electric eld incident on the antenna, and V , is the
voltage induced across a 50 load at the feed point of the antenna.
Figure 2 is an illustration of the problem space of a 3-D FDTD
program. The FDTD model uses a uniform space lattice cubic Y ee
cells. A dipole (along z-axis) is illuminated by a z-directed linearly
polarized plane wave. The time domain current I(t) owing through
the center of the dipole is calculated using Eqn. (16). During the
progress of the FDTD calculations this eld Ez (t) and current I(t)
are saved for each time step. The FDTD calculation are continued
until all transients are dissipated, so that the Fourier transform yields
the steady-state frequency domain response of the antenna. Fourier
transform of the current I(t) gives frequency domain current I()
owing through the center of the dipole. Voltage developed across
50 load is V () = 50 I(). Finally antenna factor of the dipole
antenna is evaluated using Eqn. (17).
348 Ali and Sanyal

PML
kb

Transmission line
Total field

Plane wave source

Scattering from
the dipole antenna
z

ka

ja jb
y

Figure 2. Receiving antenna case. Dipole antenna works as a EMI

sensors.

1.4.2. Antenna Factor of a Dipole Antenna

The FDTD model uses a uniform space lattice cubic Y ee cells having
x = y = z = 0.8 cm and t = 0.013 ns. This ne spatial
resolution permits direct modeling of the 0.4 cm radius wire, assumed
to be PECs. A dipole of length 50 cm and radius 0.4 cm is illuminated
by a plane wave of Gaussian impulse of maximum amplitude A =
1.0 V/m given by the Eqn. (9) with t0 = 0.26666 nsec and t =
0.08 nsec. Time domain incident linearly polarized electric eld Ez (t)
is shown in the Fig. 5 and the time domain current I(t) owing through
the center of the dipole is shown in Fig. 3.
Antenna factor of the dipole antenna is evaluated using Eqn. (17).
The amplitude part of the complex antenna factor is compared with the
published result using MoM [2] shown in Fig. 4. The agreement is quite
good considering the dierent approximations and assumptions made
in the FDTD approach related to the MoM, especially in modeling the
feed region. Fig. 5 shows the phase part of the complex antenna factor
of the dipole antenna.

1.4.3. Antenna Factor of Anritsu MP651A Dipole Antenna

Schematic diagram of a Anritsu MP 651A dipole antenna is shown
in the Fig. 6. The length of the antenna is adjustable according to
the frequency of operations. The diameter of the arms of the antenna
is dierent for dierent length. So for a particular frequency mean
diameter is taken. The length and mean diameter of the arms of the
Progress In Electromagnetics Research, PIER 69, 2007 349

0.25
Dipole Current
0.2

0.15
Current (milliamperes)

0.1

0.05

-0.05

-0.1

-0.15
Length of the dipole=0.50 m.
-0.2 Radius of the dipoles=0.4 cm.
-0.25
0 5 10 15 20 25 30 35 40 45 50
Time (ns)

Figure 3. Time domain current owing through the center of the

dipole due to unit amplitude Gaussian plane wave pulse incident on it.

70
Far Field Antenna Factor
of Dipole Antenna FDTD
60 MoM
Antenna Factor (dBm-1)

50

40

30

20

10 Length of the dipole=0.50 m.

Radius of the dipoles=0.4 cm.

0
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
Frequency (GHz)

Figure 4. Comparison of the amplitude of the far eld complex

antenna factor of a dipole antenna using FDTD and MoM results [2].
350 Ali and Sanyal

240
200 Far Field Antenna Factor of Dipole Antenna

160
120
80
Phase in degree

40
0
-40
-80
-120
-160
-200 Length of the dipole=0.50 m.
Radius of the dipoles=0.4 cm.
-240
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
Frequency (GHz)

Figure 5. Phase of the far eld complex antenna factor of a dipole

antenna.

Antenna element Antenna element

Nonmetallic region

Frequency scale Shorting bar

Figure 6. An Anritsu MP651A dipole antenna.

Anritsu dipole were measured for dierent frequencies and those values
were incorporated for numerical evaluation of Antenna Factor, is in
Table 1. There are two sets of dipole arms for two dierent ranges of
frequency. One frequency from 470.0 MHz to 1.0 GHz and the other one
is from 10 GHz to 1.7 GHz. As length and mean diameter of the dipole
are changed with frequency, so frequency separate FDTD simulation is
Progress In Electromagnetics Research, PIER 69, 2007 351

done for each frequency. For dierent value of frequency, we perform

separate FDTD simulations.

Table 1. Length and mean diameter of the Anritsu MP651A dipole

antenna at various frequencies, used for evaluating antenna factor.

Frequency Length Mean diameter

(MHz) (mm.) (mm.)
470.0 311.12 2.860024
500.0 286.58 3.038269
550.0 257.74 3.266414
600.0 233.02 3.420996
700.0 201.38 3.682818
800.0 157.50 3.795473
900.0 157.50 3.918740
1000.0 140.30 4.062659
1100.0 126.78 3.422933
1200.0 116.54 3.592283
1300.0 106.78 3.796015
1400.0 100.82 3.895904
1500.0 94.42 3.985607
1600.0 88.14 4.092991
1700.0 83.82 4.181545

The FDTD model uses a uniform space lattice cubic Y ee cells

having x = y = z = diameter of the dipole corresponding
frequency. t is calculated from Eqn. (8). The free space antenna
factor of an Anritsu dipole MP 651A has been evaluated using FDTD
and compared with the data available from the chart supplied by the
manufacturer [16], is shown in Fig. 7. The agreement is quite good
considering the dierent approximations and assumptions made in the
FDTD approach relative to the manual data, especially in modeling
the feed region and nonmetallic region shown in the Fig. 6. Phase
part of far-eld complex antenna factor of an Anritsu MP651A Dipole
Antenna is shown in Fig. 8.

1.4.4. Antenna Factor of a Broadband Dipole Antenna

A broadband dipole i.e., dipole loaded with circular disc is shown in
Fig. 9. The dimensions of dierent parts of the antenna are given
below:
Length of the central part of broadband dipole antenna = 0.54 m;
352 Ali and Sanyal

40
Far Field Antenna Factor of
Anritsu MP651A dipole antenna
35

30
Antenna Factor (dBm-1)

25

20

15 FDTD
Anritsu Manual
10

0
500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700
Frequency (MHz)

Figure 7. Comparison of antenna factor of an Anritsu MP651A dipole

antenna using FDTD and data from Anritsu Manual [16].

200

160 Far Field Antenna Factor of

Anritsu MP651A dipole antenna
120

80
Phase in degree

40

-40

-80

-120

-160

-200
500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700
Frequency (kHz)

Figure 8. Phase of the far eld complex antenna factor of a Anritsu

MP651A dipole antenna.
Progress In Electromagnetics Research, PIER 69, 2007 353

17.8 cm.

Dipole Antenna

Capacitive hats 0.54 m.

Transmission line 4.488 cm.

Figure 9. Broadband dipole antenna.

Radius of the central part of the dipole = 2.244 cm;

Radius of the capacitive hats = 8.9 cm;
The FDTD model uses a uniform space lattice cubic Y ee cells
having x = y = z = 4.488 cm and t = 0.0748 ns. This ne
spatial resolution permits direct modeling of the 2.244 cm radius wire,
assumed to be PECs. The dipole is illuminated by a plane wave of
Gaussian impulse of maximum amplitude A = 1.0 V/m given by the
Eqn. (9) with t0 = 1.496 nsec and t = 0.44880 nsec. Time varying
current owing through the center of the broadband dipole is shown
in the Fig. 10. Amplitude of the complex antenna factor of broadband
dipole obtained using FDTD has been compared with experimental,
MININEC (MoM based commercial software) simulation [3] and MoM
based numerical [4, 2] results in Fig. 11. Phase of the complex antenna
factor of broadband dipole antenna is shown in the Fig. 12. Considering
the dierences in how the feed region is modeled the agreement is quite
good. FDTD predicted antenna factor is much closer to the experiment
result, than the other available in the literature [24]. Again within
the region, near the frequency 0.1 GHz, FDTD predicted result shows
much better agreement with the measurement than the result from
MININEC [3] also.
354 Ali and Sanyal

0.8
Broadband Dipole Current
0.6

0.4
Current (milliamperes)

0.2

-0.2

-0.4

-0.6

-0.8
0 16 32 48 64 80 96 112 128 144 160
Time (ns)

Figure 10. Time domain current owing through the center of the
broadband dipole due to unit amplitude Gaussian plane wave pulse
incident on it.

140
130 Far Field Antenna Factor of Broadband Dipole
120 MOM [2,4]
110 MININEC (MoM) [3]
Measurement [3]
Antenna Factor (dBm-1)

100 FDTD
90
80
70
60
50
40 Length of the dipole=0.54 m.
30
Radius of the dipole=2.244 cm.
20
10 Radius of the capacitive hats=8.9 cm.
0
1 10 100 1000 10000 100000 1e+06
Frequency (kHz)

Figure 11. Comparison of the amplitude of the far eld

complex antenna factor of a broadband dipole using FDTD with
published Measurement, MININEC (MoM based commercial software)
simulation [3] and MOM based numerical [2, 4] results.
Progress In Electromagnetics Research, PIER 69, 2007 355

200
Far Field Antenna Factor of Broadband Dipole
160

120

80
Phase in degree

40

-40

-80
Length of the dipole=0.54 m.
-120
Radius of the dipole=2.244 cm.
-160
Radius of the capacitive hats=8.9 cm.
-200
1 10 100 1000 10000 100000 1e+06
Frequency (kHz)

Figure 12. Phase of the far eld complex antenna factor of a

broadband dipole.

2. DISCUSSION

The FDTD based simulation approach has several advantages over

conventional measurement based methods for determining antenna
factor.
Antennas can be illuminated with a plane-electromagnetic wave.
Time domain current and hence voltage across 50 load can be
calculated.
Single simulation gives all the frequency components of the results.
Illumination by a single electromagnetic wave in the direction of
bore sight is possible.
Interactions between antennas and ground planes may be
eliminated.
In order to have condence in any computer model, it must be validated
against measurement, but measurements have errors associated with
them. It is very dicult to measure plane-wave antenna factor
accurately. In the conclusion, it does appear that the manufacturer
provided data of antenna factor of the Anritsu MP651A dipole antenna
has a signicant dierence with FDTD simulated results. This is
because in the provided data they are not able to eliminate the ground
plane eect and also the proper plane wave is not possible to produce in
356 Ali and Sanyal

the measurement. Moreover in the FDTD simulation approach can get

both the amplitude and phase from a single simulations where as the
manufacturer can provid only the amplitude of the complex Antenna
Factor.
FDTD prediction of the antenna factor of EMI sensors is a very
attractive alternative if one takes into consideration the enormous
expenditure and time required for calibrating a sensor experimentally.
Also, for experimental calibration, each and every sensor is to be
calibrated individually, whereas for theoretical calibration all the
sensors constituting a particular type can be calibrated at one go
using the same approach; it is possible to predict the susceptibility of
such antennas to electromagnetic radiation incident from any direction.
Being time-domain technique, FDTD directly calculates the impulse
response of an electromagnetic system. Therefore, a single FDTD
simulation can provide either ultra wideband temporal waveforms
or the sinusoidal steady state response at any frequency within the
excitation spectrum. With FDTD, specifying a new structure to be
modeled is reduced to a problem of mesh generation rather than
the potentially complex reformulation of an integral equation. For
example, FDTD requires no calculation of structure-dependent Green
functions. The sources of error in FDTD calculations are well
understood, and can be bounded to permit accurate models for very
large variety of electromagnetic wave interaction problems. We can
easily calculate the near eld corrections on far eld antenna factor
using FDTD method.

3. CONCLUSION

According to the authors knowledge this is the rst work where the
FDTD technique is used to predict the performance of dipole antenna
when it works in receiving mode. For that, rstly, a perfectly plane
wave is generated within the FDTD problem space. Then the voltage
developed across 50 load end of the antenna is calculated due to plane
wave incident on the antenna. Finally amplitude and phase parts of
the complex antenna factor is calculated. Three cases of antenna are
considered for the validations of the theory.
Firstly, the antenna factor of a simple dipole is computed. The
result is compared with the MoM based numerical [2] results. The
agreement is quite good considering the dierent approximations and
assumptions made in the FDTD approach related to the MoM. The
FDTD computed phase information of the antenna factor is shown in
Fig. 5.
Secondly, the antenna factor of a Anritsu MP651A dipole antenna
Progress In Electromagnetics Research, PIER 69, 2007 357

is computed. The result is compared with the data available from

the chart supplied by the manufacturer [16]. Again the agreement is
quite good considering the dierent approximations and assumptions
made in the FDTD approach relative to the manual data, especially
in modeling the feed region and nonmetallic region of the Anritsu
MP651A dipole antenna system. No phase informations of the complex
antenna factor of the Anritsu MP651A dipole antenna is available in
the chart supplied by the manufacturer [16]. The FDTD computed
phase information of the complex antenna factor of the Anritsu
MP651A dipole antenna is shown in the Fig. 8.
Thirdly, the antenna factor of a broadband dipole i.e., dipole
loaded with circular disc is computed. Amplitude of the complex
antenna factor of broadband dipole obtained using FDTD has been
compared with experimental, MININEC (MoM based commercial
software) simulation [3] and MoM based numerical [2, 4] results.
Considering the dierences in how the feed region is modeled the
agreement is quite good. FDTD predicted antenna factor is much
closer to the experimental result, than the other available data
in the literature. Again within the region, near the frequency
0.1 GHz, FDTD predicted result shows much better agreement with the
measurement than the result from MININEC. The FDTD computed
phase information of the complex antenna factor of the broadband
dipole antenna is shown in the Fig. 12.
To conclude, we can say that FDTD is very ecient, accurate
and easy technique to compute the antenna factor of a dipole antenna.
FDTD technique can apply to evaluate error correction factor for the
near eld case very eciently, accurately and easily. This technique
may be extended to determine the antenna factor of other types
of antennas or to study the time domain characteristics of any
trans-receive Antenna System [17] when it is other than the free
space environment such as for Gigahertz Transverse Electromagnetic
(GTEM) Cell of characteristic impedance 50 ohms.

ACKNOWLEDGMENT

The authors would like to acknowledge Dr. S. Ghosh and

Prof. A. Chakrabarty for the theoretical discussion time to time.
358 Ali and Sanyal

REFERENCES

1. Paul, C. R., Introduction to Electromagnetic Compatibility, Wiley

Series in Microwave and Optical Engineering, Wiley, New York,
1992.
2. Ghosh, S., Some studies on loaded and unloaded wire
antennas as EMI sensors, Ph.D. dissertation, Indian Institute
of Technology, Kharagpur, Aug. 2003. [Online]. Available:
http://www.ecdept.iitkgp.ernet.in/.
3. Wang, K. and R. Nelson, Numerical simulation of the antenna
factor of a broadband dipole antenna, Proc. IEEE International
Symposium on Electromagnetic Compatibility, Vol. 1, 616619,
Aug. 1317, 2001.
4. Ghosh, S., A. Chakrabarty, and S. Sanyal, Loaded wire antenna
as EMI sensor, Progress In Electromagnetics Research, PIER 54,
1936, 2005.
5. Yee, K. S., Numerical solution of initial boundary value problems
involving maxwells equations in isotropic media, IEEE Trans.
Antennas Propagat., Vol. 14, No. 3, 302307, May 1966.
6. Sullivan, D. M., Electromagnetic Simulation Using The FDTD
Method, IEEE Press, New York, 2000.
7. Taove, A. and M. Brodwin, Numerical solution of steady-state
electromagnetic scattering problems using the time-dependent
maxwells equations, IEEE Trans. Microwave Theory Tech.,
Vol. MTT-23, No. 8, 623630, Aug. 1975.
8. Sadiku, M., V. Bemmel, and S. Agbo, Stability criteria for nite-
dierence time-domain algorithm, Proc. IEEE, Southeastcon,
4850, April 1990.
9. Taove, A. and S. C. Hagness, Computational Electromagnetic
The Finite-Dierence Time-Domain Method, Artech House,
Boston, London, 2005.
10. Sullivan, D. M., A simplied PML for use with the FDTD
method, IEEE Microwave Wireless Compon. Lett., Vol. 6, No. 2,
97, Feb. 1996.
11. Sullivan, D. M., An unsplit step 3-d PML for use with the FDTD
method, IEEE Microwave Wireless Compon. Lett., Vol. 7, No. 2,
184186, July 1997.
12. Yu, W. and R. Mittra, Conformal Finite-Dierence Time-
Domain Maxwells Equations Solver: Software and Users Guide,
Artech House, Boston, London, 2004.
13. Das, S. and A. Chakrabarty, A novel modeling technique to solve
Progress In Electromagnetics Research, PIER 69, 2007 359

a class of rectangular waveguide based circuits and radiators,

Progress In Electromagnetics Research, PIER 61, 231252, 2006.
14. Bhattacharya, A., S. Gupta, and A. Chakraborty, Analysis
of rectangular waveguide and thick windows as EMI sensors,
Progress In Electromagnetics Research, PIER 22, 231258, 1999.
15. Ghosh, S. and A. Chakrabarty, Performance analysis of EMI
sensor in dierent test sites with dierent wave impedances,
Progress In Electromagnetics Research, PIER 62, 127142, 2006.
16. Instruction Manual, Dipole Antenna MP651A/B, Anritsu Corp,
Japan.
17. Ghosh, S. and A. Chakrabarty, Estimation of equivalent circuit of
loaded trans-receive antenna system and its time domain studies,
Journal of Electromagnetic Waves and Applications, Vol. 20, 89
103, 2006.