Parametric study on the characteristics of planar

inverted-F antenna
H.-M. Chen, Y.-F. Lin, P.-S. Cheng, H.-H. Lin, C.T.P. Song and P.S. Hall
Abstract: The design and experimental demonstrations of a planar inverted-F antenna (PIFA) are
presented. The physical parameters of the antenna, including shorting plate width, antenna input
impedance, antenna thickness and ground plane dimensions, are studied. The inuence of various
parameters on antenna characteristics has been investigated using simulation software IE3D and
experimental results. Comparisons with measured results on fabricated antenna structures are
provided for simulation validation. Simple semi-empirical formulas of the operational frequency
and input impedance on the antenna have also been obtained. These results are very useful in the
design of PIFA.

Introduction

Compact antennas have been the centre of much research

interest because of the rapid progress in wireless communications. The low prole, light weight and low fabrication
cost make microstrip antennas very attractive for wireless
communications. The size of the microstrip antenna,
however, may cause inconveniences for wireless applications
at the lower microwave frequencies. To reduce the microstrip antenna size, a potential candidate is the planar
inverted-F antenna (PIFA), which is a modied form of the
microstrip antenna. The PIFA is a quarter-wavelength
shorted patch, which consists of a nite ground plane, a top
radiator, a coaxial probe and a shorting mechanism that
shorts the top radiator to the ground plane. There are
several shorting mechanisms, such as the use of a shorting
pin [14], a shorting plate [58] and a shorting plate together
with a shorting pin [9, 10].
This paper proposes a planar inverted-F antenna, where
an edge of the radiator is terminated with a shorted plate,
and fed by a coaxial probe along the centreline of the
radiator. This work includes parametric studies of the PIFA
behaviour, with variations on the shorting plate width, the
antenna height, the input impedance on the radiator and the
ground plane size. The effects of these design parameters of
the PIFA on its resonant frequencys impedance bandwidth
and radiation patterns are presented and discussed.

r IEE, 2005
IEE Proceedings online no. 20045158
doi:10.1049/ip-map:20045158
Paper rst received 19th April 2004 and in revised form 14th March 2005
H.-M. Chen and Y.-F. Lin are with the Institute of Photonics and
Communications, National Kaohsiung University of Applied Sciences,
Kaohsiung 807, Taiwan, Republic of China
P.-S. Cheng and H.-H. Lin are with Department of Electronic Engineering,
National Kaohsiung University of Applied Sciences, Kaohsiung 807, Taiwan,
Republic of China
C.T.P. Song is with Sendo International Limited, Hatchford Way, Birmingham
B26 3RZ, UK
P.S. Hall is with Department of Electronic Electrical and Computer
Engineering, School of Engineering, University of Birmingham, Edgbaston
Park Road, Birmingham B15 2TT, UK
E-mail: hmchen@cc.kuas.edu.tw

534

Antenna configuration

The conguration of the proposed PIFA is shown in Fig. 1.

The rectangular radiating patch, made of 0.2 mm-thick
copper plate, has dimensions L
W; and is located in
the middle of a 0.2 mm-thick copper ground plane with
dimensions Lg
Wg. The antenna height h is lled with an
air substrate (er 1.0). The shorting plate consists of a
vertical conducting strip with dimensions Ws
h, and it is
used not only to connect between the patch and ground, but
also to support the whole antenna. The 50 O coaxial probe
has a radius of 0.64 mm and is fed in the centreline of the
rectangular patch. The distance between the feeding
position and the shorting plate is d. The coaxial feed excites
the PIFAs TM10 mode.
3

Experimental results and discussions

A typical PIFA antenna is implemented and studied. In the

following studies, the size of the radiating patch is selected
to be L 30 mm and W 40 mm. The ground plane also
has the same dimensions with Lg 30 mm and Wg 40 mm.
For a xed antenna height of h 4 mm, the measured
return loss for different shorting plate widths is shown in
Fig. 2 and listed in Table 1 for comparison. As can be seen
in Fig. 2a, when the shorting plate width is larger than half
the width of the patch (Ws4W/2), the excited resonant
frequency of the patch is similar to the TM10 mode of the
quarter-wave patch. For a smaller shorting plate width
(WsoW/2), the proposed patch resonates at higher
frequency being similar to the TM10 mode of halfwavelength patch (see Fig. 2b). Increasing the width of the
shorting plate increases the resonant frequency as shown in
Fig. 2a.
Figure 3 shows the experimental and simulation results
for Ws 20 mm, 25 mm and 40 mm. The simulation
software IE3D was used for optimising the design
parameters. The shift in simulated resonant frequency
fc is less than 1.5% in most cases when compared with
measurements. There is good agreement between measured
and simulated results. This gives condence in using the
simulation software to examine the various antenna
parameters that inuence various performances. Surface
current densities can also be simulated and analysed readily.
IEE Proc.-Microw. Antennas Propag., Vol. 152, No. 6, December 2005

y
feed point

Table 1: Performances of the PIFA with different width (Ws)

rectangular
radiating patch

Shorting plate
width Ws (mm)

Feed position
d (mm)

Resonant
frequency
fc (MHz)

Bandwidth
(%)

7.5

4610

4.56

5.5

4775

4.82

10

7.0

4867

4.01

15

6.5

4905

2.65

20

1.5

1805

2.22

25

3.5

1946

2.26

30

5.5

2106

2.47

35

6.0

2281

3.37

40

5.5

2296

2.48

shorting
plate
WS

Wg

air substrate
ground plane
Lg

Fig. 1 Geometry of a single-feed rectangular microstrip antenna

with a shorting plate

Impedance bandwidth is determined by 10 dB return loss and fc is

the centre frequency. Antenna parameters are given in Fig. 2
0
5

0
5

15
20
25
30
35
1600

10
return loss, dB

return loss, dB

10

WS = 20 mm
WS = 25 mm
WS = 30 mm
WS = 35 mm
WS = 40 mm

15
20

W = 20mm(measured)
S

25

W = 20mm(simulated)
S

W = 25mm(measured)
S

W = 25mm(simulated)

1800

2000
2200
frequency, MHz
a

30

2400

W = 40mm(measured)
S

W = 40mm(simulated)

35
1600

1800

2000
2200
frequency, MHz

2400

Fig. 3 Measured and simulated return loss for the proposed

antenna with different shorting plate widths (Ws)

return loss, dB

Other antenna parameters are the same as in Fig. 2

10
15
20

WS = 0 mm

25

WS = 10 mm

30
4000

WS = 5 mm
WS = 15 mm
4400

4800
frequency, MHz
b

5200

5600

Fig. 2 Measured return loss for the proposed antenna with

different shorting plate widths (Ws)
er 1.0, h 4 mm, L Lg 30 mm, W Wg 40 mm

The simulated current distributions of the proposed

resonant frequencies for Ws 0 (without shorting plate),
10, 25 and 40 mm are plotted in Figs. 4ad, respectively. For
the case in Fig. 4b, the excited current distribution is very
similar to that of the TM10 mode of the half-wavelength
patch (see Fig. 4a), and is affected slightly by the shorting
plate width (Ws). As for the case in Fig. 4c, the excited
current distribution is also similar to that of the TM10 mode
of the quarter-wave patch (see Fig. 4d), and the current
distribution of the TM10 mode is strongly perturbed owing
IEE Proc.-Microw. Antennas Propag., Vol. 152, No. 6, December 2005

to the width of the shorting plate. The maximum of the

current distribution can be observed close to the shorted
edge owing to the larger shorting plate width. The currents
nd a resonant condition for the quarter-wave patch, and
the length of the current-line path decreases with increasing
shorting plate width. This is because the resonant frequency
increases as the shorting plate width Ws increases. To design
the resonant frequency, simple semi-empirical formulas
have been found very useful. An approximate equation was
derived by tting the experimental data (for Ws  20 mm):
c
fc
1
4L DL k1 DW 2 k2 DW
DW W  Ws ; k1 0:0188;
k2 0:0767; DL 2:741

where c is the speed of light in free space, DL is the

equivalent radiation slot length, similar to that of standard
rectangular patch in [11], k1, k2 and DW are correction
factors for the shorting plate effect. Equation (1) is
evaluated against measured resonant frequency, and is
shown in Fig. 5. The maximum relative difference between
the resonant frequencies is less than about 0.5%.
The radiation patterns of the proposed antenna in the
two principal planes are presented in Fig. 6. It is observed
535

that the radiation pattern of Ey in the two planes is near

omni-directional. Since current is concentrated on the
shorting plate, radiated power Ey can be seen to peak on
both planes at about y 301. As the PIFA acts like a
J

short dipole, when the ground plate is the same size as the
upper plate, the radiation from the two radiating plates
leads to nulls in the Ef (cross-polarisation) pattern of the

Fig. 4

Simulated current distributions for the proposed design

Fig. 6 Measured E-plane and H-plane radiation patterns for the

proposed design with Ws 25 mm

Other antenna parameters are the same as in Fig. 2

a Ws 0 mm
b Ws 10 mm
c Ws 25 mm
d Ws 40 mm

Other antenna parameters are the same as in Fig. 2

a E-plane (xz cut)
b H-plane (yz cut)

400

300

2200

250

2100
2000

200
150
100

1900

50

1800

measured

1700

fitting equation

1600

50
0

18

22

26

30

34

38

42

WS , mm

Fig. 5 Comparison of evaluated results of approximate equation

and experimental results for different shorting plate widths (Ws)
536

WS = 40 mm
WS = 35 mm
WS = 30 mm
WS = 25 mm
WS = 20 mm

350

2300

Zin(d ),

resonant frequency, MHz

2400

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

feed position d,

Fig. 7 Simulated input impedance against feed position with

different shorting plate widths (Ws)
Other antenna parameters are the same as in Fig. 2
IEE Proc.-Microw. Antennas Propag., Vol. 152, No. 6, December 2005

xz plane and Ef (co-polarisation) of the yz plane. In

addition, edge diffraction, owing to a nite ground plane,
introduces high cross-polarisation is noted to be higher than
the co-polarisation in the yz plane.
4

Parameter study

An important performance index of the antenna is the input

impedance measured at the coaxial feed. The objective here
is to achieve a minimum return loss at the operating
frequency. As indicated in Fig. 2a, the feed point of such

antennas has to be placed close to the shorting plate end to

obtain a 50 O impedance. The input impedance of different
shorting plate widths was also studied through simulation
analysis to obtain a simple and accurate design. Figure 7
shows the simulated input impedance Zin against feed
position d for the antenna with different shorting plate
widths (Ws). The results showed that the input impedance
can be matched to the 50 O impedance of the coaxial probe
by adjusting the location of the feed position d. For
do0.05 l, all the antennas are matched to the feed lines.
The impedance bandwidth of a PIFA is usually very
narrow, and is determined primarily by the antenna height

0
= 0

return loss, dB

10
10

15
20

20

h = 2 mm
h = 4 mm
h = 6 mm
h = 8 mm
h = 10 mm

25
30
35
1650

1750

90

90
1850

1950

2050

2150

frequency, MHz

Fig. 8 Measured return loss for the PIFA with different antenna
heights (h)
er 1.0, L Lg 30 mm, W Wg 40 mm, Ws 25 mm

Table 2: Performances of the PIFA with different antenna

height (h)

180
a

Antenna height
h (mm) (in l)

Feed position
d (mm)

Resonant
frequency
fc (MHz)

Bandwidth
(%)

2 (0.0135 l)

3.0

2029

1.23

4 (0.0259 l)

3.5

1946

2.26

6 (0.0381 l)

4.5

1907

2.67

8 (0.0497 l)

5.5

1862

3.49

10 (0.0602 l)

8.0

1807

4.76

= 0

10

20

The width of shorting plate is kept constant at Ws=25 mm

90

90

return loss, dB

5
10
15
20
25

180

30 40 mm2

50 60 mm2
70 80 mm2
80 90 mm2

30
1600

1700

1800

1900

2000

2100

frequency, MHz

Fig. 9 Measured return loss for various ground plane sizes

(Lg
Wg)
er 1.0, L 30 mm, W 40 mm, Ws 25 mm, h 4 mm
IEE Proc.-Microw. Antennas Propag., Vol. 152, No. 6, December 2005

30 40 mm2

70 80 mm2

50 60 mm2

80 90 mm2

Fig. 10 Measured E-plane (xz plane) and H-plane (yz plane)

radiation patterns at resonance for the PIFA with various ground
plane sizes
Other antenna parameters are the same as in Fig. 9
a E-plane (xz cut)
b H-plane (yz cut)
537

and dielectric constant of the substrate. To increase the

bandwidth of a PIFA, one can choose to increase the height
of the substrate or to use a substrate with smaller dielectric
constant. In this study, we choose the dielectric constant of
1.0 (air substrate) and vary the antenna height to obtain the
variation of the antenna bandwidth. Figure 8 shows the
measured return loss of the PIFA with xed shorting plate
width (Ws 25 mm) when the antenna height h is varied.
When the antenna height h is increased, the resonant
frequency decreases from 2029 MHz to 1807 MHz and the
impedance bandwidth increases from 1.23% to 4.76%. The
relation between the antenna height, resonant frequency
and bandwidth is listed in Table 2.
The PIFA is designed to operate with a ground plane and
the ground plane is an integral part of the antenna
assembly. In fact, the ground plane size is the limiting
factor for the PIFA in many applications. The effects of
ground plane size on the resonant frequency, impedance
bandwidth, radiation patterns and gain were studied. In this
study, the PIFA consists of a rectangular ground plate with
dimensions Lg
Wg mm2 and a shorting plate width of
25 mm and 4 mm height. Figure 9 shows the measured
return loss for a few ground plane sizes. Two observations
were made. First, the resonant frequency decreases with an
increase in the ground plane size, and second, the
impedance bandwidth increases when the dimensions of
the ground plane are increased. In this case, the ground
varies from 30
40 mm2 to 80
90 mm2. Figures 10a and
10b show the measured radiation patterns at resonant in the
xz and yz planes of the PIFA with various ground plane
sizes respectively. When the ground plane size is enlarged,
the co-polarisation in the xz plane is almost omnidirectional on the front side of the patch and has a larger
dip at around y 1501. In the yz plane, the back radiation
is reduced with increasing ground plane sizes. Figure 11
shows the measured peak gain of the PIFA for various
ground plane sizes. It is seen that gain is affected strongly by
ground plane size, and the peak gain increases as ground
plane size is enlarged. Table 3 summarises the effect of the
4.5
30 40 mm2
50 60 mm2
70 80 mm2
80 90 mm2

4.0

antenna gain, dBi

3.5
3.0
2.5
2.0
1.5
1.0
0.5
0
1650

1750

1850

1950

2050

frequency, MHz

Fig. 11
sizes

Measured peak antenna gain for various ground plane

Other antenna parameters are the same as in Fig. 9

538

Table 3: Performances of the PIFA with various ground

plane sizes (Lg3Wg)
Ground plane
size Lg
Wg
(mm2)

Feed
position
d (mm)

Resonant
frequency
fc (MHz)

Bandwidth
(%)

Gain
(dBi)

30
40

4.5

1946

2.26

0.75

50
60

6.0

1881

2.66

1.84

70
80

4.5

1830

2.84

3.38

80
90

5.0

1744

3.04

3.95

ground plane sizes on resonant frequency fc, bandwidth and

gain.
5

Conclusions

A measurement of a PIFA mounted on a nite ground

plane was performed to investigate the effects of the
shorting plate width. The results determined by simulations
and measurements showed very good agreement. Moreover, a simple and semi-empirical approximate equation has
been derived for the estimate of the resonant frequency for
different shorting plate width. The effects of ground plane
sizes on resonant frequency, bandwidth, gain and radiation
patterns are also studied.
6

Acknowledgment

This work was supported by the National Science Council

of Taiwan, under Grant 92-2213-E-151-023.
7

References

1 Salonen, P., Keskilammi, M., and Kivikoski, M.: New slot

congurations for dual-band planar inverted-F antenna, Microw.
Opt. Technol. Lett., 2001, 28, (5), pp. 293298
2 Yeh, S.H., Fang, S.T., and Wong, K.L.: Dual-band shorted patch
antenna for dual ISM-band application, Microw. Opt. Technol. Lett.,
2002, 32, (1), pp. 7980
3 Chen, H.T., Wong, K.L., and Chiou, T.W.: PIFA with a meandered
and folded patch for the dual-band mobile phone application, IEEE
Trans. Antennas Propag., 2003, 51, (9), pp. 24682471
4 Yeh, S.H., Wong, K.L., Chiou, T.W., and Fang, S.T.: Dual-band
planar inverted F antenna for GSM/DCS mobile phones, IEEE
Trans. Antennas Propag., 2003, 51, (5), pp. 11241126
5 Abedin, M.F., and Ali, M.: Modifying the ground plane and its effect
on planar inverted-F antennas for mobile phone handsets, IEEE
Antennas Wirel. Propag. Lett., 2003, 2, pp. 226229
6 Huynh, M.C., and Stutzman, W.: Ground plane effects on planar
inverted-F antenna performance, IEE Proc., Microw. Antennas
Propag., 2003, 150, (4), pp. 209213
7 Row, J.S.: Experimental studies of the dual-band planar inverted slot
antenna with a U-shaped slot, Microw. Opt. Technol. Lett., 2003, 37,
(5), pp. 359361
8 Mak, C.L., Chair, R., Lee, K.F., Luk, K.M., and Kishk, A.A.: Half
U-slot patch antenna with shorting wall, Electron. Lett., 2003, 39,
(25), pp. 17791780
9 Tung, H.C., and Wong, K.L.: A shorted microstrip antenna for
2.4/5.2 GHz dual-band operation, Microw. Opt. Technol. Lett., 2001,
30, (6), pp. 401402
10 Mok, W.C., Chair, R., and Lee, K.F.: Wideband quarter-wave patch
antenna with shorting pin, IEE Proc., Microw. Antennas Propag.,
2003, 150, (1), pp. 5660
11 Garg, R., Bhartia, P., Bahl, I., and Ittipiboon, A.: Microstrip antenna
design handbook (Artech House, Norwood, MA, 2001)

IEE Proc.-Microw. Antennas Propag., Vol. 152, No. 6, December 2005