This article has been accepted for publication in a future issue of this journal, but has not been
fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/LAWP.2016.2536678, IEEE
Antennas and Wireless Propagation Letters
A Dual-Band Highly Miniaturized Patch
Antenna
Ahmad A. Salih, and Mohammad S. Sharawi, Senior Member, IEEE
Abstract A highly miniaturized dual-band patch antenna is
proposed for small form factor (SFF) devices. The antenna is
designed to cover two wireless local area network (WLAN) bands
at 2.4 and 5.2 GHz. The antenna is miniaturized using a shorting
post and a novel defected ground structure (DGS). The patch size
is 8.8mm by 10 mm which means 74% miniaturization ratio is
achieved at 2.4 GHz. The dual-band property is achieved by
etching a U-slot in the ground as a part of the proposed DGS.
The antenna is fabricated on Rogers RO4350 board and its
radiation characteristics are measured. The measured results
have good agreement with simulated ones.
Index Terms dual-band, electrically small antennas (ESA),
miniaturization, patch antenna, WLAN.
I. INTRODUCTION
M ANY wireless devices have been developed successfully
in recent years such as smart cell phones, Global
Positioning System (GPS) receivers and wireless sensors [1].
Most of these handheld wireless devices have size limitations.
Due to the need for small wireless devices, the term Small
Form Factor (SFF) wireless devices was introduced to refer to
wireless devices that have small dimensions (height, width,
weight and configuration). Such SFF wireless devices are
suitable candidates for wearable wireless device applications.
SFF wireless devices consume lower power than their
conventional counterparts and this is important when
considering long life devices that can be turned on for a long
time [2]. For SFF wireless devices, small antennas are a must.
Small antennas with sizes much less than one wavelength at
their operating frequencies are referred to as Electrically Small
Antennas (ESA) [3]. Many mathematical definitions were
developed for ESA, one of the widely used ones is to satisfy
ka < 1 ([3]) where k is the wave number (2/) and a is the
radius of the smallest sphere containing the radiating part at
the antenna operating frequency. ESA parameters such as
bandwidth, efficiency and gain decrease with the decrease of
their electrical size ka. Most of ESAs have the familiar
omnidirectional dipole antenna radiation pattern [3].
Patch antennas are widely used due to their attractive
properties such as light weight, low profile and ease of
Manuscript received October 17, 2015. This work was supported by
KACST, Saudi Arabia, through the National Technology Plan for Science and
Technology (NSTIP) under Project No. 12-ELE3001-04.
The authors are with the Electrical Engineering Department, King Fahd
University for Petroleum and Minerals (KFUPM), Dhahran 31261, Saudi
Arabia (e-mail: eng.ahmad054@gmail.com; msharawi@kfupm.edu.sa).
1536-1225 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
fabrication. Many methods for miniaturizing dual-band patch
antennas were reported in literature such as shorting antennas
via a shorting post or a shorting wall like those reported in [4],
[5] where 60% miniaturization ratio was achieved in [4]. The
use of defected ground structures (DGS) to miniaturize patch
antennas and achieve dual-band property was reported in [6],
[7] and the highest miniaturization ratio obtained was 68% in
[6]. Etching slots on the patch to increase the current path
length is used extensively in miniaturizing patch antennas and
the highest size reduction ratio reported was 64% in [8]. Using
a coplanar waveguide (CPW) with slits [9] and CPW with
strips [10] were reported to miniaturize dual-band patch
antennas and 61% size reduction was achieved [9]. The
calculations of miniaturization ratios depend on calculating the
ratio for each dimension (length and width) separately and
then taking the average of them. From the aforementioned
quick survey about dual-band patch antenna miniaturization
techniques, the highest miniaturization ratio reported thus far
was 68% in [6], but for single band patch antennas the
miniaturization ratio is higher and can reach 83% as achieved
in [11]. Most of the highly miniaturized patch antennas have
very low radiation efficiency [12].
In this work, a compact dual-band patch antenna is designed
to cover two WLAN bands at 2.4 (802.11b/g/n) and 5.2
(802.11 a/ac) GHz. 74% miniaturization ratio is obtained by
shorting the patch via a shorting pin and etching a novel DGS
to match the antenna at the desired frequencies. The dual-band
property is achieved by using a U-slot etched in the ground.
The novelty of the design comes from getting the highest
miniaturization ratio with relatively acceptable radiation
efficiency in both bands.
The rest of the paper is organized as follows: Section II
presents the antenna design details, Section III presents the
results and discussions with parametric studies and Section IV
concludes the paper.
II. ANTENNA DESIGN DETAILS
Fig. 1 shows the proposed patch antenna geometry and the
detailed dimensions are listed in Table I. The proposed
antenna is simply a patch antenna miniaturized using a
shorting post close to the feeding position and a novel DGS.
The dual-band property is achieved by etching a U-slot in the
ground. As the current goes to the ground through the shorting
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/LAWP.2016.2536678, IEEE
Antennas and Wireless Propagation Letters

x
y
z L2
L1
L3 W2
L Lp via L
via W3 via W1
Feed L6
Feed Feed
L4
Wp
L5

t (a) (b)
W W

(a) (b) (c) Fig. 3. Current distribution for the proposed patch antenna at 5.2 GHz (a) top
layer, (b) bottom layer.
Fig. 1. The patch antenna geometry (a) top view, (b) side view, (c) bottom
view.
0
TABLE I
DETAILED DIMENSIONS OF THE PROPOSED PATCH -5
Parameter Value (mm) Parameter Value (mm) L2=4mm
L 20 L4 8.9 L2=5mm
-10
L2=6mm
W 18.8 L5 6

S11 (dB)
L2=7mm
Lp 10 L6 1.8 -15 L2=8mm
Wp 8.8 W1 1
L1 6 W2 1.2 -20
L2 5.9 W3 0.6
L3 2.6 t 0.762 -25

-30
post, this makes the ground part of the antenna. The patch 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6
Frequency (GHz)
antenna is fed with an SMA connector acting as a feeding Fig. 4. Effect of the U-slot width (L2) on antenna behavior.
probe at its edge. The antenna is fabricated on a double sided
Rogers RO4350 substrate with dielectric constant of 3.48, loss TABLE II
EFFECT OF THE U-SLOT WIDTH (L2) ON THE ANTENNA
tangent of 0.004 and thickness of 0.76 mm. The total antenna RESONANCE
size along with its GND plane is 18.8mm by 20mm by Value 1st resonance 2nd resonance Freq. Difference
0.76mm which represents an ESA at 2.45 GHz even when the (mm) (GHz) (GHz) (GHz)
4 2.8 5.14 2.34
circle surrounds all the substrate. The radius a of the circle is 5 2.58 5.12 2.54
13.86 mm so that ka = 0.7 < 1. 6 2.41 5.17 2.76
7 2.1 5.23 3.13
8 1.88 5.3 3.42
III. RESULTS AND DISCUSSIONS
The proposed patch antenna was modeled, optimized and To investigate more how the U-slot can control the antenna
simulated using HFSSTM (version 15). Fig. 2 and Fig. 3 show resonance, parametric studies on the U-slot width and arm's
the proposed patch antenna current distributions at 2.43 and length were conducted and their effect on the antenna
5.2 GHz, respectively. At 2.43 GHz, the highest current resonance was observed. The sweeps were conducted when all
intensity traces two of the patch edges. Although the U-slot is other parameters were set to the values in Table I. Fig. 4
etched on the ground, it affects the current on the patch. As the shows the effect of the U-slot width (L2) when it is changed
current go around the U-slot position in the ground, its from 4 to 8 mm. Changing L2 will affect the lower band and
electrical length increases thus explaining the miniaturization has no effect on the higher band which indicates that the
rule of the DGS. The current path length is found to be 32 mm frequency difference between the lower and higher band can
which is approximately /4. At 5.2 GHz, the current is over be controlled by changing L2 and this conclusion is shown
two edges of the patch and in the ground is surrounding the with exact numbers in Table II. Fig. 5 shows the effect of the
DGS. The current path is approximately 23 mm which U-slot arm's length (L3) when changing it from 1 to 5 mm.
corresponds to about /2. Changing L3 will affect both frequency bands with the same
amount and this is shown clearly in Table III as all numbers in
the frequency difference column are almost the same. To
conclude, by increasing L2 the lower band resonance can be
reduced with reduced matching levels and changing L3 will
shift both bands with the same amount of frequency, such that
larger L3 values will give lower resonance frequency with
worse matching levels.
Fig. 6 shows the fabricated patch antenna top and bottom
(a) (b) layers. The fabricated antenna was measured using an Agilent
N9912A FieldFox VNA. The simulated and measured
Fig. 2. Current distribution for the proposed patch antenna at 2.43 GHz (a) top
layer, (b) bottom layer. resonances of the antenna are shown in Fig. 7 for the optimum

1536-1225 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/LAWP.2016.2536678, IEEE
Antennas and Wireless Propagation Letters

L3=1mm
-5 L3=2mm
L3=3mm
S11 (dB)

L3=4mm
-10 L3=5mm

-15

-20
1.5 2 2.5 3 3.5 4 4.5 5 5.5 6
Frequency (GHz)
Fig. 5. Effect of the U-slot arm's length (L3) on antenna behavior.

TABLE III
EFFECT OF THE ARM'S LENGTH (L3) ON THE ANTENNA RESONANCE
Value Lower resonance Higher resonance Freq. Difference
(mm) (GHz) (GHz) (GHz)
1 2.6 5.35 2.75
2 2.45 5.2 2.75
3 2.23 4.9 2.67 Fig. 8. Patch antenna radiation pattern measurement setup, inset is a zoomed
4 1.92 4.64 2.72 in version.
5 1.72 4.4 2.68
5 100

0 80
Gain

Efficiency (%)
Gain (dB) Efficiency
-5 60

-10 40

-15 20

-20 0
2 2.5 3 3.5 4 4.5 5 5.5 6
Frequency (GHz)
Fig. 9. Patch Antenna measured maximum gain and efficiency

The antenna radiation characteristics were measured at

(a) (b)
Fig. 6. The fabricated antenna top and bottom layers (a) top view, (b) side
view
Fig. 7. The patch antenna measured and simulated resonance
values in Table I. The measured resonance (circled red curve)
had lower matching than the simulated one (dotted blue curve
in Fig. 7) when the connector model was not incorporated in
the initial model. After modeling an SMA connector and
incorporating it in the simulation model, better agreement with
measured results was observed (solid green curve). The small
ripples in the measured curve might be due to the cable and
connection junction (cable-SMA) which were not accounted
for in the simulation model (see [13]). The antenna has a
measured -6 dB bandwidth of 40 MHz at 2.43 GHz and 1.2
GHz at 5.2 GHz.
MVG-Italy, in a Satimo Star-Lab near field chamber as shown
in Fig. 8. The measured maximum gain and efficiency are
shown in Fig. 9. The antenna had measured maximum gains of
-1.7 and 2.4 dB at 2.43 and 5.2 GHz, respectively. The gain at
2.43 GHz is lower than the simulated gain of 0.8 dB and this
due to the lower matching levels (lower total efficiency)
between the simulated and the fabricated antenna. This
mismatch can be attributed to the connector effect and its size
with respect to the antenna ground and hand soldering. Also,
the fact that the ground is very small makes some current to
flow back to the cable and that will affect the efficiency as
was shown in [13]. For the higher band at 5.2 GHz, the
measured gain and the simulated one are in very close
agreement as the maximum simulated gain value was 3 dB.
The measured efficiency of the antenna was -5.27 dB (30%)
and -0.9 dB (81%) at 2.43 and 5.2 GHz, respectively, which is
considered acceptable for miniaturized dual band patch
antennas. Fig. 10 shows the simulated and the measured 2D
radiation patterns (total E-field) in xz and yz planes at 2.43
and 5.2 GHz. The antenna has almost omnidirectional pattern
as expected for small antennas with linear polarization.
Table IV shows a comparison between the proposed patch
antenna and other dual-band miniaturized patch antennas that
appeared in literature in terms of miniaturization ratio, gain

1536-1225 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/LAWP.2016.2536678, IEEE
Antennas and Wireless Propagation Letters

TABLE IV
COMPARISON BETWEEN THE PROPOSED PATCH ANTENNA AND OTHER ANTENNAS FROM LITERATURE
Reference Band 1 Band 2 Patch Miniatur- Ka factor containing Maximum Gain (dB) Efficiency (%)
(GHz) (GHz) ization (%) the antenna Ground Band 1 Band 2 1 2
[4] 2.45 5.8 60 - 0.5 1.5 - -
[5] 1.7 8.1 46 1.6 - - - -
[6] 900 2.45 68 0.8 2.6 3.5 - -
[7] 3.5 5.8 58 - 5.4 6.4 - -
[8] 2.7 4.5 64 0.6 - -4 - -
[9] 2 5.32 60 0.75 2.5 4.3 - -
[10] 1.8 3.4 60 1.2 - - - -
This work 2.43 5.2 74 0.7 -1.7 2.4 30 81

0 -1 0 -1
30 30 30 30
is obtained by using a shorting post and a novel DGS. The
-6 -6 antenna is fabricated on Rogers RO4350 board with size of
60 -11 60 60 -11 60 18.8mm by 20mm by 0.76mm. The measured -6 dB
-16 -16
bandwidth was 40 MHz and 1.2 GHz at the lower and
higher bands, respectively. The antenna has a measured
90 90 90 90
gain of -1.7 and 2.4 dB at 2.43 and 5.2 GHz, respectively.

120 120 120 120

REFERENCES
150 150 150 150
180 180 [1] H. Wong, K. Luk, C. H. Chan, Q. Xue, K. K. So and H. W. Lai,
(a) (b) Small Antennas in Wireless Communications, Proceedings of the
IEEE, vol. 100, no. 7, pp. 21092121, July 2012.
02 03
[2] Matthew Weaver and Duane Wessels, Make projects: Small Form
30 30 30 30 Factor PCs, O'REILLY, 2008, pp.14.
-3 -2 [3] J.L. Volakis, C.C. Chen, and K. Fujimoto, Small antennas:
60 -8 60 60 60
miniaturization techniques & applications, McGraw Hill, 2010.
-7
[4] G. Liu, D. Wang, F. Jin and S. Chang, Compact Broad Dual-Band
-13 -12 Antenna Using A Shorted Patch with A Thick Air Substrate for
90 90 90
Wireless Body Area Network Application, International
90
Symposium on Microwave, Antenna, Propagation and EMC
Technologies for Wireless Communications (MAPE), pp. 1821,
2011.
120 120 120 120
[5] S. H. S. Esfahlani, A. Tavakoli, and P. Dehkhoda A Compact
Single-Layer Dual-Band Microstrip Antenna for Satellite
150 150 150 150 Applications, IEEE Antennas and Wireless Propagation Letters,
180 180
(c) (d) vol. 10, pp. 931934, September 2011.
[6] M. Taouzari, J. El Aoufi, A. Mouhsen, H. Nasraoui and O. El
Fig. 10. Simulated and measured 2-D radiation pattern. Simulated: Dashed Mrabat, 900 MHz and 2.45 GHz Compact Dual-Band Circularly-
lines. Measured: Solid lines. (a) xz-plane at 2.43 GHz. (b) yz-plane at 2.43 Polarized Patch Antenna for RFID Application, Conference on
GHz. (c) xz-plane at 5.2 GHz. (d) yz-plane at 5.2 GHz. Microwave Techniques (COMITE), pp. 14, 2015.
[7] S.K. Singh, P. Consul and K.K. Sharma, Dual-band gap coupled
and efficiency. The proposed patch antenna has the highest microstrip antenna using L-slot DGS for wireless applications,
International Conference on computing, communication and
miniaturization ratio with acceptable gain and efficiency. automation (ICCCA), pp. 13811384, 2015.
Notice that no measured radiation efficiency was mentioned [8] A. Saghir, S. M. Abbas, M. U. Afzal, T. Tauqeer and M. H. Tariq,
in almost all previous works. Since ESAs cannot perform Compact Dual-band Microstrip Antenna Design Using Slits,
International Conference on computer, control and communication
without their ground (since it is a part of the radiator), a
(IC4), pp. 14, 2013.
column is added to Table IV for the calculated ka factor [9] W.C. Liu and W.R. Chen, CPW-fed compact meandered patch
when a sphere containing the antenna ground is considered. antenna for dual-band operation, Electronics Letters, vol. 40, issue
The smallest ka factor reported was 0.6 (in [8]) with very 18, pp. 10941095, September 2004.
[10] S. Nagar, U. Nagar and R. S. Meena, CPW - Fed dual-band patch
low realized gain when compared to our proposed one. The antenna for mobile applications, 5th International Conference on
tradeoff is clear. If the GND plane is increased (since its a Computational Intelligence and Communication Networks (CICN),
part of the antenna) the resonance frequency of the structure pp. 15, 2013.
[11] Prabhakar H.V., U.K. Kummuri, R.M. Yadahalli and V. Munnappa,
decreases and its efficiency starts increasing in both bands, Effect of various meandering slots in rectangular microstrip antenna
while the bandwidth improves in the lower band and ground plane for compact broadband operation, Electronics Letters,
degrades in the higher one. vol. 43, issue 16, pp. 848850, August 2007.
[12] M.S. Sharawi, M.U. Khan, A.B. Numan, and D.N. Aloi A CSRR
Loaded MIMO Antenna System for ISM Band Operation, IEEE
IV. CONCLUSIONS Transactions on Antennas and Propagation, vol. 61, No. 8, pp. 4265
4274, August 2013.
A miniaturized dual-band patch antenna is designed for [13] L. Liu, Y.F. Weng, S. W. Cheung, T. I. Yuk, and L. J. Foged,
small form factor wireless devices. The dual-band antenna Modeling of cable for measurements of small monopole antennas,
resonates at 2.4 and 5.2 GHz. The patch has a compact size IEEE Loughborough Antennas and Propagation Conference (LAPC),
of 8.8mm by 10 mm which means 74% miniaturization ratio pp. 1-4, 2011.

1536-1225 (c) 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.