Professional Documents
Culture Documents
Review Article
ISSN 1751-8725
Microstrip patch antenna miniaturisation Received on 9th September 2014
Accepted on 18th December 2014
techniques: a review doi: 10.1049/iet-map.2014.0602
www.ietdl.org
Abstract: The microstrip patch antenna (MPA) has been in use and has been studied extensively during the past three
decades. This antenna, which consists of a metallic patch printed on a dielectric substrate over a ground plane, offers
several advantages including ease of design and fabrication; low profile and planar structure; and ease of integration
with circuit elements. The minimum dimension of a conventional MPA is in the order of half a wavelength. In recent
years, with the advent of new standards and compact wireless devices, there has been a need to reduce the size of this
type of antenna. This study discusses some of the principal techniques that have been reported in the literature to
reduce the size of an MPA. These miniaturisation techniques include material loading, reshaping the antenna, shorting
and folding, introducing slots and defects in the ground plane and the use of metamaterials. The major features and
drawbacks of each of these approaches are highlighted in this study along with their effects on the antenna
performance metrics.
1 Introduction of the new standards also require the use of multiple antenna
elements especially for multiple-inputmultiple-output (MIMO)
Wireless communication devices are ubiquitous nowadays and their applications. Thus accommodating multiple antenna elements in
use is steadily on the rise. These devices include AM and FM radios, limited space, typically specied by the device manufacturer, is a
cellular mobile phones, tablets, laptops, satellite phones difcult task and challenges the engineers to look for novel ways
and receivers, global positioning system (GPS) devices, to design miniaturised antennas. Although, the size of such
radio-frequency identication (RFID) systems, to name a few. antennas should be small, they must maintain desirable radiation.
The antenna, being an integral part of these devices, plays an An antenna which can t within a half radian sphere is called an
important role in dening the performance of these devices. electrically small antenna (ESA), which has been studied
Therefore the design of an antenna for any wireless device should extensively. Particularly, a number of theoretical works have been
be carried out with care in order to guarantee good system-level published on the topic of performance limits of small antenna. All
performance. theories have reached the same conclusion that the size of an
Among various kinds of antennas, printed antennas have received antenna can only be reduced at the expense of its bandwidth and
considerable attention during the past few decades because of their gain [1, 2]. These theories provided a lower limit on the antenna
low-prole nature and their ease of integration with associated Q-factor given its dimensions. Although these theories are useful
electronics, which make them very suitable for use in compact for dening performance metrics, they do not offer any systematic
wireless devices. A printed antenna is normally fabricated using clues for designing practical ESAs. Recently, the design of ESA
printed circuit technology. Printed antennas were rst presented has been an area of interest for many research groups and many
during the 50s of the past century but they did not gain much novel miniaturised antenna structures have appeared in the
importance until the early 80s. Many designs of printed antennas literatures [35]. These include the miniaturised versions of
have appeared in the literature since the 80s. Several of these different types of antennas including the patch, the printed
designs were thoroughly analysed analytically as well as monopoles, PIFA, among others.
numerically and their theory was developed for better The MPA is a planar printed antenna, which is well known, and
understanding of their performance characteristics. The most whose theory of operation is well developed. In this paper, the
commonly used printed antennas are microstrip patch antennas design of an MPA is discussed, with a focus on how to
(MPAs); printed monopoles and dipoles; slot antennas; loop miniaturise it. All of the major and novel techniques, which have
antennas; and planar inverted F-shaped antennas (PIFA). appeared in the literature related to MPA miniaturisation are
Although other parts of a communication device have seen a surveyed in this review paper. This paper serves as an essential
considerable reduction in their size because of the use of starting point for students and professionals involved in antenna
micro-fabrication technology, reducing the size of the antenna is design and miniaturisation by showing the pros and cons of each
still a difcult and challenging task. A conventional antenna method, thus making it easier to choose the appropriate method
resonates at a frequency when its length is on the order of a for the application at hand. In addition, we point out some
half-wavelength of that frequency. Many of the new standards for important aspects which should be considered for antenna
wireless communications including Wi-Fi, fourth generation/ miniaturisation problems.
long-term evolution, worldwide interoperability for microwave This paper is organised as follows. The basics of MPA are
access etc. are dened in the frequency range of 700 MHz6 GHz. described in Section 2. The main approaches reported in the
Thus, the conventional length of an antenna operating at the lower literature for MPA miniaturisation are presented and discussed in
band of these standards would be very large, that is, 214 mm at Section 3. A summary of the miniaturisation techniques and their
700 MHz. However, such a length far exceeds the specication of effects on the performance of the antenna are discussed in
many practical devices, including the RFID-based cards, mobile Section 4. Conclusions and some general remarks are presented in
phones, tablets, among other portable terminals. Moreover, most Section 5.
2 Patch antenna quality factor (Q) of the antenna are also obtained from the same
model. The resonant frequency of a rectangular MPA is a function
MPAs are fabricated by placing a conductor on a substrate backed by of the material properties of the substrate, its thickness and the
a ground plane. Although they can be designed in various shapes, dimensions of the patch. The resonant frequency ( fr), obtained
circular and rectangular patches (see Fig. 1) are the ones that are from the cavity model is given by [8]
widely used, since their design procedure is simple and well
developed [6]. MPAs have several advantages: (i) they have been
mp2 np2 pp2
1
extensively studied and their theory is well understood, making (fr )mnp = + + (1)
them easy to design; (ii) their planar geometry makes them easy to 2p me h L W
fabricate and integrate with other electronic devices; (iii) they can
be fabricated on a variety of cheap available substrates, making where h is the height of the substrate, L is the length of the patch, W
them a cost-effective choice for many applications; (iv) they are is the width of the patch, and are the permeability and permittivity
rugged and low prole, which make them ideal for use in a of the substrate, respectively. The integers m, n and p correspond to
number of practical applications such as in GPS receivers, tablets, the modes of operation. In a standard MPA, h L and h W . If
personal digital assistants (PDAs) and so on. In addition to the condition h < W < L is met, the lowest resonance corresponding
circular and rectangular shapes, MPAs with other geometries also to the dominant mode occurs when m, n and p are 0, 1 and 0,
exist, for instance triangular and annular ring-slot shapes [6]. In respectively. The resonant frequency of the dominant mode is thus
the next section, we will show how changing the patch geometry given by
can help to reduce its size.
MPAs are typically analysed using the cavity model [7]. An MPA
1
can be considered as a cavity lled with a dielectric with non-perfect (fr )010 = (2)
electric conductor (PEC) side walls. Thus, the radiation from this 2L me
cavity takes place because of leakage from these side walls. To
nd the radiation characteristics and resonant frequency of the Apart from the rectangular MPA, the theory of circular MPAs (see
MPA, the elds inside the cavity are solved rst. The top and Fig. 2) is also well developed and it is also widely used [8]. The
bottom parts of the cavity are considered as PECs, whereas the circular MPA can also be modelled using the cavity model [7]. In
sides are assumed to be perfect magnetic conductors (PMCs). The this case, the cavity is circular with its top and bottom surfaces
eld distribution inside the cavity is found by applying appropriate assumed to be PEC discs while the boundary around the circular
boundary conditions on the cavity walls. The solution is then used periphery of the cavity is assumed to be PMC. By solving the
to nd the resonant frequency of the patch for different modes, as elds inside in the cavity, various parameters of the antenna are
well as to determine radiating elds. The input impedance and the found. The resonant frequency of the dominant mode of a circular
MPA miniaturisation using a magnetodielectric substrate has been In the practical implementation of a quarter-wavelength MPA,
analysed in details in [15]. A magnetodielectic substrate design was placing a continuous conducting sheet at the edge between the
presented in [16]. The MPA fabricated on the proposed substrate was patch and the ground plane is difcult. An easier approach to
found to be 65% smaller, as compared with a conventional patch fabricating a quarter-wavelength MPA is to add an array of
resonating at 2.45 GHz. The MPA had a fractional bandwidth of shorting pins near the edge of the patch, as shown in Fig. 4.
0.5% and radiation efciency was 45%. A number of works have appeared in the literature that presented
Many other works have also appeared in the literature which the design or analysis of miniaturised MPAs using the shorting or
followed approaches that are similar to the ones mentioned folding techniques [1927]. In [19], a shorted patch was folded to
previously. A modied ceramic substrate was used in [17], make it a l/8 MPA. The radiation efciency of the antenna was
whereas [18] used a perforated substrate to miniaturise an MPA. In found to be 90% while it had a bandwidth of 4%. The schematic
summary, ceramic substrates and modied engineered substrates diagram of the shorted-folded patch is shown in Fig. 5. In [27],
can provide a signicant reduction in the patch size. However, the results of a parametric study of using a single, double or
major disadvantages of this method of size-reduction are the cost multiple shorting posts with a circular MPA have been presented.
of such substrates as well as reduction of the obtained bandwidth. Various parameters were analysed. It was found that the MPA
could be reduced by more than a factor of 3, as compared with a
standard patch via the use of optimal placement of the shorting
posts. A theoretical analysis of an MPA with a shorting post has
3.2 Shorting and folding
been provided in [24].
Folding an MPA and the use of shorting posts has been used to MPA can therefore be miniaturised using folding and shorting.
reduce the size of MPAs, and to render it electrically small [6]. However, this method comes at a cost of greatly decreasing the
For a half-wavelength rectangular MPA, the E-eld distribution gain and directivity of the antenna. Moreover, this method
under the patch has a sinusoidal pattern with a maximum E-eld complicates the geometry of the MPA at times and also makes it
at the radiating edges and zero in the middle. If an electric wall is non-planar, which adds to the complexity of the structure and its
placed at the middle of the patch, and the other half is removed, it conformity. However, when properly applied, this method has
would still resonate at the same frequency. Such a patch is called a little effect on the efciency of the antenna.
quarter-wavelength MPA. The theoretical analysis shows that a
quarter-wavelength patch has the same Q as that of its 3.3 Reshaping or introducing slots
half-wavelength counterpart [6]. However, the fact that the antenna
aperture has reduced results in the decrease in antenna directivity, An MPA can also be miniaturised by making changes in the shape of
thus directly affecting the antenna gain. the patch or by adding slots in the MPA. Miniaturised patches can be
shaped and optimised to obtain a large electrical length in limited
space using the genetic algorithm [28], running on
high-performance computing platforms. Fractal geometries are also
used to obtain miniaturised MPAs with good efciency [5, 29].
Fractals are space lling contours in which electrically large
features can be effectively packed in a relatively small space, with
a reduction in bandwidth.
A miniaturised MPA suffers from higher ohmic losses, which lead
to a lower radiation efciency. To mitigate this problem, an
Fig. 4 Quarter-wavelength MPA fabricated using shorting posts [6] Fig. 5 l/8 shorted/folded patch antenna [19]
etched out from underneath the MPA. This resulted in a 68% antenna shown that the proposed DGS provided greater miniaturisation
miniaturisation in terms of the size of the MPA. The proposed DGS than did the other two DGS-based designs. The use of DGS
was analysed and compared with other cross-shaped and structures usually provide lower efciency and yield narrower
dumbell-shaped DGS-based miniaturised patch antennas. It was operating bandwidths. In addition, re-tuning of the antenna is
1 h2 m
= 2 (7)
1 + h2 m1
Fig. 9 Geometry of the annular ring patch on an MTM substrate [58] From the analysis of the above equation, it was found that it was
material loading high dielectric substrates, ceramic high degree of miniaturisation, expensive materials limited bandwidth
substrates, magnetodielectric easy design procedure
substrates
shorting and folding shorting pins, shorting wall, up to four times miniaturisation, no standard design procedure, makes antenna
folding cost-effective solution geometry complex, non-planar because of
folding, very low gain and directivity
reshaping a patch or fractal antenna, engineered can provide wider bandwidth, make antenna geometry complex, affects the
introducing slots conductors, slots in the patch up to eight times miniaturisation radiation characteristics, no standard design
procedure, poor polarisation purity
modifications in slots in ground plane, use of DGS up to eight times miniaturisation, low efficiency increases the back lobe level no
ground plane antenna geometry remains planar standard design procedure
and simple
use of MTMs use of ENG, MNG or DNG high degree of miniaturisation limited bandwidth, low efficiency, complex
substrates, use of MTM inspired antenna geometry, no standard design procedure
techniques
possible to design a small antenna by choosing the core as an MNG with the models used to analyse such structures as such models
medium and using DPS for the outer medium. The computed usually ignore the polarisation of the eld that might cause
far-eld radiation patterns showed that such an antenna had better different behaviours as compared with the normal incidence/
radiation characteristics than a rectangular patch on an ENG/DPS non-polarised models used in most analysis to derive the effective
medium. medium properties. Thus caution has to be taken and full
There are no naturally occurring homogenous ENG or MNG understanding of the physics is required.
media, and realising a practical miniaturised MPA based on the
above design is not possible unless the above-mentioned media are
synthesised articially. Articially created MNG or ENG structures 4 Summary and features of miniaturisation
are frequency dependent and highly dispersive. Furthermore, the methods
size of the unit cells making the MNG or ENG medium sometimes
make it impossible to realise a practical antenna. Nevertheless, Table 1 summarises the various techniques discussed in this paper
many MTM-inspired MPA designs have been presented in the for MPA miniaturisation. The table highlights the features of each
literature, among which many are for MPA miniaturisation. technique, and lists its major advantages and disadvantages. As
A magnetic-permeability-enhanced MTM substrate for MPA seen from the table, almost all the mentioned techniques lead to
miniaturisation has been presented in [60]. The substrate consisted signicant miniaturisation. However, some techniques such as
of periodically arranged SRRs placed vertically between the patch folding or MTM-inspired can make the antenna structure more
and the ground plane. This resulted in signicant miniaturisation complex and non-planar. Moreover, most techniques reported so
of the patch antenna. The area of the MPA was reduced by a far provide little insight into a general design procedure, and they
factor of 10 as compared with a conventional MPA printed on an do not mention how the technique can be applied to design
FR4 substrate. However, since the design called for a vertical antennas in other bands. Hence, more attention should be given to
arrangement of the SRRs, the substrate was much thicker than the this issue in further works. It is also important to mention that
normal substrate. many of the patch miniaturisation gures reported in the literature
In [4], an MTM-inspired technique was used for MPA only mention the decrease in patch size (or radiating structure size)
miniaturisation. A CSRR was inserted in the substrate horizontally and do not take into account the size of the ground plane. Since
between the circular patch and the ground plane, which resulted in the size of the ground plane greatly affects the performance of an
a decrease in antenna size as well as its bandwidth. This decrease MPA, it cannot be neglected when dening the size of the antenna
was 75% at 2.45 GHz, as compared with that of a conventional
MPA fabricated on an FR4 substrate. However, this was also
accompanied by a decrease in the bandwidth and the radiation
efciency of the antenna. The antenna had a fractional bandwidth
of only 0.4% and an efciency of 28.1%.
An MNG medium created by three helices was placed between a
circular patch and the ground plane to reduce the size of the patch
[61]. This resulted in 60% miniaturisation in the patch area. The
antenna resonated at 735 MHz with a fractional bandwidth of
0.5% and a maximum gain of 7.9 dBi. In [62], an 80% reduction
in the size of patch was achieved by loading it with MTM
transmission-line. The miniaturised MPA utilising CSRR, etched
out in the ground plane, which was mentioned in the previous
section [52], can also be termed as an MTM-inspired antenna
since it used an MTM structure for miniaturisation. The same is
true for other CSRR-based MPAs discussed in the previous section.
In summary, various MTM-based as well as MTM-inspired
techniques have been implemented for MPA miniaturisation.
Although they have been successful in reducing the size, this has
Fig. 11 Comparison of bandwidths of various miniaturised patch antenna
been at a substantial cost in terms of complex material use, very
designs
narrow operating bandwidths and low radiation efciency. In
(Black = material loading, red = folding and shorting, pink = reshaping or introducing
addition, in MTM miniaturisation methods, care should be taken slots, green = modication in ground plane and blue = use of MTM)