Substrate Integrated Wave Guide Antennas

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/304195642
Substrate Integrated Waveguide Antennas
Chapter · January 2015

DOI: 10.1007/978-981-4560-75-7_57-1
CITATION READS
1 3,275
3 authors:
Tarek Djerafi Ali Doghri

Institut National de la Recherche Scientifique Université de Montréal
100 PUBLICATIONS 1,044 CITATIONS 17 PUBLICATIONS 112 CITATIONS
SEE PROFILE SEE PROFILE
Ke Wu
Montreal Polytechnic
886 PUBLICATIONS 20,080 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Innovative 94 GHz Components (Ph.D.) View project
IEEE Student Branch of Bordeaux View project
All content following this page was uploaded by Ali Doghri on 25 June 2016.
The user has requested enhancement of the downloaded file.

Handbook of Antenna Technologies
DOI 10.1007/978-981-4560-75-7_57-1
# Springer Science+Business Media Singapore 2015
Substrate Integrated Waveguide Antennas

Tarek Djerafi*, Ali Doghri and Ke Wu
Poly-Grames Research Center, École Polytechnique de Montréal, Montreal, QC, Canada
Abstract
Research and development progress in the area of millimeter-wave antennas has been associated since
recent years with substrate integrate waveguide (SIW). SIW was proposed and studied as a class of
efficient integrated transmission lines compatible with planar technologies, offering incomparable self-
consistent shielding and high-quality factor performances. SIW can be designed as open wave-guiding
structures and energy leakage will take place when the uniformity of those guides is perturbed or they are
not excited in an appropriate mode. This leakage effects may be used positively for the design of antennas
by deliberately introducing perturbations in these guides so that they radiate in a controlled fashion. With
the advantages of broadband, efficiency, and high gain, a specific benefit of such antennas is their
compatibility with SIW from which they are derived, thus facilitating an integrated design. In this chapter,
H-plane horns, leaky-wave structures (long and periodic) and tapered slot antennas are presented. Special
attention is given for tapered slot antennas and arrays with their application in the design of passive
imaging systems.
Keywords
Antenna array; Antipodal structure; Imaging system; Horn antenna; Leaky-wave; Millimeter-wave;
Polarization; Slot antenna; Rod antenna; Substrate integrated waveguides (SIW); Tapered slot antenna;
Twist; Directional coupler; Vivaldi antenna
Introduction
The frequency range from 30 to 300 GHz is generally defined as millimeter waves or extremely high-
frequency waves. Wavelengths over this frequency band are ranged from 10 to 1 mm, thus naming it the
millimeter-wave band or simply millimeter-wave (MMW or mmW). Compared to lower frequency bands,
electromagnetic waves in this band are subject to high atmospheric attenuations and they are absorbed by
atmospheric gases. Beyond this millimeter-wave range, electromagnetic radiation is referred to as
terahertz (THz) radiation. Millimeter-wave bands offer exciting opportunities for various bandwidth-
demanding and spectrum-sensitive applications such as short-range communications in the 60 GHz band
and automotive radar and imaging systems in the frequency windows of 77 and 94 GHz (Meinel 1995;
Solbach and Schneider 1999; Schoebel and Herrero 2010). Since a number of years, emerging applica-
tions are being reported such as imaging sensors for security and biomedical applications (Salmon 2004).
There is also a great deal of interest in millimeter imaging arrays for other applications such as remote
sensing, radioastronomy, and plasma measurement (Nanzer 2012).
In commercial system developments, low-cost, mass-producible, high-performance and high-yield
microwave, and millimeter-wave technologies are critical for developing successful and sustainable
*Email: tarek.djerafi@polymtl.ca
Page 1
DOI 10.1007/978-981-4560-75-7_57-1
commercial broadband systems, in particular, front end and antenna modules. Specifications of antenna
components and systems for those applications should satisfy not only desired electrical performances as
gain, low side lobe level (SLL), and low cross-polarization, but also stringent mechanical criteria of size,
weight, and shielding. The integrability with other active and passive system elements is also considered
as a main key criterion, which generally defines the ultimate cost and consequently the commercial
success. The millimeter-wave bands are useful for enabling the design of densely packed systems
integrated with high-gain antenna arrays.
Antennas in standard technology or in SIW can be classified as either resonant or nonresonant types.
Resonant antennas, such as microstrip patches and dipoles, resonate at discrete frequencies. Nonresonant
antennas or traveling-wave antennas do not resonate but rather leak energy out along their wave
propagating path, thereby producing more efficient broadband structures. The resonant structures do
not provide an easy-to-accept performance in the millimeter-wave range. This is because of very high
conductor losses arising from immense current densities (singularities) at the strip edges, especially in the
feeding network. The substrate becomes electrically thicker at higher frequency and more surface wave
modes (parasitic modes) may exist. This can be deleterious for standard microstrip antennas with radiation
efficiency reduction and radiation pattern perturbations, which result from the diffraction of surface waves
at the edges of the antenna structure in an uncontrolled manner (Pozar 1983).
In this chapter, different constraints and limits of SIW antennas are examined with focus on
no-resonance structures. At millimeter-wave, the energy is much easier to radiate out from open
transmission lines. As such, a number of leaky-wave or surface-wave antennas can be synthesized and
developed. They include tapered slot antenna (TSA), printed Yagi-Uda antenna or log-periodic dipole
array, dielectric rod antenna, long slot leaky-wave, partially reflective surface-based antenna, etc. For
more background information and development history, readers are referred to (Walter 1965; Oliner
et al. 2007; Suntinjo et al. 2008; Caloz et al. 2011). Tapered slot antenna (TSA), also known as flared notch
or Vivaldi antenna, is among the most promising candidates satisfying all the requirements described
above (radiation, mechanical, integrabilty, and cost). TSA has some unique characteristics of symmetrical
patterns in two planes, high gain in addition to having wide bandwidth characteristics in terms of radiation
performance and impedance characteristics. TSA has proved to be of particular interest due to its potential
as effective radiating elements in an array, and the possibility of using it at very high frequencies.
The integration of structure is a crucial issue for front-end and antenna system developments. The
conservative waveguide solutions, which have been widely deployed for numerous applications, present a
number of challenging problems in connection with size, cost, weight, and fabrication complexity of feed
networks and antenna systems. Various discontinuities and lossy structures within the feed networks
should be avoided or reduced or used with caution since noise emitted or interference created by such
components into the receiver could affect the overall performance. Substrate integrated waveguide (SIW)
with the advantages of self-consistent field shielding, low cost, and light weight is an excellent candidate
to substitute the traditional solutions for antenna array design as well for whole system development
(Djerafi and Wu 2013).
This chapter presents an overview and a summary of recent developments for leaky-wave antennas
(LWAs). An LWA makes use of a guiding structure that supports wave propagation along the length of the
structure with the wave radiating or “leaking” continuously along the structure. Such antennas may be
uniform, quasi uniform, or periodic in topology. After reviewing the basic physics and operating
principles behind, a summary of some recent advances for those types of structure is given and discussed.
Horn antenna based on SIW presents some constraint in connection with thickness. Recent advances on
SIW horn antenna will be presented.
The performance of TSAs with three different profiles will be compared over the Ka-band and
corrugation effects will be detailed. Mutual coupling effects are critical on the performance of antenna
Page 2
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 1 A potential SIW-based platform in which SIW antennas, couplers, cavities, filters are all combined with active and DSP
circuits on the same substrate or master board
arrays. The coupling in H and E-plane arrangement of two TSAs is studied as well as 90 configuration.
This configuration is useful to build up antenna arrays with dual or circular polarization capabilities.
Substrate Integrated Waveguide

A new generation of high-frequency integrated circuits called “substrate integrated circuits – SICs” was
proposed and demonstrated many years ago (Wu 2001). This disruptive concept has unified the hybrid
and monolithic integrations of various planar and nonplanar circuits that are made in single substrate
and/or multilayer platforms. Basically, any nonplanar structures can be made or synthesized in planar
form, which presents the foundation of SICs. Being synthesized on a planar substrate, the substrate
integrated image guide (SIIG), for example, can be combined in a hybrid way with the SIW and the
Substrate Integrated Nonradiative Dielectric guide (SINRD) on the same substrate, operating with
different modes and/or different frequency ranges. These guides can also be combined with standard
waveguide, microstrip, Coplanar waveguide (CPW) or slotline, leading to attractive hybrid schemes of
planar and nonplanar structures. In fact, an easy-to-handle low-cost hybrid design strategy is of critical
importance for the development of high-volume millimeter-wave ICs and systems as illustrated in Fig. 1.
These technologies have well been studied in the millimeter wave range. Since the transmission line
technology is instrumental for developing high-frequency electromagnetic hardware, the choice of an
appropriate waveguide or line structure is critical for millimeter-wave developments and applications.
The transmission lines of choice should allow high-density integration and mass-production scheme at
low cost. Rectangular waveguides have widely been used in the development of microwave- and
millimeter-wave components and systems with their salient features such as low insertion loss, high
quality factor (Q-factor) and high power capability, etc. However, they are also characterized by their
bulky size, stringent manufacturing precision and nonplanar geometry. Therefore, it is impossible to
design and develop microwave and millimeter-wave integrated circuits with this technological platform.
Benefiting from the properties of low profile, easy fabrication, and low cost, microstrip-like circuits
including coplanar waveguides (CPW) and strip lines are presently the principal choices of integration for
the development of microwave and millimeter wave circuits. Unfortunately, such printed circuits suffer
Page 3
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 2 The topology of a typical single-layered SIW. Metalized via and slot arrays for creating equivalent metallic fence or
wall configurations with: (a) cylindrical via arrays, (b) long slot trenches
from significant losses and packaging problems. In fact, the performances of microstrip-like circuits are
fundamentally limited by physical properties such as field or current singularities occurring at the stripline
(Zhu and Wu 2002).
SIW structure preserves most of the advantages associated with conventional metallic waveguides,
namely high Q-factor (low loss) and high power-handling capability with self-consistent electromagnetic
shielding. The most significant advantage of SIW technology is its capacity of enabling a possible
complete integration of all the components on the same substrate, including passive components, active
elements, and even antennas as presented in Wu et al. (2003, 2012), Deslandes and Wu (2006), Bozzi
et al. (2011). SIW techniques can be used to solve a series of headache problems, which renders it hugely
popular in the community today (Vye 2011). A remarkable problem arising at high frequency is the
appearance (trapping) of surface waves that generally decrease the antenna efficiency. The SIW can
effectively control this phenomenon. Since SIW components are bounded by conducting surfaces on both
sides of the substrate, they exhibit the merits of extremely low (completely negligible) radiation/leakage
loss and insensitive to outer interferences. The SIW technology has already experienced a rapid devel-
opment over more than one decade. This development allows the demonstration and applications of
innovative passive and active circuits, antennas, and systems at microwave and millimeter-wave frequen-
cies covering a very broad frequency range from subgegahertz to subterahertz. In addition, the SIW
technique can be combined with other SICs platforms to create multiformat and multifunction devices and
systems.
SIW Techniques and Design Basics

The SIW (also known as postwall waveguide or a laminated waveguide (Uchimura et al. 1998)) is a
rectangular waveguide-like structure in an integrated planar form, which can be synthesized and fabri-
cated by using two rows of conducting cylinders, vias or slots embedded in a dielectric substrate that is
electrically sandwiched by two parallel metal plates as illustrated in Fig. 2.
The operating frequency range is delimited by the monomode propagation of quasi-TE10 wave as its
cut-off frequency is only related to equivalent width aeq of the synthesized waveguide as long as the
substrate thickness or waveguide height is smaller than this width. This equivalent width will be discussed
in the following section.
Via Configurations
Round metalized via holes are used to create the electric sidewalls or fences of SIW through two parallel
via arrays. The discontinued current flow along the via- or slot-synthesized metalized sidewalls does not
allow the propagation of TM modes. In addition, a large width-to-height ratio of SIW supports the
Page 4
DOI 10.1007/978-981-4560-75-7_57-1
propagation of TEm0 modes. Since these techniques are amenable to arbitrarily shaped perforations, the
limitation of circular via shapes is no longer mandatory. Rectangular slot trenches were found to be
advantageous for lower leakage and better definition of the SIW sidewalls. This is important for some
designs such as those iris and window coupling geometries found in the filter design. Rounded corners
increase the overall mechanical stability, allowing for a better metallization, which often cannot be
avoided in the fabrication process due to the finite diameter of laser beams in the case of using a laser-
based micromachining process. Figure 2 shows two different slot trench (SIW) configurations:
(a) cylindrical via arrays and (b) longer slots for SIW operation between the first and second stop
bands. Note that the SIW structure is a periodic geometry, which is subject to the guided-wave phenomena
of all periodic waveguides such as bandgap (stopband) effects. Such stopbands, caused by the distributed
Bragg reflection, occur around frequencies at which the periodic spacing p is equal to a multiple of half a
guided wavelength (Patrovsky et al. 2007).
SIW Propagation Modes

A three-dimensional view of the SIW is illustrated in Fig. 2. The PCB's ground planes form the broad
walls of the SIW and a linear array of metalized via holes is used for the narrow walls. Since only vertically
oriented currents can flow on the via-hole walls, this structure is capable of containing mode patterns with
only vertical components of the electric field vector. Mode patterns with horizontal components of the
electric field vector have horizontal components of the current distribution on the via hole walls, which
will radiate. The SIW is thus ideally suited for containing the dominant mode pattern in a rectangular
waveguide. The thickness h of waveguide does not have an effect on the propagation constant. This gives
the designer certain degree of freedom in the choice of the SIW substrate.
The cut-off frequency of modes TEm0 can be derived using:
m
f cm0 ¼ pffiffiffiffiffi (1)
2aeq me
and the corresponded guided wavelength is:
l0
lg ¼ sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi (2)
2
er 1 ffc
This expression gives a good approximation of the field propagation in the SIW. This equation shows that
there are two important parameters: e that represents the dielectric constant inside the waveguide and a the
width of the metallic waveguide that is not defined here in the case of the SIW.
The modeling and design of SIW interconnects and circuits can be made on full-wave analysis tools,
either general purpose commercial packages or specialized numerical codes. Among the most common
techniques adopted for the design of SIW structures, there are the Finite-Difference Time Domain
(FDTD) method or the Finite-Difference Frequency Domain (FDFD) (Xu et al. 2003), the Boundary
Integral-Resonant Mode Expansion (BI-RME) method (Cassivi et al. 2002), the Method of Lines (MoL)
(Yan et al. 2005), and the Transverse Resonance Method (Deslandes and Wu 2006).
Using those different numerical methods, the propagating parameters of the dominant mode, like
wavenumber k or complex propagation constant g can be derived.
The SIW propagation constant of the mode TE10 can be obtained using a simulation software like HFSS
or CST or using measurement. To characterize the propagation and attenuation constants of the SIW line,
Page 5
DOI 10.1007/978-981-4560-75-7_57-1
simulations are conducted using the HFSS software. The permittivity and height of the substrate used are
respectively 2.94 and 0.762 mm in the calculations. The width of the line is 3.75 mm, p = 0.4 mm, and
d = 0.4 mm according to Fig. 2.
By considering two SIW lines with different length l2 and l1, and after generating simulated
S-parameters, phase constant b and attenuation constant a are calculated using:

∠ S 21, l ∠ S 21, l
b¼ 2 1
(3)
Dl
and

ln S 21, l1 = S 21, l2
a¼ (4)
Dl
and
Fig. 3 Phase constants of TE01 and TE20 modes simulated using a full wave simulator
Fig. 4 (a) E-field distribution of the first mode, (b) H-field distribution of the first mode, (c) E-field distribution of the second
mode, and (d) H-field distribution of the second mode
Page 6
DOI 10.1007/978-981-4560-75-7_57-1
g ¼ a þ ib (5)
Figure 3 shows the phase constant for the first and second modes; the second mode’s cut-off frequency is
twice that of the first mode just like the standard rectangular waveguide. The corresponding E and H field
distributions are illustrated in Fig. 4.
Equivalent Width
The SIW can be modeled by a conventional rectangular waveguide (RW) through the so-called equivalent
width aeq. This parameter is calculated such that the resulting dielectric-filled rectangular waveguide has
the same cut-off frequency of the fundamental TE10 mode as its corresponding SIW structure. This
determines the propagation characteristics of the TE10 mode. Physical parameters of via-holes d and p are
set to minimize the radiation (or leakage) loss as well as the return loss (Deslandes and Wu 2003). The
equivalent rectangular waveguide width can be approximated according to the geometrical parameters
illustrated in Fig. 2a as follows:
d2
aeq ¼ aSIW (6)
0:95p
The cut-off frequency of the first mode is defined by:

1
c d2
f c ðTE10 Þ ¼ pffiffiffiffi : aSIW (7)
2 er 0:95p
And for the second mode by:

1
c d2 d3
f c ðTE20 Þ ¼ pffiffiffiffi : aSIW (8)
er 1:1p 6:6p2
Loss Considerations
The energy in transmission may be lost or dissipated through different physical mechanisms including
dielectric losses, conductor losses, and radiation losses. Since the inner part of SIW is filled with a
dielectric material, an adequate choice of dielectric material and conductor quality can reduce the
contribution of the first two loss mechanisms (Bozzi et al. 2014).
• Attenuation because of the dielectric losses is related to tand of the dielectric substrate:
k 2 tan d
ad ¼ (9)
2b
• Attenuation because of the conductive loss:

Rs
ac ¼ 3
2bp2 þ a3 k 2 (10)
a bbk
with:
Page 7
DOI 10.1007/978-981-4560-75-7_57-1
0.5
αdiel+αcond
0.4 αrad
αdiel+αrad+αcond
α (Nepers/m)
0.3
0.2
0.1
0
25 30 35 40
Frequency (GHz)
Fig. 5 Attenuation constant versus frequency for an SIW structure

rffiffiffiffiffiffiffiffi
om0
Rs ¼ ðsurface resistanceÞ (11)
2s
The radiation leakage is caused by the gaps between the metalized via and slots. Radiation or leakage
leads to two consequences, namely additional signal losses and undesired interferences.
2:84 6:28
1 d p
1

aSIW aSIW d dB
aR ¼ rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 (12)
m
4:85 2aSIW
lg 1
In order to ensure that the synthesized waveguide section becomes radiationless or free from leakage loss,
parametric effects of p and d were studied on those issues (Deslandes and Wu 2003). To simplify the
analysis, dielectric and conductor losses are not considered, the loss solely comes from radiation. It is
found that the following requirements can be put forward to minimize the return and leakage losses, that
is, the diameter of hole should satisfy some geometric constraints:
lg
d< (13)
5
p 2d (14)
At millimeter and submillimeter frequencies, planar circuits usually suffer from radiation originating at
bends and discontinuities. 90 SIW and microstrip bend losses were analyzed (Djerafi and Wu 2012a) in
the 77 GHz band. The major part of losses in microstrip is related to the radiation compared to an
insignificant radiation loss in SIW bend.
Using Eq. 5 and the same lines used to estimate the phase propagation constant below, Fig. 5 shows the
attenuation versus frequency with only the contribution of the radiation loss and with the metallic and
dielectric loss. The dielectric loss represents the major part of the losses. The attenuation constant
decreases when frequency increases, especially the radiation loss.
Page 8
DOI 10.1007/978-981-4560-75-7_57-1
Table 1 SIW line configurations allowing for a spatial arrangement

Type H-plane SIW line Two plane SIW line C-type SIW line L-type SIW line
Structure
Circuit model
Principal SIW Parameters
Substrate Materials
Low-loss material is the foundation for developing high-performance integrated circuits and systems.
This becomes more critical for power budget as frequency increases to the millimeter-wave ranges and
beyond. This is because it is relatively difficult to amplify over those ranges. Thermal effect, dielectric
nonuniformity and metallic surface roughness may have to be taken into account for better and accurate
design. This is especially important for antenna developments. The SIW can theoretically be constructed
with any available substrate. The most used ones are Rogers RT/duroid ®5880 glass microfiber reinforced
PTFE composite and RT/duroid ®6002 for conventional PCB processing, which are easily sheared with
laser and machined to the required shape. The holes can easily be drilled mechanically into these
machinable materials compared to ceramics which can only be processed by the laser perforation and
other special techniques. All these materials have an excellent dimensional stability. Of course, a good
thermal stability of the material of choice should also be considered in the design. These selections will
not just affect the performances but also defines power handling capabilities (Cheng et al. 2008).
SIW Line Configurations

The SIW structures can be arranged in different configurations. Depending on applications in term of
allowed space, integrability, task and performance, one of them can specially be preferred. Table 1
summarizes some attention-grabbing configurations. The H-to E-plane interconnection shown in (b) is
very useful in the design of multipolarization antenna feeding. To realize the corner, a vertical line is
inserted into a horizontal line. The vertical and horizontal layers making up the corner were fabricated
using a conventional printed circuit board (PCB) process, before manually introducing the vertical layer
in the horizontal one (Djerafi et al. 2012a).
Comparing to other types of planar transmission lines, a large form factor of SIW is a bottleneck for its
integration with other conventional integrated circuits. Miniaturization techniques for SIW therefore
become very crucial in some designs (Ding and Wu 2008; Djerafi et al. 2012b). The proposed structures
present transversely folded waveguides in different shape, namely, C-type, T-type, and L-type. In 2005, a
2-layered T-type structure was proposed for the miniaturization of an SIW for the first time with SIW in
Page 9
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 6 Dispersion characteristics of different SIW lines
multilayered substrates (Grigoropoulos and Young 2004). Subsequently, C-type folding approaches as
described in column (b) were also implemented with SIW in multilayered substrates (Grigoropoulos
et al. 2005). L-FSIW was proposed (Doghri et al. 2013) to build a series of couplers. The cross section of
the L-FSIW is equivalent to 2-D transmission line with a shunt capacitance that models the effect of the
corner region.
For the C-type SIW line, the propagation constant is obtained by solving:
tan ðs þ d Þk x þ tan ða s d Þk x ¼ 0 (15)
and for L-type SIW line by:

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
pffiffiffiffiffi 0
bf10 ¼ ð2pf meÞ2 b 2 (16)
where
2 0
pffiffiffi 13
2t
6 B 2 1 , C7
, 2 6 B a C7
b10 ¼ , cot 1 68a, tf 2 melnB pffiffiffi C7 (17)
a 4 @ 2C f A5
In Fig. 6, dispersion characteristics of the dominant mode in the two types of folded waveguide are
compared with that of the conventional version of the original unfolded SIW (Ding and Wu).
All three lines have the same substrate filling. It is clearly observed that the cut-off frequencies of the
L-type SIW line and C-type SIW line are lower than the classic SIW line. The shift in frequency is due to
the effect added by the corners.
Characteristic Impedance
The impedance is an important characteristic especially when the SIW is connected to other planar and
nonplaner structures. One definition of impedance in a rectangular waveguide is the wave impedance.
This impedance, however, does not take into account the geometry of the transmission line. Another
impedance definition more often employed in the design of matching networks is the characteristic
Page 10
DOI 10.1007/978-981-4560-75-7_57-1
impedance. The characteristic impedance in a transmission line supporting a TEM wave can be defined in
one of the three following ways (D’Orazio 2004):
VV
Z PV ¼ (18)
2Pt
2Pt
Z PI ¼ (19)
II
V
ZV I ¼ (20)
I
In a rectangular waveguide, however, the choice of voltage and current is not unique and these definitions
do not produce the same results. The choice of which definition is more suitable depends on the targeted
applications in question. The three definitions of characteristic impedance in an SIW are given by:
2h
Z PV ¼ ZTE (21)
aeq
p2 h
Z PI ¼ ZTE (22)
8aeq
ph
ZV I ¼ ZTE (23)
2aeq
where ZTE is the wave impedance of the TE mode.
SIW Transitions
Transitions from SIW to microstrip or coplanar waveguide or from SIW to air-filled rectangular wave-
guide are necessary to couple high-Q waveguide components to active planar ones. In addition, these
transitions are necessary to be able to experimentally characterize the SIW components. They should have
the following properties, namely broadband, low loss, and minimum VSWR.
Although the design of transitions is specific to a particular geometry, some basic guidelines can be
appreciated. The design of broadband transitions, in general, requires that the level of the characteristic
impedances be similar (impedance matching) and that the field distributions of the coupling structures be
similarly oriented (field matching). The design of transitions between two dissimilar structures is
complicated by the excitation of higher order modes at the discontinuity.
Different waveguide-to-SIW transitions are detailed in the literature. Most of these transitions are based
on waveguide-to-microstrip or to-CPW (Cassivi et al. 2002; Deslandes 2010).
Microstrip/CPW-to-SIW Transition
The microstrip-to-SIW transition shown in Fig. 7 is based on the quarter-wave transformer principle. The
main parameters of this transition are the width of the SIW (aeq), the width of the microstrip taper (w) and
the taper length (l). aeq can be computed using Eq. 6. For each substrate having thickness h and
permittivity er, an optimum taper width is obtained by solving this equation (Deslandes 2010):
Page 11
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 7 Microstrip-to-SIW transition: (a) configuration and (b) example of a fabricated prototype
Fig. 8 GCPW-to-SIW transition: (a) configuration and (b) example of a fabricated prototype
8
>
> 60 h w w
>
< ln 8 þ 0:25 ; <1
1 h w h h
¼ (24)
we >
> 120p w
>
: ; >1
h½w=h þ 1:393 þ 0:667lnðw=h þ 1:444Þ h
0:627er þ1 er
1 4:38 e 1
þr p 1 ffiffiffiffiffiffiffiffiffiffi
¼
2 2
e 1þ12h=w
(25)
we aeq
For a given substrate (h and er) and SIW width (aeq), to find w, Eqs. 24 and 25 are equalized, which yields
the optimum taper width.
l is optimized to minimize the return loss over the full bandwidth, this length can be computed using
some analytical equations described in Lu (1997).
Patrovsky et al. (2007) proposed a CPWG-to-SIW transition using an E-field coupling. The coupling
slots are cut on the top surface of the SIW and placed next to the short circuit termination of the SIW. The
coupling slots act like a magnetic dipole antenna with a strong E-field across the slot in the center but
weaker at the end of the slot. In Li et al. (2009) Fig. 8, the sidewalls of the SIW are tapered along the
triangle-shaped coupling slot in such a way that the direction of the electric field on the coupling slot is
always perpendicular to the SIW sidewalls. This allows for a smooth transition. The tapered coupling slot
also serves as an impedance transformer to transform any arbitrary impedance line in SIW to the CPWG
port impedance exactly as the microstrip transition does.
To assure a single mode operation in the CPW, the following condition must be satisfied:
c
W þ 2S þ 2D < pffiffiffiffi (26)
2fmax er
W, S, and D are defined in Deslandes and Wu (2003), fmax is the maximum frequency of operation.
The optimal taper length is l4 which gives a wider bandwidth.
Page 12
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 9 Metallic waveguide-to-SIW transition using a fin line: (a) configuration and (b) simulated E-field magnitude
distributions by HFSS at 34 GHz, along the transition XZ plan and YZ plan
Fig. 10 WR10-to-SIW line transition using a stepped impedance: (a) configuration, (b) photograph of a back-to-back
transition
Rectangular Waveguide-to-SIW Transition

In (Xia et al. 2006), a radial probe extended from an SIW inserted into a height-tapered waveguide is used,
losses of about 2.5 dB are achieved at 33 GHz. In order to design the transition to a rectangular waveguide,
two step transformers are used in Moldovan et al. (2006) with parallelepiped alumina probe to improve the
scattering parameter, more than 3 dB of loss is measured in W band. In Li et al. (2009), a longitudinal slot
window etched on the broad wall of SIW couples the energy between SIW and rectangular waveguide.
This transition shows a return loss of less than 15 dB over a frequency range of 800 MHz at 35 GHz with
loss of 2.5 dB in measurement and more than 1 dB in the simulation. In Li et al. (2010), a novel transition
structure is realized by using a finline probe inserted into a WR-28 waveguide with an insertion loss of
1.4 dB. To measure this transition, two parts of the waveguide sandwich the substrate, which normally
increases the width of the metallic waveguide. In Djerafi et al. (2012c), a WR-28-to-SIW transition based
on ALTSA (Antipodal Linearly Tapered Slot Antenna) as illustrated in Fig. 9 is presented. A design
constraint was to make the transition without any modification on the waveguide. A back-to-back
transition has been fabricated, and measured results confirm performances of this transition. In Djerafi
et al. (2013), a transition from the WR10 waveguide to SIW structure has been designed and fabricated.
The proposed transition consists of a multistep impedance transformer milled into an aluminum block.
The transition is illustrated in Fig. 10 where the parameters to be optimized in this transition are the aspect
ratios and lengths of the steps.
Page 13
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 11 (a) SIW-waveguide transition, (b) a surface mountable pyramidal horn antenna
Nonresonant SIW Antenna

The SICs and related antennas technologies converge toward the development of system-on-substrate
(SoS) technology that allows for a complete and seamless integration of antennas and circuits. SIW
supports the design of different antennas involving various elements and feeds as detailed in Wu
et al. (2012). The more adaptable and popular antenna elements: slot (Wei et al. 2013) and patch backed
by cavity. To miniaturize the size of antennas, different techniques can be used in the design of slot array
antennas (Ding and Wu 2009, 2010). Printed antennas such as the microstrip patch, which might fulfill
design requirements in the microwave range, do not provide sufficient performances in the millimeter-
wave range. This is because of very high conductor losses arising from immense current densities
(singularities) at the strip edges, especially in resonant structures. Disadvantages of resonant elements
include also, the fact that it is usually narrowband, with bandwidths of a few percent being typical. Also,
the radiation efficiency of the microstrip antenna tends to be lower than some other types of antennas, with
efficiencies less than 70 % being typical.
An antenna operating in a traveling-wave configuration could provide the bandwidth and the efficien-
cies needed. Traveling-wave antennas are a class of antennas that use a traveling wave on a guiding
structure as the main radiating mechanism. On the other hand, leakage can also be generated with
enclosed waveguide structures with carefully designed geometrical asymmetry, open apertures, or slots.
Therefore, leaky-wave or surface-wave antennas can be developed for millimeter-wave applications
including tapered slot antenna (TSA), printed Yagi–Uda antenna or log-periodic dipole array, dielectric
rod antenna, long-slot leaky-wave antenna. SIW supports the design of different antennas involving
various elements and feeds as detailed in Wu et al. (2012). Over the last decade, various nonresonant
antennas in SIW technique have been studied, which will be presented in the following section.
Horn Antenna
Horns provide high gain, relatively wide bandwidth, and they are not difficult to make. There are three
basic types of rectangular horns. The rectangular horns are ideally suited for rectangular waveguide
feeders. The horn acts as a gradual transition from a waveguide mode to a free-space mode of electro-
magnetic waves. The radiation fields from aperture antennas, such as slots, open-ended waveguides,
horns, reflector and lens antennas, are determined from the knowledge of the fields over the aperture of the
antenna. The theory of waveguide aperture antennas is generally used to study the horns. However, phase
error occurs due to the difference between the lengths from the center of the feeder to the center of the horn
aperture and the horn edge.
Horn antennas are now commonly used for many applications including microwave communications,
feeds for reflector antennas, and radar elements. In Li and Wu (2004), an integrated H-plane horn antenna
Page 14
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 12 (a) Geometry of H-plane SIW horn, (b) electric field distribution simulated at 35 GHz, (c) and (d) the parameters used
in the analysis
was proposed. In Wang et al. (2010), a dielectric load was added to the SIW H-plane horn antenna in order
to have a narrow beamwidth in both E-plane and H-plane. In Wong et al. (2008), a broadside horn antenna
was implemented by the SIW technology.
E-Plane SIW Horn (Waveguide Transition)

The design of an appropriate SIW-waveguide transition passes through the design of a multiple stepped
waveguide quarter-wave impedance transformer. The transformer makes use of a multiple reflection
scheme. This transition can be used for designing a surface mountable horn antenna feed by SIW line. The
waveguide port of the transition is flared into a large opening and forms an E-plane horn.
The proposed transition consists of a multistep impedance transformer milled into an aluminum block.
The transition is illustrated in Fig. 11. The parameters to be optimized for this transition are the aspect
ratios and lengths of the steps. The detail of the design has been well established (Pozar 2011).
The impedance in each section of the quarter-wave transformer is defined by:
rffiffiffi
2 m
p hi
e
Z i ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
l0 2 (27)
8a 1 2a
where hi is the height of each section. The input impedance is defined by SIW line.
Since most of microwave integrated circuit modules need a metal housing, this structure can be
installed as part of an external housing. A prototype presented by Li and Wu (2004) has an HPBW of
15 in H-plane and 27 in E-plane and a cross polarization below 30 dB.
H-Plane SIW Horns

The horn antenna integrated into substrate is a waveguide aperture flared into a large opening on the
H-plane by keeping the perpendicular dimension constant. The radiation is focused in the H-plane. The
geometry is shown in Fig. 12.
Page 15
DOI 10.1007/978-981-4560-75-7_57-1
2
A
lh 2
¼ R0 þ ; (28)
2

A
ah ¼ arctan ; (29)
2R0
s
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

lh 1
Rh ¼ A aeq ; (30)
A 4
To have a single mode excitation, the width and height of the SIW line must satisfy:
l0 l0
pffiffiffiffi < aeq < pffiffiffiffi (31)
2 er 2 er
and
l0
h < pffiffiffiffi (32)
2 er
where l0 is wavelength in free space.

The design rules follow the same principles of the conventional horn. The corresponding pattern can be
computed using:

1 þ cos ðyÞ I y, j ¼ 0
F H ðy, j ¼ 0 Þ ¼ ðH plane patternÞ (33)
2 Iðy ¼ 0 , j ¼ 0 Þ

bb
sin sin y sin ’
1 þ cos ðyÞ 2
F E ðy, j ¼ 90 Þ ¼ ðE plane patternÞ (34)
2 bb
sin y sin ’
2
where
, , , ,
I ðy, ’Þ ¼ ejð 2b Þðb sin y cos ’þAÞ ½CðS 2 ÞjS ðS 2 ÞCðS 1 ÞjS ðS 1 Þ
R1 p 2
p 2 , , , (35)
þ ejð 2b Þðb sin y cos ’AÞ C t jS t C t þ jS t ;
R1
2 2 1 1
sffiffiffiffiffiffiffiffiffiffiffi
, 1 bA pR1
S1 ¼ R1 b sin y cos ’ (36)
pbR1 2 A
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

, 1 bA pR1
S2 ¼ R1 b sin y cos ’ (37)
pbR1 2 A
Page 16
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 13 Optimal gain for different lengths and apertures

sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
1 bA pR
t ,1 ¼
1
R1 b sin y cos ’ þ (38)
pbR1 2 A
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

, 1 bA pR1
t2 ¼ R1 b sin y cos ’ þ (39)
pbR1 2 A
C(x) and S(x) are Fresnel integrals, which are defined as

ðx ðx
p 2
C ðxÞ ¼ cos 2 t dt and S ðxÞ ¼ sin p2 t2 dt
0 0
The directivity is calculated by
4p H
D¼ et eph ðAhÞ (40)
l2
where
8
et ¼
p2
p2 2 2

eH
ph ¼ ðC ðp1 Þ C ðp2 ÞÞ þ ðS ðp1 Þ S ðp1 ÞÞ
64t

pffi 1
p1 ¼ 2 t 1 þ
8t

pffi 1
p2 ¼ 2 t 1 þ
8t
Page 17
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 14 SIW horn antennas (a) standard, (b) SIW horn with a dielectric lens (the lens has a constant permittivity, (c) SIW horn
antenna with dielectric lens (the lens has different permittivity), (d) SIW horn with a 2 block printed transition to the air, (e)
SIW horn antenna with a triangular grated transition and a superelliptical taper

1 A 2 l
t¼
8 l R0
Figure 13 shows an example of the gain obtained using HFSS for different lengths (R0) and apertures A.
The substrate used has a dielectric constant of 2.94 and a thickness of 2 mm (aeq/2)
For each length, there is an optimal aperture that gives the maximum gain. The figure demonstrates that
by increasing the length, the gain enhancement is less and less significant.
For a very thin substrate thickness with respect to propagation wavelength, the electromagnetic waves
will be confined within the substrate instead of radiating. The mismatch between the SIW horn and the air
increases which results in a poor bandwidth.
Matching Bandwidth and Radiation Improvements

One of the solutions to improve the horn performance is the insertion of lens in front of the radiating
aperture. Generally, dielectric lenses are constructed by extending the same dielectric used to build the
horn. In the lens-corrected horn, the lens acts as a focusing element, which increases the front to back ratio
and reduces the front phase error. When the substrate becomes thinner, generally below l60, the lens effects
become less relevant. The different developed topologies in the literature are shown in Fig. 14 as well as
the fabricated counterpart in Fig. 15.
SIW H-plane sectorial horn antenna is given in Fig. 15a. The rectangular waveguide and sectorial horn
antenna are integrated by using the same single substrate based on the SIW technology in Wang
et al. (2010). A reasonable length of 15 mm was determined to obtain an acceptable quadratic phase
error and single mode on the aperture. The simulated gain and side lobe level are 6.73 dBi and 12.08,
respectively; the simulated beamwidths of the E-plane and the H -plane are 180 and 48 , respectively.
Efforts for improving the impedance matching of planar horn antenna have been reported in recent papers.
Solutions based on integrated dielectric lens (Fig. 14b, c) or printed transition for SIW were proposed
(Fig. 14d, e).
In Wang et al. (2010) a dielectric slab is placed in front of the aperture of the horn. This slab serves as the
dielectric guiding structure in the H-plane. In the H-plane, for a horn with maximum gain, the aperture
phase distribution along the H-plane is nearly uniform without the dielectric loading. This antenna is
loaded by a dielectric elliptical slab with a length of 1.4 mm. In this case, the dielectric loading serves as
the phase corrector in the H-plane to compensate the quadratic phase error caused from the shortened
length. The gain is 5.75 dBi, the back sidelobe level is 11.78 dB and the beamwidths of the E- and H-plane
Page 18
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 15 Example of the fabricated horn antennas according to Fig. 14
are 178 and 63 , respectively. Different shapes of dielectric loading can be used (ellipse, rectangle,
hyperbola, and parabola). In Cao et al. (2014) a planar SIW horn antenna with a modified dielectric lens
was proposed, by drilling air vias in the substrate, different effective dielectric constants can be realized by
adjusting via diameter. Therefore, the SIW horn antenna loaded by a dielectric slab with gradually
decreasing dielectric constant is designed.
The presented solution in Esquius-Morote et al. (2013a) is focused on improving the reflection
coefficient of H-plane SIW horn antennas with additional parallel plate block. The horn aperture
practically behaves as a slot in an infinite ground plane with a front-to-back ratio (FTBR) close to
unity. The proposed printed transition does not significantly modify the radiation characteristics of a
conventional SIW horn. The main radiating element of an SIW horn with the proposed transition is indeed
the open-ended termination of the last parallel plates waveguide. Therefore, the differences in radiation
performances are mainly due to diffraction effects at the transition edges. Modifications of the transition
geometry have been presented in Esquius-Morote et al. (2013b) to improve the radiation performances of
SIW horn antennas (Figs. 14d and 15d).
The tapered ridge was proposed in Mallahzadeh and Esfandiarpour (2012), consisting of a simple
arrangement of vias on the side flared wall within the multilayer substrate. VSWR below 2.5 for the whole
frequency range of 18–40 GHz was realized by simulation.
Leaky-Wave Antenna
Leaky-wave antennas (LWA) are nonresonant, consisting of a series of perturbations on a metallic or
dielectric waveguide in question. The perturbations produce radiation since they cause energy to leak
away along the waveguide. Leaky-wave antennas have been studied for more than four decades,
exhibiting interesting features due to their traveling-wave nonresonant nature. Leaky-wave antennas
have the inherent property that the beam scans with frequency. For scanning applications, this is an
important advantage. Most of the early leaky-wave antennas were based on enclosed rectangular or NRD
waveguides, in which some kind of cuts or holes were introduced in order to produce the leakage of power
along the length of the waveguide. In the last two decades, the research was focused in the millimeter-
wave band, where new LWA designs have been proposed to overcome the technological problems
associated to these higher frequencies.
Leaky-wave antennas support a fast wave on the guiding structure, where phase constant b is less than
free-space wavenumber k0. The leaky-wave is therefore fundamentally a radiating type of wave, which
Page 19
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 16 Leaky-wave long slot antenna: (a) straight longitudinal slot, (b) sine squared meander pattern, and (c) Taylor power
distribution meander pattern
radiates or “leaks” power continuously as it propagates on the guiding structure (and hence the name of
the antenna). The operation is therefore quite different from a slow-wave or surface-wave type of antenna,
where radiation mainly takes place at the end of the antenna.
Depending upon their architecture, leaky-wave antenna can be categorized as either uniform or periodic
or quasi uniform periodic. Uniform antennas are homogeneous along the length of the guiding structure.
Periodic structure has regular perturbations along their length. They both are derived from specific
guided-wave counterparts and follow the same rules of formation of leaky wave. These two types can
be constructed on the top wall (broad wall) or lateral wall (narrow wall). Another distinction that can be
made is between a one-dimensional (1D) leaky-wave antenna and a two-dimensional (2D) leaky-wave
antenna.
Uniform Leaky-Wave Structure
Long Slot Leaky-Wave Antenna

Figure 16 shows different configurations of SIW uniform leaky-wave structure. Usually, a fast wave
(b < k0) travels in enclosed guided-wave structures such as rectangular metallic waveguide and SIW. A
physical cut that disturbs the guided wave along the longitudinal propagation direction will produce a
leaky wave because the condition of forming a leaky wave has been fulfilled in this case. The long slot on
the broadside of waveguide has to be precisely machined with a specific meandering profile in order to
synthesize the desired radiation properties. Cheng et al. (2011) introduces a design process of SIW long
slot leaky-wave antenna.
Using substrate with permittivity er, the beam direction y at the operation frequency is defined as:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2ffi
l0 c
y ¼ arcsin ¼ arcsin er (41)
lg 2aeq f
aeq, the SIW equivalent width is generally selected to ensure the monomode operation.
As shown in Fig. 17, the scanning range of uniform leaky-wave antennas using a standard air-filled
waveguide is about 12 with 4 GHz of frequency sweep around 35 GHz. The variation of pointing angle in
the same bandwidth increases when the SIW line is used, with permittivity of 3. The scanning angle is
three times in SIW compared to the hollow waveguide. In Cao et al. (2014) planar beam scanning
substrate integrated waveguide (SIW) slot leaky-wave antenna (LWA) is proposed for enhancing scan-
ning range and gain flatness using a modified composite right/left-handed transmission line (CRLH TL)
structure.
It is assumed that the leaky-wave antenna is terminated with load which absorbs a fraction R of input
power. Generally, the length L is considered equivalent to 20 wavelength to ensure efficiency of 95 %
(R = 0.05). The slot width should be around 0.08 lg.
Page 20
DOI 10.1007/978-981-4560-75-7_57-1
80
εr=1
70
εr=2.2
Pointing angle (Deg)

60 εr=3
50
40
30
20
33 34 35 36 37
Frequency (GHz)
Fig. 17 Scanning range of uniform slot with different permittivity
Different meander contour can be considered other than the straight longitudinal slot; sine squared
plotted in Fig.16b is defined by in Whetten and Balanis (1991)
p
x ¼ 0:5 þ 0:2sin2 z a (42)
L
The slot starts at waveguide center (x = 0.5 a at z = 0) and returns at the center at the end of the slot.
Different contours can be defined regarding the distribution of the radiated power. This type of meander
contour provides the necessary aperture field taper for superior antenna far-field amplitude performance.
Since the power within the waveguide is continuously lost along the slot, the ideal meander pattern must
account for both the amount of power to be radiated at a given point along the slot, and the amount of
power remaining in the guide. Thus, at the leading edge of the slot (nearest to the generator), the slot can
remain near the centerline of the guide and still radiate sufficient power. Farther along the slot, as the
power level decreases within the guide, the slot must meander closer to the sidewall in order to radiate the
desired power level. The exact meander contour is usually very difficult to determine. Starting from the
required amplitude distribution (Taylor (shown in Fig. 17c), Tchebechev, sinusoidal, etc.), the radiation
per unit length a is defined by:
1
jAðzÞj2
aðzÞ ¼ ðL 2 ð (43)
z
1 2 2
AðzÞ dz AðzÞ dz
1R 0 0
Then the relation between offset x, and radiation per unit length a should be determined. Most of previous
works determined the radiation properties of a long slot leaky-wave antenna by a numerical analysis of the
fields in the slot. In Cheng et al. (2011), Ansoft HFSS was used to determine radiation per unit length, and
normalized phase propagation per unit length, according to different offset. In Scharp (1982) the relation
is defined as:
Page 21
DOI 10.1007/978-981-4560-75-7_57-1
20
Taylor distribution
10
Sin contour
0
Gain (dBi)
–10
–20
–30
–40
–50
–90 –60 –30 0 30 60 90
Angle (Degree)
Fig. 18 Simulated patterns of leaky-wave antenna

vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
z
u
u
a u h lc 3 a
xðzÞ ¼ arcsinu
u2d ll L z
L (44)
p u g ð
t
1 L aðyÞdy
0
d for the width of the slot and h is the substrate thickness, the illumination being a function of the distance
z divided by the length L.
Figure 18 gives simulated radiation patterns for an LWA antenna with Sin contour and Tayler
distribution. Dielectric with 2.2 of permittivity and 0.5 mm of thickness with length of 115 mm (20 lg
at 35 GHz) is used in the two designs. The antenna with Taylor distribution shows higher gain, narrower 3-
dB beam-width and better sidelobe levels. However, the sin contour slot is very simple to design and in
general its performance is still acceptable.
Lateral Slot Leaky-Wave Antenna

Using its half-mode (HM) version, the SIW structure offers another advantage when used in the design of
leaky wave structures. When only the mode TE10 is propagating in SIW waveguide, the maximum value
of E-field is located at the vertical center plane along the propagating direction and the center plane can be
considered as an equivalent magnetic wall. As a result, the half-mode substrate integrated waveguide
(HMSIW) can be simply achieved by dividing the SIW into two equal parts. HMSIW was proposed by
Hong et al. (2006). In an SIW structure, the ratio of width a to height b is usually very large when an SIW
is cut into two half waveguides along the center plane, the resulting open aperture between the top and
bottom planes is thus very close to a magnetic wall. The dominant mode in such a structure is just like one
half of the dominant mode TE10 in a complete SIW. In the area close to the aperture, the field is
out-diffused a little bit to satisfy its boundary conditions. On the other hand, an HMSIW can also be
regarded as a half of microstrip line, which is terminated by a short-circuited plane in the middle. The first
higher order mode of the microstrip line has been demonstrated to be an efficient leaky-wave line source.
Usually an electric wall is inserted in the middle plane of a microstrip line to suppress the dominate mode
(quasi-TEM mode). Interestingly, the working mechanism of a half microstrip line excited by its first
high-order mode is exactly the same as that of an HMSIW. Since the leaky-mode is the fundamental mode
in the HMLWA, the excitation becomes easier, and complicated feeding structures are not needed
anymore. A leaky-wave antenna based on the HMSIW technique was proposed and investigated in Xu
Page 22
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 19 (a, b) Configuration of HMSIW LWA, (c) a fabricated prototype
Z
a Z b Z c
Theta Theta
Theta
X Y X Ph1 Y
Ph1 X Y
Ph1
Fig. 20 Radiation pattern of HMLWA with different thickness at 35 GHz: (a) 5 mil, (b) 10 mil, (c) 20 mil
et al. (2008). The antenna is characterized by many attractive advantages, including compact size, easy
fabrication, low cost, low loss, wide bandwidth, and direct integration with planar circuits.
The proposed LWA topology is shown in Fig. 19, to improve the return loss, a section of HMSIW with
gradually tapered width from W2 to W1 is adopted. L is the antenna length, b is the thickness of substrate
and the dielectric constant of the used substrate is 2.2, W 2 is the width between the metal via holes and
leakage aperture (also called aperture depth when seen from antenna angle), and W 0 is the width between
the metal via holes and the other edge of waveguide (input HMSIW lines). The width of additional
dielectric is designed less than l/4 to avoid the undesired channel modes.
The far-field radiation pattern of this antenna indicates a quasi omnidirectional characteristic, which is a
hollow cone centered on the axis. The beam will then be in the form of a conical beam that opens along the
negative z axis at an angle y that increases with frequency, approaching a donut-shaped pattern at
broadside. The angle of the cone can be controlled using the thickness as shown in Fig. 20. As the used
substrate is thicker as the aperture of the cone is smaller.
Periodic Structures
This type of leaky-wave antenna consists of a slow-wave structure that has been modified by periodically
modulating the structure in some fashion. A typical example is a rectangular waveguide that is loaded with
a dielectric material (so that the TE10 mode is a slow wave) and then modulated with a periodic set of holes
or slots, as shown in Fig. 21. Many of the features common to 1D periodic leaky-wave antennas may be
appreciated by consideration of this simple (but practical) structure. The antenna can be designed to
radiate from backward to forward. In rectangular waveguide, (Hyneman 1959) shows a closely spaced
slotted traveling wave array. All slots have the same length, the waveguide does not taper, and the
Page 23
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 21 (a) Configuration of periodic slot LWA, (b) example of array architecture
waveguide is filled with air. Dion (1958) describes relationships among parameters within nonresonant
slotted arrays.
It is assumed that the relative permittivity of the filling material is sufficiently high so that the TE10
mode is a slow wave over the frequency region of interest. This will be the case provided
2
p
er > 1 þ (45)
k 0 aeq
According to Floquet’s theorem, the infinite numbers of space harmonics of guided mode inside the
periodic LWA are related to,
bn ¼ b0 þ 2np=p (46)
The first kind of leaky-wave antennas provides radiation into the forward quadrant and can yield scanning
from broadside to forward end-fire directions. The scanning range for periodic leaky-wave antennas
reaches from backward end-fire through broadside directions into a part of the forward quadrant. The
dominant mode on the former type represents a fast wave, while the latter type is a slow wave structure. As
a result, the dominant mode on periodic leaky-wave antennas does not radiate and radiation is achieved by
using one of its space harmonics.
The first higher-order radiating harmonic b1 inside the dielectric-filled rectangular waveguide for
which (bn/k0) 1 is related as:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
b1 pffiffiffiffi c nl0
¼ er 1 pffiffiffiffi (47)
k0 2af er d
Dispersion behavior inside the LWA is explained through the attenuation constant a and the propagation
constant b. An angle of radiation f0 is a function of space harmonic bn. If b1 < 0, antenna radiates in the
backward direction, as frequency is increasing b1 > 0, antenna radiates in the forward direction.
pffiffiffiffi
1 er b1
f0 ¼ sin (48)
k0
The distance d0 corresponding to this condition is
l
d0 ¼ pr2 (49)
1 þ jcos f0 j
Page 24
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 22 The reflection cancelation forward-wave LWA
Since the slots cause a perturbation to the TE mode of the SIW, to suppress the reflection, the slot lengths
are tapered linearly at both ends as shown in Fig. 21. The slot length is transformed from zero to slot length
and keep the other parameters of the slots constant. To enhance the bandwidth, the SIW width can be
tapered as a cosine function at both ends.
The periodic frequency scanning antenna generally has a poor radiation efficiency around broad side
due to the existence of open stop band in the narrow frequency band. The reason is that the constructive
addition of all series or shunt admittances along the transmission line forces input impedance to become
zero. The addition of reflection cancelation stub at the distance of lg/4 from excitation slot avoids the
undesired open stopband. The dual slot structure was proposed in Yamamoto et al. (2002), but the
structure is of broadside nature only. The periodic array of eight radiating transverse slots is embedded
on the broad wall of the guiding structure. The distance (S1) between the excitation slot and reflection
cancelation slot is approximately one quarter of a guided wavelength (lg) as shown in Fig. 22. The total
phase shift between the two slots is approximately 180 . Hence, the reflected waves from the two slots are
added in out-of-phase resulting in a very low reflection coefficient at the input. The slot pairs are
periodically spaced at 1 lg, hence progressive phase is multiple of 2p, resulting in the fact that all the
slot pairs are added in phase.
One-port LWA, steering from backward to forward directions including the broad side, is studied and
implemented in the SIW technology. The azimuth plane scanning of multibeam antenna (MBA) is
obtained through the phase control of array of nine 1-D elevation plane frequency scanning LWAs,
each fed through the vertical corner on the array port contour of a Rotman lens. The complete antenna
prototype fabricated on single Rogers RT/Duriod 6002 substrate with er = 2.94 and 0.508 mm thickness
is simulated through Ansoft HFSS.
The periodic frequency scanning antenna has poor radiation efficiency because of the existence of an
open stopband around broad side in the narrow frequency band. In Guntupalli (2014), SIW LWA using the
concept of reflection cancelation was developed. Directional pattern characteristics were studied and
demonstrated from 37 to +18 by changing frequency from 58 to 78 GHz at an interval of 1 GHz.
A matched termination was designed, validated and integrated to realize single port LWA. A 2-D scanning
multibeam conformal array was realized by utilizing a Rotman lens BFN feeding nine LWAs in the
azimuth plane for E-plane steering and natural frequency scanning property of 1 8 LWA for H-plane
steering. The whole 3-D circuit is constructed after fabricating nine LWAs and then inserted into the slots
made on the Rotman lens BFN (Fig. 23). The antenna array has seven inputs for beam switching, four
dummy ports for sidewall absorption and nine LWAs on the contour of the Rotman lens. The lens is
Page 25
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 23 Two-dimensional scan conformal multibeam array antenna (Figure from http://creer.grames.polymtl.ca/ressources/
realisations/)
Fig. 24 Open periodic waveguides
synthesized to distribute the desired amplitude to the radiating elements located on the output ports of the
Rotman lens contour.
Laterally Open Periodic Waveguides

The scheme of a lateral periodic open SIW LWA is shown in Fig. 24. This configuration is similar to the
HMLWA with the addition of a second row of conducting vias separated a distance P, which behaves as an
inductive partially reflective surface (PRS) (Martinez-Ros et al. 2010) and allows the propagation of the
fundamental mode in an SIW. In the SIW LWA introduced by Deslandes and Wu (2005), there is a lateral
wave that radiates from the open end of the truncated substrate. The radiation mechanism was the basis for
identifying additional limitations originating from Snell’s law at the dielectric-air interface. In general
case, the periodic modulation of the conductors that gives the periodic LWA a larger scanning range and a
more flexible design scheme than the uniform leaky-wave antenna. The lateral LWA radiates in the form
of a space leaky-wave created in the discontinuity between the grounded microstrip lateral edge and the
extended dielectric-filled ground plane as any periodic LWA.
Antipodal Tapered TSA
TSA
The Tapered Slot Antenna (TSA) consists of a tapered slot cut in a thin film of metal with or without an
electrically thin substrate on one side of the film as shown in Fig. 25. The slot is narrow toward one end (s)
for efficient coupling to devices such as mixer diodes. Away from this region (Lant), the slot is tapered and
Page 26
DOI 10.1007/978-981-4560-75-7_57-1
a W
W0
y
b c
a
Lant
Fig. 25 Differently shaped tapered-slot antennas: (a) Nonlinear: Exponential (Vivaldi, ETSA), Fermi, Parabolic (PTSA), Dual
Exponentially (DETSA), (b) Linear (LTSA) and (c) Constant Width Slot
a traveling wave propagating along the slot radiates in the end-fire direction (w0). Antenna is in principle a
transition from a guided wave to an unbounded wave or vice versa. Looking at typical TSA, this definition
is quite intuitive. As the width of the slot increases, the characteristic impedance increases as well
providing smooth transition to the free space characteristic impedance of 120p O. The antenna was first
proposed by Lewis et al. (1974). Prasad and Mahapatra (1979) first introduced the linearly tapered slot
antenna (LTSA) in 1979. Their antenna of about one free space wavelength long was etched on a piece of
25-mil alumina substrate. In the same year, Gibson (1979) reported a TSA with exponentially tapered
profile, also known as – Vivaldi antenna, which demonstrated a bandwidth of 8–40 GHz.
An SIW-based antipodal LTSA was demonstrated in Hao et al. (2005) and width corrugated version in
Djerafi and Wu (2012b). In the antipodal version of the TSA, the metallization on either side of the
substrate is flared in opposite directions to form the tapered slot. When the SIW waveguide is used to feed
the antipodal TSA, the bandwidth limitation caused by balun can be removed and, thus, wideband
characteristics are indeed obtainable. The SIW TSAs have many advantages such as low profile, low
weight, wideband characteristics, and they are easy to integrate with other planar devices. The TSA's
antennas are very sensitive to the thickness and the permittivity of their supporting substrate. They can
exhibit higher cross-polarized lobes in the (E or H)-plane and sidelobe in the H-plane. This is not only
very critical in element design but affects considerably the array performances. The different material,
antenna shape, as well as mutual coupling have been widely studied for one side TSAs unlike the SIW
antipodal version (Fig. 26).
In the antipodal version, a balanced microstrip feed is used in a double-sided arrangement for the
antenna. When SIW is used to feed the antipodal TSA (Fig. 27), the bandwidth limitation caused by balun
Page 27
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 26 SIW TSA
2.8
2.7
2.6
2.5
W0 (aeq)
2.4
2.3
2.2
2.1
2
0 1 2 3 4 5 6 7 8 9 10 11 12
Permittivity
Fig. 27 SIW TSA end-fire slot width (w0) versus the used material permittivity
can be removed. The performance of the antenna is determined both by its geometry and by the thickness
and permittivity of the supporting dielectric substrate.
Slot-line field components are not confined to the substrate alone. They extend into the air regions
above the slot and below the substrate also. Therefore, the energy is distributed between the substrate and
the air regions. Consequently, the effective dielectric constant for a slot-line (ere) is less than the substrate
permittivity er. Galejs (1962) has shown that the zeroth-order value of er for a slot on an infinitely thick
substrate is the average dielectric constant of the two media, that is, the aperture can be calculated using
the following equation
c
w0 ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi (50)
flow ðer þ 1Þ=2
To cover the entire monomode bandwidth, the lower frequency is equal to the SIW first mode cut-off,
which gives
pffiffiffiffi
2aeq er
w0 ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi (51)
ðer þ 1Þ=2
Page 28
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 28 Electric field at various cross sections along the transition from SIW waveguide to end-edge of the antenna
120
SIW h=1 mm
Stripline h=1 mm
100
SIW h=0.5 mm
Stripline h=1 mm
80
Impedance (Ω)
60
40
20
0
10 20 30 40 50 60
Frequency (GHz)
Fig. 29 SIW and stripline impedance
Figure 27 shows that it is possible to reduce the width of the antenna efficiently using dielectric with
permittivity lower than 4.
Balun
The tapered antipodal finline smoothly transforms the incident TE10 (Fig. 26 cut A) waveguide mode to a
quasi microstrip mode which is propagating in the area of overlapping antipodal fins of the transition
(Fig. 26 cuts B and C). E field in the transition area concentrates and rotates by almost 90 until waveguide
mode transforms to quasi slot mode (Fig. 26 cut D). The E field transformation process in different cut
planes of the transition is shown in Fig. 28. Smooth transformation of the field mode allows the transition
to operate in a wide frequency band. The region II (plane BB’ to CC’) act as balun and impedance
transformer. In addition, region II transforms the impedance of the TE10 waveguide mode to high
impedance of slot line component.
Figure 29 shows the SIW waveguide impedance according to the Eq. 21 and the impedance of the
overlapping fins at Cut BB’ considered as parallel plat line. When the thickness of SIW changes, the
cut-off frequency remains the same and the impedance changes. The parallel plat impedance is higher than
its SIW counterpart, when the overlapping section diminishes, the impedance increases ensuring a
wideband operation. The limit of width of the overlapping section is the half of the SIW width..
Page 29
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 30 TSA antenna with corrugation
Corrugation
TSA width w should be larger than 2l. A degradation of radiation pattern has been observed for a narrower
TSA width (Schaubert 1993). This degradation in the radiation pattern is a significant problem for the
effective design of compact TSAs. Pattern improvement for a narrow width antenna can be achieved by
using a corrugation structure, which makes this technique suitable for the formation of antenna arrays
where a small spacing between antenna elements is needed. A Fermi TSA antenna with a corrugation was
proposed in Sugawara et al. (1998) and in Sato et al. (2003). This technique was verified with SIW in
Djerafi and Wu (2012b). It has been successfully demonstrated that the use of corrugations along the sides
as shown in Fig. 30 can reduce the antenna width and improve the VSWR over a wide frequency range, as
well as suppress the side lobe levels. The corrugation structure is used to alter the phase of currents
flowing along the outer edge of the substrate, changing the orientation of the electric field at the edge of the
antenna substrate.
To explain this improvement, simulated E-field distributions along the antipodal LTSA structures with
and without corrugation are shown in Fig. 31. The E-field counterparts in the transverse plane over the
radiation region are described. In the case without corrugation, the electrical field orientation at the edge of
the antenna substrate is opposite to that in the antenna aperture. Some reference claims that the alternation
of strips and slots behaves like a slow-wave structure that increases effective aperture size resulting in
lower reflection of the radiated wave. As observed in Fig. 31 this phenomenon is not observed. The
current distribution illustrated in Fig. 32 clearly shows that the energy is more confined in the central
region and radiation from the side is limited.
The effective antenna aperture is thus increased by the presence of the corrugation, which ensures a
plane wave phase front over the antenna aperture. In standard antenna, the antipodal nature of the antenna
gives rise to very high levels of cross-polarization, particularly at high frequencies because the skew in the
slot fields is close to the throat of the flare. As shown in the field distribution in the transverse view in
Fig. 31, the field in the center region is more highlighted in the structure without corrugation. Out of this
perturbed region, the field is parallel to the two sides of the substrate compared to the one in the antenna
without corrugation which is arranged in an arc.
Physical dimension of the slot corrugation is constrained by the fabrication process. In fact, the smallest
possible slot width, feasible under this process, is 236 um in this work. The slot lengths are selected to be
around lr/4, where lr is the wavelength in the dielectric substrate. The SIW antipodal LTSA is optimized
Page 30
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 31 Simulated E-field magnitude distributions obtained by HFSS at 28 GHz, along the antenna and in transverse cut (a)
antipodal LTSA without corrugation, (b) antipodal LTSA with corrugation
Fig. 32 Simulated current distribution comparison between corrugation and no-corrugation structures
around 28 GHz. Substrate RT/Duroid 6002 with thickness h of 508 um and dielectric constant of
er = 2.94 is used.
Figure 33 shows simulated pattern results of antennas with and without corrugation. Without corruga-
tion, the E-plane pattern beamwidth is excessively large and the SLL in H-plane is 12 dB. Once the
corrugation is used, a noticeable decreasing of 3 dB beamwidth can be observed in the E-plane, while the
Page 31
DOI 10.1007/978-981-4560-75-7_57-1
Without corrugation With corrugation

0
–5
–10
Normalised Gain (dB)

–15
–20
–25
–30 E-Plane
H-Plane
–35
CrossPolar
–40
–90 –60 –30 0 30 60 90
Angle (Degree)
Fig. 33 Simulated radiation patterns with and without corrugation obtained by HFSS at 28GHz
Table 2 Typical properties of end fire traveling antenna

Type Maximum gain Low sidelobes

Directivity (dB) 10 log 10 lL0 10 log 4 lL0
qffiffiffiffi qffiffiffiffi
Beamwidth 55= lL0 77= lL0
beamwidth in the H-plane increases. The sidelobe in the H-plane for this antenna is 23 and 33 dB in the
E-plane. The cross-polarization level with corrugation is found to be better than 27 dB at the frequency of
design and without corrugation is 15 dB.
The impact of the slot width is verified for its dimensional variation from 6 to 12 mil. A small effect on
the SLL level is observed without gain degradation. For the slot depth, the variation affects the gain of the
antenna by changing the effective aperture. It is possible to optimize the corrugation slot to achieve the
same beamwidth in both E- and H- planes.
Maximal Gain Versus Length

Generally, directivity increases as the length, L, of a TSA is increased. For lengths between three and eight
wavelengths, the increase is linear according to Formula in Table 2 defined in Yngvesson et al. (1985). In
this table, the tapered dielectric rod antennas are classified by Zucker as low sidelobes traveling wave.
Figure 34 shows the studied linear and constant with TSA antennas. The CWTSA is a kind of tapered
slot antenna for which the width of the radiating slot is constant over the antenna length. Since the slot line
that feeds the antenna is generally narrow, there should be a tapering section between the feeding slot and
the radiating slot in order to achieve a better impedance matching. If such a transition is not used, there
could be a high return loss due to the abrupt change in the characteristic impedance.
Figures 35 and 36 show the gain variation with antenna length. In the LTSA version the gain increases
linearly between 1 l0 and 3 l0 and saturates after 14 dBi. The real constraint with this version is the angle a
(according to Fig. 25). The taper angle in the case of the LTSA version is: 5 2a 15 . If the length
increases, the antenna will be out of the acceptable angle range after a specific value. The linear
dependence of the gain on L/l0 is an important characteristic of all end-fire traveling-wave antenna. For
Page 32
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 34 (a) Linear taper, (b) CWTSA with linear taper
20
18
16
14
Gain (dBi)
12
10
8
6
Gain
4
High Gain
2 Low Sidelobes
0
0 1 2 3 4 5 6 7 8
Length (λ0)
Fig. 35 Effect of antenna length variations on gain of LTSA
CWTSA, the typical numbers are gain of 12–18 dBi with length of the antenna L in the range of 3 l0 to
8 l0. For longer antennas, the coefficient multiplying L/l0 decreases somewhat.
Figures 37 and 38 show the effect of length variations on E- and H-plane patterns for LTSA and
CWTSA. The CTWSA shows that the beam width and gain essentially following the high gain standard
curve for length below 5 l0. In the TSA, the E-plane beam with curve has a very steep slope for length of
4–5 l0, exactly as standard curves. After 3 l0, the beamwidth in the E- and H-plane follow the low side
lobe curve. While TSA elements have many features in common with the general classpffiffiffi of traveling-wave
antenna, there are some differences, for example, the beamwidth often follows (1= L) while the E-plane
varies more slowly with L. The E- and H- bandwidths follow the low side lobe type curve in the case of
LTSA and high gain curve for the CWTSA version.
Substrate Thickness
There is a trade-off in substrate thickness; a too “thick” substrate launches surface modes, which degrade
the antenna performance, whereas a too “thin” substrate yields higher cross-polarizations and less
symmetrical beam patterns. The optimum substrate thickness h has experimentally been defined by
Yngvesson et al. (1985) as a function of factor, teff :
pffiffiffiffi
teff ¼ bð er 1Þ=l0 (52)
Page 33
DOI 10.1007/978-981-4560-75-7_57-1
21
19
17
Gain (dBi)
15
Gain
13
Hight Gain
Low Sidelobes
11
9
2 3 4 5 6 7 8 9 10 11 12
Length (λ0)
Fig. 36 Effect of antenna length variations on gain of CWTSA
140
E-Plane
130
H-Plane
120 High Gain
110 Low Sidelobes
HPBW (Degree)
100
90
80
70
60
50
40
30
20
0 1 2 3 4 5 6 7
Length (λ0)
Fig. 37 Effect of antenna length variations on E- and H-plane patterns for LTSA
An acceptable range of teff for a good TSA opeacceptable range of teff for a good TSA operation is ration is
determined to be between 0.005 and 0.03.
Considering the criteria of 3 dB gain bandwidth, the CWTSA gain is still in this bandwidth from teff of
0.015–0.053 as shown in Fig. 39. The width of the bandwidth is still in the same order of Yngvesson range
(in this simulation er is fixed and thickness modified). With a smaller length, the LTSA gain variation
shows a larger teff range (Fig. 40). This condition is not applicable to antipodal SIW TSAs, which is less
subject to substrate mode. LTSA is designed using substrate with 2.94 of relative permittivity and
0.508 mm of thickness, which is equivalent to teff = 0.034 at 28 GHz. Obviously, this value is out of
the Yngvesson range. The designed antenna still works adequately in both simulation and measurement.
To examine the possibility of using material out of the defined range, two different thicknesses are tested
here: t = 0.05 mm and t = 1.25 mm. This is equivalent to teff of 0.0034 and 0.101, respectively. Figure 41
illustrates E-plane radiation patterns of the two selected structures. For the two substrate, SLL is still low
and an acceptable gain is achieved. These results demonstrate the possibility to design the antipodal
Page 34
DOI 10.1007/978-981-4560-75-7_57-1
70
E-Plane
H-Plane
60
High Gain
Low Sidelobes
50
HPBW (Degree)
40
30
20
10
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Length (λ0)
Fig. 38 Effect of antenna length variations on E- and H-plane pattern for CWSA
20
18
3dB
16
Gain (dB)
14
12
10
6
0 10 20 30 40 50 60 70
teff (x1000)
Fig. 39 Simulated gain obtained by HFSS for CWTSA at 28 GHz versus substrate thickness
corrugated TSA with materials out of Yngvesson’s range. In this chapter, new criteria will be established
for the SIW antipodal version.
Prototypes
Several TSA designs have been reported; as shown in Fig. 25. The tapered slot section which constitutes
the radiating region can take on any of three geometric profiles:
– Nonlinear taper (Vivaldi, tangential or parabolic),

– Linear taper (LTSA),
– Constant width (CWTSA).
Vivaldi
The Vivaldi antenna has been studied since its first appearance. It undergoes many improvements in the
last three decades, including introducing corrugated edges to form a better radiation pattern (Sugawara
Page 35
DOI 10.1007/978-981-4560-75-7_57-1
12
11.5
11
Gain (dB) 10.5
10
9.5
0 20 40 60 80 100 120 140
teff (x1000)
Fig. 40 Simulated gain obtained by HFSS for LTSA at 28 GHz as a function of substrate thickness
–5
–10
–15
–20
–25
–30 b=2.50mm
b=0.05mm
–35
–90 –60 –30 0 30 60 90
Angle (degree)
Fig. 41 Simulated radiation pattern obtained by HFSS at 28 GHz in connection with substrate thickness of 0.05 and 2.5 mm
et al. 1998), or changing its beam width by changing the shape, length, dielectric constant, and dielectric
thickness. As one of the improvements, antipode Vivaldi antenna was proposed (Kim and Chang 2004).
The exponential taper profile is defined by the following equation:
yðxÞ ¼ AeRx þ y0 (53)
where R specifies how the slot is widening, parameter A is determined by the length L and also the width
w0 of the taper and parameters y0 is equal to L-Lant. Very wide band performance can be acquired using an
antipodal Vivaldi antenna with its easy wideband transition from microstrip line to radiation part of the
antenna (Yang 2008).
Dual Exponentially Tapered Slot Antenna (DETSA), also known as Bunny-Ear Antenna (BEA), is an
end-fire traveling wave antenna used for UWB applications. This antenna is a special form of Tapered Slot
Antenna (TSA). It is created by tapering the outside edge of the slotline conductor of a Vivaldi antenna.
The outer slot taper of the DETSA is described by an exponential function. This adds additional antenna
design degrees of freedom.
Page 36
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 42 (a) Top and bottom photograph of the fabricated SIW corrugated antipodal LTSA (L = 25.6 mm, aeq = 5 mm,
w = 8.44 mm), (b) example of array using LTSA
Fermi
The designed feed antennas are based on the TSA concept of Sugawara et al. with a Fermi taper. It
provides symmetric radiation patterns and low side lobe level in both E-and H plane, The Fermi-TSA
opens out the shortest while the Vivaldi opens out with the longest extension and the linear TSA is in the
middle. The Fermi-TSA profile is defined by the following:
w0
yðxÞ ¼ (54)
1 þ eRðxbÞ
where R and b are variable. The antipodal version was suggested first in Gazit (1988).
The Parabolic
The inner side is shaped as a parabola
yðxÞ ¼ Ax2 þ y0 (55)
The parabolic type opens out the shortest while the Vivaldi opens out with the longest extension and the
Fermi-TSA linear is in the middle. Consequently, the beamwidth is narrowest for the PTSA and wide for
the FTSA and widest for Vivaldi (Wang et al. 2009).
LTSA
The linearly tapered slot antennas (LTSA) have been reported for years (Simons et al. 1992; Yngvesson
et al. 1985, 1989; Nesic 1991; Wu et al. 1998). Due to their salient features such as narrow beam width,
high element gain, wide bandwidth, and small transverse spacing between elements in arrays, it attracts
lots of interests in many applications. Practically, the performance of the LTSA is mainly determined by
the feeding system. The angle of the aperture should be lower than 15 .
In this section, different types of the SIW TSA antenna are fabricated and tested.
Page 37
DOI 10.1007/978-981-4560-75-7_57-1
0
Simulated
–5 Measured
–10
–15
S11 (dB)
–20
–25
–30
–35
–40
–45
20 25 30 35 40
Frequency (GHz)
Fig. 43 Measured and simulated return losses
–5
–10
–15
–20
–25
–30
E-Plane
–35 H-Plane
Cross Polar
–40
–90 –60 –30 0 30 60 90
Angle (degree)
Fig. 44 Measured radiation patterns at 28 GHz
LTSA
SIW antipodal LTSA is optimized to work around 28 GHz for LMDS or future 5G systems. Substrate
RT/Duroid 6002 with thickness b of 0.508 mm and dielectric constant of er = 2.94 is used. Parameters of
the optimized structure are shown in Fig. 42. Simulated and measured return losses of the proposed
antenna are shown in Fig. 43. The return losses are greater than 15 and 10 dB in the simulation and
measurement, respectively. Such an observed degradation is probably caused by the mismatch effect of
transitions.
E-plane and H-plane patterns of the antenna as well as the cross-polarizations at 28GHz are measured
and shown in Fig. 44. The achieved gain is 12.2 dBi against the simulated 11.9 dBi. This difference is
within the calibration-related tolerance range of the antenna reference in used anechoic chamber. The 3 dB
beamwidth is 59 in the E-plane and 48 in the H-plane and the first primary side lobe is located at
22 dB. It can be seen that the measured results are in good agreement with the simulated ones. The cross-
polarization level shows a 4 dB degradation compared to the simulated counterpart.
Page 38
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 45 Top and bottom photograph of the fabricated SIW corrugated antipodal Fermi-TSA (L = 65.61 mm, w = 6.8 mm,
aeq = 3.8 mm)
0
S11 Simulated
−5 S11 Measured
−10
−15
−20
S11 (dB)
−25
−30
−35
−40
−45
30 31 32 33 34 35 36 37 38 39 40
Frequency (GHz)
Fig. 46 Simulated and measured S-parameter characteristics of the SIW corrugated Fermi-TSA
Corrugated Fermi
A single corrugated antipodal Fermi-TSA has been designed to cover the band of interest around 35 GHz
for imaging array applications. Substrate RT/Duroid 6002 with thickness of 0.508 mm and dielectric
constant of er = 2.94 is used. To realize the desired gain, antenna length L1 has been increased to reach the
value of 8 l0. The antenna width has been slightly modified due to the restrictions related to the design of
antenna arrays where the width of the antenna is 0.79 l0. The photograph of the fabricated SIW corrugated
Fermi-TSA (top and bottom) is shown in Fig. 45.
Simulated and measured reflection coefficients versus frequency are presented in Fig. 46. It can be
noticed that the antenna bandwidth covers the entire considered frequency band of 28.57 % at 35 GHz for
both measured and simulated results that are obtained by the use of HFSS.
Figure 47 presents simulated and measured copolar E-plane and H-plane radiation patterns. From these
results, it can be seen that there is a good agreement between the measured and simulated radiation
patterns. The 3 dB beam widths are approximately 25.33 and 28 for both E-plane and H-plane patterns,
respectively. The side lobes are lower in the E-plane compared to the H-plane. At 35 GHz, the antenna
gain is about 19.4 dBi based on simulations and almost 18.4 dBi is obtained in practice according to
measured results. The cross-polarization level is found to be better than 20 dB thanks to the overlapped
flaring metal in the SIW version. The excellent radiation properties and compact size of this antenna make
it ideal as either a standalone antenna with a broad pattern or as an array element.
Page 39
DOI 10.1007/978-981-4560-75-7_57-1
0
E−plane Simulated H−plane Simulated
E−plane Measured H−plane Measured
−5
−10
Amplitude (dB)
−15
−20
−25
−30
−35
−90 −60 −30 0 30 60 90
Angle θ (degree)
Fig. 47 E-plane and H-plane radiation pattern performances of the SIW corrugated Fermi-TSA at 35 GHz
Fig. 48 Photography of the fabricated: (a) planar corrugated PTSA and (b) corrugated L-Folded PTSA
Corrugated SIW Planar and Folded Antipodal Parabolic Tapered Slot Antenna
A folded tapered slot antenna is presented in Doghri et al. (2013), this antenna represents a three
dimensional version of Parabolic Tapered Slot Antenna. The PTSA is folded along its longitudinal center
line which gives the possibility to have 45 and 45 polarizations.
Planar PTSA
To design an L-folded TSA, the first step is to design a planar TSA with respect to the required half power
beamwidth, side lobe level, gain, etc. over all the desired operating frequency bandwidth. The geometrical
parameters such as width, dielectric thickness, taper profile, length, and so on are optimized to meet
requirements. Several TSA designs have been reported. For the demonstration purpose of the L-folded
TSA PTSA (Parabolic Tapered Slot Antenna) is used in this design to obtain an optimal matching and
bandwidth. The corrugation is also used for effectively reducing the width of the TSAs without degrading
the antenna performances. The fabricated planar prototype is shown in Fig. 48, radiation and matching
performances in Figs. 49 and 50 successively.
Folded SIW Line

An SIW line folded along its longitudinal center line is used to properly feed the L-folded TSA (LFSIW).
Figure 51 illustrates a cross-sectorial view of SIW and LFSIW with the transverse electric and magnetic
field lines of the fundamental TE10 mode. The junction acts as a tapered matching section between the two
Page 40
DOI 10.1007/978-981-4560-75-7_57-1
Simulations Measurements
0
E−Planes
–5 Cross Polar
H−Planes
–10
Normalized Gain (db) –15
–20
–25
–30
–35
–40
80 60 40 20 0 20 40 60 80
Angle (degree)
Fig. 49 Simulated and measured radiation patterns of the corrugated PTSA at 35 GHz
−5
−10
⏐S11⏐(dB)
−15
−20
−25
−30 Non Folded ATSA

Folded ATSA
−35
30 32 34 36 38 40
Frequency (GHz)
Fig. 50 Measured return loss of the corrugated APTSA and corrugated L-Folded APTSA at 35 GHz
waveguide types. The taper is optimized to achieve the minimum reflection coefficient. The optimized
length of the L-Junction is about 1.41 lg at the center frequency. The back-to-back structure layout is
illustrated in Fig. 52 with the taper shape. The back-to-back configuration is used to measure perfor-
mances of the junction. The fabricated prototype is shown in Fig. 52b. The fabrication processes involve
laser micromachining permitting arbitrarily shaped perforations, long and large rectangular vias with
rounded corners that define SIW walls. After machining the structure, vias are electroplated. A back-to-
back configuration is used to measure the performance of the junction because it was not possible to
measure directly the L-shaped SIW line.
Figure 52a shows the measured reflection and transmission coefficients. The back-to-back junction
achieves a return loss of better than 28 dB and an insertion loss of better than 0.9 dB over the 30–40 GHz
bandwidth. The quality of the design is demonstrated by the E-field distribution inset in Fig. 52.
Page 41
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 51 Geometry of (a) SIW and (b) LFSIW
0
Port1 Port2
−10
S-Parameters (dB)
Alignment Posts
−20
−30
−40 S21
S11
−50
30 32 34 36 38 40
Frequency (GHz)
Fig. 52 (a) Back-to-back L-junction with measured results and simulated E-field magnitude distributions obtained by HFSS at
35 GHz, along the junction. (b) Assembled back-to-back L-junction
Folded PTSA
The third design step consists in folding the planar TSA. The previously introduced L-junction is used to
feed the L folded PTSA. The main geometrical parameters are not altered by the folding step. The
fabrication of this antenna involves two PCBs that are illustrated with dimensions in Fig. 53. The
fabricated prototype is shown in Fig. 48b. Two posts are used for alignment purposes. When assembled,
the two corners are metalized for a proper electrical connection between the two PCBs.
Similar performances are observed when the proposed L-folded antenna is compared with the
nonfolded counterpart. Figure 54 shows the measured and simulated radiation patterns in the E and
H-planes, corresponding to the 45 and 45 planes. The measured gain is 11 dBi while the simulated
gain is 13 dBi. The 3 dB beamwidth is 38 in the E and H-planes. The cross polarization is 10.7 dB in
simulation and 8.1 dB in measurement. The folded version has a better symmetry in the E and H-planes.
However, the gain and the cross polarization are reduced.
Coupling
Mutual coupling between elements is an important issue in designing antenna arrays. It modifies
impedance, radiation pattern, beamwidth, and directivity of an array, and even degrades the performance.
Therefore, it is a parameter that must be taken into account during the design process. The previously
described antenna topology has been studied.
Using HFSS simulations to evaluate the isolation levels between two adjacent elements (Fig. 55). The
isolation is about 13 dB between the elements aligned along their E-plane and separated by 0.68 l0. The
dielectric is subtracted in the central region and the isolation between the two elements is 3 dB lower. To
explain the coupling effect and a possible improvement, simulated E-field distributions are shown in
Fig. 56 along the antipodal Fermi-TSA structures with and without the substrate removal. The effective
antenna aperture is thus increased by such a removal, which ensures a plane wave phase front over the
Page 42
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 53 Geometry of the corrugated folded TSA with Lant = 24.5 mm, t1 = 0.76 mm, aSIW = 4 mm, L_slot = 13.4 mm,
W_slot = 0.76 mm, wt = 9.3 mm, ht = 1.6 mm, t2 = 1.2 mm, lc = 1.23 mm, wc = 0.13 mm, h = 0.76 mm. (a) top and (b)
bottom view of main PCB, (c) top and (d) bottom view of inserted PCB
Simulations Measurements
0
H-Planes
E-planes
−5
Cross Polar
−10
Normalized Gain (dB)
−15
−20
−25
−30
−35
−40
80 60 40 20 0 20 40 60 80
Angel (degree)
Fig. 54 Simulated and measured radiation patterns of the corrugated L-folded PTSA at 35 GHz
Page 43
DOI 10.1007/978-981-4560-75-7_57-1
−10
−15
S21 (dB)
−20
Standard antenna elements

Air gaped antenna elements
−25
30 31 32 33 34 35 36 37 38 39 40
Frequency (GHz)
Fig. 55 Simulated isolation between the antennas
Fig. 56 Simulated mutual coupling for standard two element Fermi-TSA and for air gaped two element Fermi-TSA
Page 44
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 57 Simulated E-field magnitude distributions obtained by HFSS in transverse cut (a) case1: square configuration, (b)
case2: V configuration; (c) case3: X configuration
antenna aperture. As shown in the transverse view of the field distribution in Fig. 56, the field in the center
region is more highlighted in the structure without the subtraction. In the modified antenna, the field is
more spread along the transverse section. In fact, the air gap creates discontinuities and the field is
confined along the air gap. When used in the two element array, the air slotted version gives more than
1 dB of improvement. The gain of one simulated Fermi-TSA element in the presence of a second passive
element is 2 dB lower than that of an isolated element. In the H-plane, as the field is confined around the
material, the coupling is very small between two elements disposed in the parallel arrangement.
Circular polarization can be ensured in wide bandwidth with high gain performances based on LTSA
and sequential rotation. This element can then be integrated into a subarray cell to determine the optimal
geometry. The four elements are physically rotated relative to each other and the feed phase is individually
adjusted for each arrangement as shown in Fig. 57. In the figure, the field distribution in the transverse
view when just one port is excited is illustrated. The mutual coupling effect changes with configuration,
the first cell shows a better gain performance. The coupling is very high between two elements disposed in
the parallel arrangement as in V configuration and directly affect the center of the parallel element. The
total gain is in the same level of one element. The 3rd case is the complement of the first one to build a
larger antenna array. The field distribution in the three cases gives some idea about the expected
asymmetrical radiation pattern.
Application to Imaging
Passive millimeter-wave radiometer for imaging has found applications primarily in the security and
defense industries. It is desired because of its ability to image through typical obscurants such as smoke,
dust, and materials like plastic and textile, in certain frequency ranges. Certain frequency windows present
low atmospheric losses (centered at 35, 94, 135, and 220 GHz). Many designers typically build passive
sensors to operate within these windows in order to exploit low atmospheric attenuation and increased
image contrast compared to lower frequencies.
Page 45
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 58 (a) Topology of the SIW dual polarization system, (b) photograph of the fabricated SIW dual polarization system
−10
S-Parameters (dB)
−20
−30
S11-Simulated
−40 S21-Simulated
S11-Measured
−50 S21-Measured
−60
30 31 32 33 34 35 36 37 38 39 40
Frequency (GHz)
Fig. 59 S-parameter characteristics of the prototyped antennas
The antenna architecture for imaging application should satisfy defined conditions in term of high gain
and low side lobe level (SLL). In addition, low cost and light weight characteristics are always required for
highly integrated, reliable and portable systems. In fact those systems are generally duplicated to achieve a
better resolution, which increase the weight and total volume.
Thanks to the intrinsic shielding of the proposed SIW feeding, a compact feed can be obtained without
resorting to any coupling in the feeding network that would deteriorate isolation and therefore increase
cross-polarization. Different antennas based on TSA element with orthogonal polarization are presented
in the following. An array of 128 elements shows low SLL and height gain with just 200 g of the total
weight is used to build image.
Dual-Polarized end-Fire Antenna

Progressive plane wave can be decomposed into two linearly polarized waves formed orthogonal vectors;
which are also orthogonal to the direction of propagation. In the case of the radiometry these orthogonal
components are referred to as the vertical and horizontal polarization. The natural radiation emitted by the
human is unpolarised, but after interaction with certain objects around it becomes polarized. One of the
Page 46
DOI 10.1007/978-981-4560-75-7_57-1
−5
−10
Amplitude (dB)
−15
−20
−25
Co-polarization
−30 Cross-polarization
−35
−90 −60 −30 0 30 60 90
Angle (θ)
Fig. 60 Radiation pattern performances at 35 GHz
a 0
−5 b
Simulated S21
Port2 Measured S21
−10 Simulated S11
S11 & S21 (dB)
Measured S11
−15
−20 Port1
−25
c
−30
−35
28 30 32 34 36 38 40
Frequency (GHz)
Fig. 61 (a) Measured and simulated S-parameters of the proposed twist, with E-field magnitude distribution simulated at
35 GHz, (b, c) shows the fabricated SIW 90 twist front and back view respectively
challenging issues in utilizing such full polarimetry is the need of a good discrimination between the H-
and V- polarization, which finally leads to an accurate imaging synthesis. Therefore, achieving good
isolation between the H- and V- polarization port, along with the important antenna parameters including
antenna gain, impedance matching, co- and cross- polarization discrimination, etc., is crucial. Three
different configurations for building dual polarization TSA are presented in this section.
A Dual Polarization Fermi Tapered Slot Antenna Setup Using E-Plane Corner
Compared to other types of tapered slot antennas, the corrugated Fermi-ATSA provides almost the same
E-plane and H-plane patterns, presenting at the same time low side lobe levels. Two 45 and 45 vertical
arm rotated E-plane bends are arranged to feed two Fermi antennas in posture of 90 , as illustrated in
Fig. 58a. The photograph of the fabricated prototype is shown in Fig. 58b.
Measured and simulated S-parameters of the dual polarization setup are shown in Fig. 59 (El Khatib
et al. 2012a). The isolation between the two ports is almost better than 20 dB in simulation over the entire
32–40 GHz band and in the measured one. The measured performance matches well with the simulation.
Figure 60 presents measured copolar (E-plane) radiation patterns of the design at the 0 , 90 planes angle,
the two ports are fed in phase simultaneously. The E field amplitude in the 0 plane is the geometrical
Page 47
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 62 SIW dual-polarized end-fire antenna (a) assembling scheme (b) fabricated prototype
0
−5
−10
−15
S-Parameters (dB)
−20
−25
−30
−35
Measured S21
−40 Measured S11
Measured S22
−45 Simulated S11
Simulated S21
−50 Simulated S22
−55
32 33 34 35 36 37 38
Frequency (GHz )
Fig. 63 Measured and simulated return losses and isolations
addition of the E field in the two planes. In counterpart, the 90 plane is the subtraction of the two. The
radiation curves show some little dissymmetry due to the presence of the other element. Difference
between the E field at 0 and 90 plane is 22 dB in the broadside direction.
A Dual Polarization Fermi Tapered Slot Antenna Setup Using Twist

In the second configuration, two PTSAs are disposed perpendicular to each other which give us the
vertical and horizontal polarization. To bring the signal of the vertical antenna to the horizontal an SIW
90 twist is used (Doghri et al. 2013).
The SIW 90 twist consists of a progressive deformation of an L-Shaped SIW line shown in Fig. 61.
The width is progressively decreasing in one plane and increasing in the other. The signal from the
horizontal plane will move to the vertical one and vice versa. To validate the proposed device, a prototype
operating over the Ka-band (26.5–40 GHz) was designed based on RT/Duroid 6002 substrate with
dielectric constant of er = 2.94 and substrate thickness of 0.76 mm. The fabricated prototype involving
two LEGO-like interconnected PCBs is shown in Fig. 61b, c. Figure 61a illustrates the measured
Page 48
DOI 10.1007/978-981-4560-75-7_57-1
Measurements Simulations
0
−5
−10
−15
−20
−25
−30
−35
X-Polar E-Plane
−40
H-Plane
−45 X- Polar H-Plane
E-plane
−50
80 60 40 20 0 20 40 60 80
Angle (degree)
Fig. 64 Simulated and measured radiation patterns corresponding to the vertically polarized antenna (fed by the twist) at
35 GHz
Fig. 65 (a) Geometry of the E-plane junction with aSIW = 4 mm, Wslot = 0.762 mm and Lslot = 15.2 mm, (b) E-field
magnitude distribution simulated at 35 GHz
performance of the proposed twist, a reflection coefficient of less than 18 dB and an insertion loss better
than 1 dB over the Ka-band is achieved.
Two high gain and broadband corrugated PTSAs are used as radiating elements. Some slots and holes
are used for alignment purposes. The fabricated prototype is shown in Fig. 62, two K connectors are added
to feed the access. The measured and simulated return losses and isolations are shows in Fig. 63. The
structure achieves a return loss better than 10 dB and an isolation higher than 32 dB from 32 to 38 GHz
(22 % bandwidth centered at 35 GHz). Due to the shielding of the structure, the interference in the feeder
is minimized to a negligible level.
Simulated and measured radiation characteristics of this dual-polarized antenna when one port is
excited while the other port is matched are shown in Fig. 64. The vertically polarized antenna fed by
the twist achieves a gain of 14.44 dBi at 35 GHz while the horizontally polarized antenna achieves a gain
of 14.28 dBi. The two patterns have a cross-polarization of better than 27 dB in the 0 direction. A small
asymmetry is observed in the co- and cross-polarization patterns because of the presence of the other
elements. This solution shows a better isolation than the first solution with corner.
Page 49
DOI 10.1007/978-981-4560-75-7_57-1
b 0
−10
a
S-Parameter (dB)
−20
−30
−40
Simulated S11
−50 Measured S11
−60
30 32 34 36 38 40
Frequency (GHz)
c −2
Simulated S31
d 190
Simulation
S-Parameters (dB)
−2.5 Simulated S21 Measurement
Phase Imbalance
Measured S21
185 d
−3 Measured S31
180
(°)
−3.5 175
−4 170
30 32 34 36 38 40 30 32 34 36 38 40
Frequency (GHz) Frequency (GHz)
Fig. 66 Measured and simulated: (a) photograph of the fabricated prototype, (b) input return losses, (c) transmission
coefficients (d) output phase differences
Fig. 67 (a) and (b) E-field magnitude distribution when feeding the horizontal and vertical antennas, (c) fabricated prototype
Four Elements Dual-Polarized end-Fire Antenna

Compared to two-element configuration, four elements will give a better symmetry between the two
vertical and horizontal planes. The horizontal polarization is obtained thanks to the use of 3-port E-plane
junction. Figure 65 illustrates the proposed E-plane junction. The proposed E-plane junction consists of a
progressive deformation of a T-shaped junction to rectangular waveguide with TE20. The resulting mode
is then divided in two arms to feed two TE10 mode SIW lines with equal amplitude and 180 phase
difference.
As depicted in Fig. 65, the E-plane junction divides and rotates the mode signal of the input SIW line in
H-plane to the mode signal of the two output SIW lines in E-plane. An experimental prototype operating
over the frequency range of 30–40 GHz was designed using RT/Duroid 6002 substrate from Rogers
Corporation, having substrate thickness of 0.76 mm. The equivalent synthesized bilateral walls of the
SIW are rectangular metalized slot arrays with width 0.762 mm (the minimum is defined by the
metallization process, in presented case 0.254 mm).
The photograph of the fabricated prototype is shown in Fig. 66a. The prototype was measured using a
vector network analyzer and a Line-Reflect-Line (LRL) calibration to de-embed the connectors, transi-
tions and 90 bend effects. As shown in Fig. 66b, c, return loss and insertion loss of, respectively, less than
Page 50
DOI 10.1007/978-981-4560-75-7_57-1
−10
S-Parameters (dB)
−20
−30
−40 Simulated S11

Simulated S21
Simulated S22
−50 Measured S22
Measured S21
Measured S11
−60
30 32 34 36 38 40
Frequency(GHz)
Fig. 68 Measured and simulated return losses and isolations
a 10
Measured Measured Simulated Simulated
X-Polor E-Plane E-Plane X-Polar E-Plane E-Plane
0
−10
−20
−30
−40
−50
−90 −70 −50 −30 −10 10 30 50 70 90
Angle(degree)
b 10
Measured Measured Simulated Simulated
X-Polor E-Plane E-Plane E-Polar E-Plane E-Plane
0
−10
−20
−30
−40
−50
−90 −70 −50 −30 −10 10 30 50 70 90
Angle(degree)
Fig. 69 35 GHz simulated and measured E-plane radiation patterns (a) the horizontally polarized array and (b) vertically
polarized array
Page 51
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 70 (a) and (b) Simulated E-field magnitude distribution obtained by HFSS at 35 GHz, along the unequally distributed
1/16 and 1/8 principal power dividers, (c) fabricated 1/16 power dividers
16 and 0.46 dB, are observed from 30 to 40 GHz. In addition, as shown in Fig. 66d, the E-plane junction
provides a 180 phase difference with phase imbalance of less than 3 over all the frequency range.
The proposed junction is used to feed four PTSA elements. The prototype shown in Fig. 67 involves
three Lego-style PCB blocks accurately aligned using posts and slots.
Figure 68 shows the measured and simulated return losses and isolations. In this case, ports 1 and
2 correspond to the vertical and horizontal ports, respectively. The structure achieves a return loss better
than 10.6 dB and isolation better than 26 dB from 30 to 40 GHz (28 % bandwidth). Figure 69 compares
simulated and measured radiation patterns of the horizontally and vertically arranged arrays, respectively.
The horizontally polarized array achieves a gain of 14.6 dBi at 35 GHz while the vertically polarized array
fed by the E-plane junction achieves a gain of 14 dBi. Thanks to the symmetry of the array arrangement, a
good symmetry of the radiation patterns is observed.
Fermi Tapered Slot Antenna Array for Imaging Applications

A Villeneuve distribution is used for sidelobe control in the proposed 8 by 16 array (El Khatib
et al. 2012b). Figure 70a shows the 1/16 power divider, the 4-stage H-plane SIW power divider makes
use of an equal and unequal-split T-junction power divider to achieve the required aperture distribution.
H-plane bends are used to compensate different delays generated by the imbalanced power splitting. The
equal power division in the fourth stage of the power divider represents a compromise that is needed due
to the tight dimensions that are dictated by the element spacing. All transmission coefficients fluctuate
around the ideal coupling factor in the band of 30–40 GHz. Similar to the previous 1/16 power divider, an
8-way T-junction power divider is designed in order to accommodate the feeding for eight samples of the
above mentioned 1 16 linear array, thus forming the proposed integrated 8 16 planar antenna array.
Page 52
DOI 10.1007/978-981-4560-75-7_57-1
a b
0 E-plane Measured
H-plane Measured
−10
Amplitude (dB)
−20
−30
−40
−50
−60
−90 −60 −30 0 30 60 90
Angle θ (degree)
Fig. 71 (a) Photograph of the proposed three-dimensional architecture of 128 element antenna array, (b) E-plane and H-plane
radiation pattern performances of the 8 16 SIW corrugated Fermi-TSA planar array at 35 GHz
Fermi element described in the SIW TSA section is selected to ensure initial high gain and broadband
characteristics for the array. At 35 GHz, the simulated gain is about 19.4 and 18.4 dBi is obtained in
measurement. The mutual coupling between the elements of antenna arrays must be taken into account
during the design process. The previously described antenna topology has been studied to evaluate the
isolation levels between two adjacent elements along their E-plane.
The isolation is about 13 dB between the elements aligned and separated by 0.68 l0. The dielectric is
subtracted in the central region and the isolation between the two elements is 3 dB lower. In the modified
antenna, the field is more spread along the transverse section. In fact, the air gap creates discontinuities
and the field is confined along the air gap. When used in the two element array, the air slotted version gives
more than 1 dB of improvement. The gain of one simulated Fermi-TSA element in the presence of a
second passive element is 2 dB lower than that of a single element. The coupling is very small between
two elements disposed in the parallel arrangement.
Figure 71 illustrates this novel planar array composed of 8 16 antenna elements of corrugated
antipodal Fermi-TSA, and fed with an SIW power divider. A 45 rotated E-to-H-plane interconnect
ensures, with eight 1 16 and one 1 8 power dividers, the construction of one block feeding network
for 128 antenna element array. The whole antenna is fabricated with standard PCB process (7 PCB
plaques). The gain of the planar array is 27 dBi, and the SLL is better than 26 dB in both E-plane and
H-plane as shown in Fig. 71b, the beamwidth is about 5.15 in the E-plane and 6.20 in the H-plane. The
total weight of the array is less than 200 g, showing an important advantage of SIW technology in
payload’s design efficiency.
Imaging System
There is a number of different ways to obtain a passive radiometric image from a scene (Goldsmith
et al. 1993). Phased arrays have been limited in their success at millimeter wavelengths due to difficulties
designing efficient phase shifter as well as relatively large feed system. Mechanically scanned systems
have been employed for different imaging applications in the millimeter range, but suffer the very major
handicaps of expensive, relatively unreliable mechanical systems together with low data rate from the
single receivers utilized. Another way of acquiring an image is to implement a FPA (Focal Plane Array),
which uses an array of independent detector elements in the focal surface of an imaging lens or reflector.
Recently, moderate-cost systems using small FPAs combined to mechanical scanning have been built
allowing real-time imaging (Doghri et al. 2011).
Page 53
DOI 10.1007/978-981-4560-75-7_57-1
Fig. 72 Heated square object imaged by three different antenna configurations with: (a) single element, (b) linear array, (c)
planar array
For passive imaging system, the antenna is used together with a very sensitive receiver to detect and
provides measurements of the electromagnetic radiation intensity received within a solid angle subtended
by the antenna beam. It is necessary to accommodate the antenna in this system design with high gain, low
side lobe level and low cross-polarization. Array antenna schemes can be used to accomplish these goals.
However, they present a number of problems in connection with cost, weight, and fabrication complexity
of the feed networks. Various discontinuities within a feed network should be avoided or used with
caution since noise emitted or interference created by such components into the receiver could desensitize
and confound the radiometric measurement. Images issued from a simple receiver composed of a front-
end circuit, which has two prime tasks, namely, the input frequency-band selection and the amplification
of the received signal into a proper level for the detector. A novel SIW prototyping approach was used to
Page 54
DOI 10.1007/978-981-4560-75-7_57-1
build up the receiver. This approach consists of SIW subcircuits realizing simple functions and intercon-
nects as LEGO building blocks to build up a complex function out of basic elements. Putting the SIW
transmission lines in front of each other with sufficient pressure passes the electromagnetic wave through
with a negligible loss.
The signal from the detector is fed to the analog to digital (A/D) input of the microcontroller and then
sent to the computer through a USB port and processed by the data acquisition software developed on the
Matlab platform.
The tests were done using respectively one element, a planar array composed by 16 elements and a 2D
array composed by 128 elements shown in Fig. 71 shows the three images corresponding to the three
scenarios. By increasing the antenna gain, the resolution is increased as shown in Fig. 72. The 2D array
gives a closer image of the square as it presents a higher gain and a better beam symmetry in the two planes
compared to one element and the planar array. The image resolution can be further improved by increasing
the number of array elements or by increasing the gain of each element.
Conclusion
It is well recognized that the antenna is one of the most important system components that limit or enhance
system performance, depending on the design of such a component. Generally, antenna elements cannot
be conveniently integrated in chip-set because either they are too large or the required performance such
as efficiency cannot be achieved by integrated components. In some cases, they could simply be
considered as part of the package that embeds the chip-set, which may be of importance for millimeter-
wave system design. In most high-gain antenna applications, array geometries are always required that
involve both radiating elements and feed network. Special feed networks in the form of beam-forming
networks present the engine for the development of multibeam, beam-scanning and beam-agile systems.
In this chapter, nonresonant SIW antenna structures well suited for a wide range of applications over
millimeter-wave and terahertz frequency bands are presented and discussed.
Different dielectric loaded H–plane SIW horn antennas are presented. Through a proper choice of the
structure and parameters, the maximum gain for the given size or the minimum size for a given gain may
be achieved.
Two different kinds of leaky-wave structures have been discussed in this chapter. One is related to
uniform guided-wave structures while the other consists of an array of periodic guided-wave structures.
They share common features but also exhibit many interesting differences in connection with electrical
and mechanical aspects. They exhibit low losses, very flexible radiation characteristics, and excellent
mechanical simplicity. Tapered slot antennas based on SIW feed with different shapes are presented with
simulation and measurement results. Configurations which ensure dual polarization and applications of an
array on imaging system are detailed. Other type of SIW-based nonresonant antenna can be fed by SIW or
integrated to the SIW antenna like Yagi, Lenses, Rod.
Advances in thick film technology have been made such as photoimageable and photoetchable
conductor pastes, which have enabled system-in-package (SiP) techniques to be demonstrated well into
the up-millimeter-wave frequency range (500 GHz, for example). LTCC technology has been known to
well synergize with SIW techniques since it offers multilayered 3D passive integration. Proposed LEGO-
style blocks fabricated and assembled with easy-to-connect and manipulate PCB pieces have been studied
and demonstrated in different components as well as for imaging system. This technique will play a
critical role with LTCC and PCB fabrication processes.
Page 55
DOI 10.1007/978-981-4560-75-7_57-1
Cross-References
▶ Beam-scanning Leaky Wave Antennas
▶ Broadband and Multiband Planar Antennas
▶ Millimeter-Wave Antennas and Arrays
▶ Phased Arrays
▶ Transmission Lines
▶ Waveguide Slot Array Antennas
References
Bozzi M, Georgiadis A, Wu K (2011) Review of substrate-integrated waveguide circuits and antennas.
IET Microwaves Antennas Propag 5:909–920
Bozzi M, Pasian M, Perregrini L (2014) Modeling of losses in substrate integrated waveguide compo-
nents. In: International conference on numerical electromagnetic modeling and optimization for RF,
microwave, and terahertz applications (NEMO), Pavia, 14–16 May 2014, pp 1–4
Cai Y, Qian Z-P, Zhang Y-S, Jin J, Cao W-Q (2002) Bandwidth enhancement of SIW horn antenna loaded
with air-via perforated dielectric slab. IEEE Antennas Wirel Propag Lett 13:571–574
Caloz C, Jackson DR, Itoh T (2011) Leaky-wave antennas. In: Frontiers in antennas: next generation
design & engineering. McGraw-Hill, New York
Cao W, Chen ZN, Hong W, Zhang B, Liu A (2014) A beam scanning leaky-wave slot antenna with
enhanced scanning angle range and flat gain characteristic using composite phase-shifting transmission
line. IEEE Trans Antennas Propag 62(11):5871–5875
Cassivi Y, Perregrini L, Arcioni P, Bressan M, Wu K, Conciauro G (2002) Dispersion characteristics of
substrate integrated rectangular waveguide. IEEE Microwave Wirel Compon Lett 12(9):333–335
Cheng YJ, Wu K, Hong W (2008) Power handling capability of substrate integrated waveguide inter-
connects and related transmission line systems. IEEE Trans Adv Packag 31(4):900–909
Cheng YJ, Hong W, Wu K, Fan Y (2011) Millimeter-wave substrate integrated waveguide long slot leaky-
wave antennas and two-dimensional multibeam applications. IEEE Trans Antennas Propag
59(1):40–47
D’Orazio William (2004) Study of substrate integrated waveguide circulators for millimeter wave
applications. In: Mémoire présenté en vue de l'obtention du diplôme de maîtrise ès sciences appliquées
(génie électrique)’Thèse – École polytechnique
Deslandes D (2010) Design equations for tapered microstrip-to-substrate integrated waveguide transi-
tions. In: International microwave symposium, Anaheim, 23–28 May 2010, pp 704–707
Deslandes D, Wu K (2003) Single-substrate integration technique of planar circuits and waveguide
components. IEEE Trans Microwave Theory Tech 51:593–596
Deslandes D, Wu K (2005) Substrate integrated waveguide leaky-wave antenna: concept and design
considerations. In: Microwave conference proceedings, APMC 2005. Asia-pacific conference pro-
ceedings, Suzhou, pp 346–349
Deslandes D, Wu K (2006) Accurate modeling, wave mechanisms, and design considerations of a
substrate integrated waveguide. IEEE Trans Microwave Theory Tech 54(6):2516–2526
Ding Y, Wu K (2008) Miniaturization techniques of substrate integrated waveguide circuits. In: IEEE
MTT-S inter microwave workshop series (IMWS) on art of miniaturizing RF and microwave passive
components, Chengdu, pp 63–66
Page 56
DOI 10.1007/978-981-4560-75-7_57-1
Ding Y, Wu K (2009) A 4 x 4 ridge substrate integrated waveguide (RSIW) slot array antenna. IEEE
Antennas Wirel Propag Lett 8:561–564
Ding Y, Wu K (2010) T-type folded substrate integrated waveguide (TFSIW) slot array antenna. IEEE
Trans Antennas Propag 58(5):1792–1795
Dion A (1958) Nonresonant slotted arrays. IRE Trans Antennas Propag 1958:360–365
Djerafi T, Wu K (2012a) 77 GHz planar Butler matrix without crossover. IEEE Trans Antennas Propag
60(10):4949–4954
Djerafi T, Wu K (2012b) Corrugated substrate integrated waveguide (SIW) antipodal linearly tapered slot
antenna array fed by quasi-triangular power divider. Prog Electromagn Res C 26:139–151
Djerafi T, Wu K (2013) Substrate integrated waveguide (SIW) techniques: the state-of-the-art develop-
ments and future trends. J Univ Electron Sci Technol China 42(2):171–192
Djerafi T, Ghassemi N, Kramer O, Youzkatli-el-Khatib B, Guntupalli AB, Wu K (2012a) Small footprint
multilayered millimeter-wave antennas and feeding networks for multi-dimensional scanning and
high-density integrated systems. Radioengineering 21(4):935–945
Djerafi T, Aubert H, Wu K (2012b) Ridge substrate integrated waveguide (RSIW) dual-band hybrid ring
coupler. IEEE Microwave Wirel Compon Lett 22(2):70–72
Djerafi T, Ghiotto A, Wu K (2012c) Antipodal fin-line waveguide to substrate integrated waveguide
transition. In: IEEE MTT-S international microwave symposium digest, Montreal, pp 1–3
Djerafi T, Patrovsky A, Wu K, Tatu SO (2013) Recombinant waveguide power divider. IEEE Trans
Microwave Theory Tech 61(11):3884–3891
Doghri A, Djerafi T, Ghiotto A, Wu K (2011) Early demonstration of a passive millimeter-wave imaging
system using substrate integrated waveguide technology. In: Mediterranean microwave symposium
(MMS), 2011 11th, Hammamet, pp 215–218
Doghri A, Djerafi T, Ghiotto A, Wu K (2013) SIW 90-degree twist for substrate integrated circuits and
systems. In: IEEE MTT-S international microwave symposium digest, Seattle, pp 1–3
Doghri A, Ghiotto A, Djerafi T, Wu K (2014) Broadband E-plane junction for three-dimensional substrate
integrated waveguide circuits and systems. IEEE Microwave Wirel Compon Lett 24(11):739–741
El Khatib BY, Djerafi T, Wu K (2012a) Substrate integrated waveguide vertical interconnects for three-
dimensional integrated circuits. IEEE Trans Compon Packag Manuf Technol 2(9):1526–1535
El Khatib BY, Djerafi T, Wu K (2012b) Three-dimensional architecture of substrate integrated waveguide
feeder for fermi tapered slot antenna array applications. IEEE Trans Antennas Propag
60(10):4610–4618
Esquius-Morote M, Fuchs B, Z€ urcher J-F, Mosig JR (2013a) A printed transition for matching improve-
ment of SIW horn antenna. IEEE Trans Antennas Propag 61(4):1923–1930
Esquius-Morote M, Fuchs B, Zurcher J, Mosig JR (2013b) Novel thin and compact H-plane SIW horn
antenna. IEEE Trans Antennas Propag 61(6):2911–2920
Galejs J (1962) Excitation of slots in a conducting screen above a lossy dielectric half space. IRE Trans
AP-10:436–443
Gazit E (1988) Improved design of the Vivaldi antenna. Proc Inst Elect Eng pt H 135:89–92
Gibson PJ (1979) The Vivaldi aerial. In: Proceedings of the 9th European microwave conference,
Brighton, pp 101–105
Goldsmith PF, Hsieh C-T, Huguenin GR, Kapitzky J, Moore EL (1993) Focal plane imaging systems for
millimeter wavelengths. IEEE Trans Microwave Theory Tech 41:1664–1675
Grigoropoulos N, Young PR (2004) Compact folded waveguides. In: 34th European microwave confer-
ence, Amsterdam, pp 973–976
Grigoropoulos N, Izquierdo BS, Young PR (2005) Substrate integrated folded waveguides (SIFW) and
filters. IEEE Microwave Wirel Compon Lett 15(12):829–831
Page 57
DOI 10.1007/978-981-4560-75-7_57-1
Guntupalli AB (2014) Low-cost integrated waveguide antenna front-end solutions for fifth generation
cellular systems and beyond. Dissertation thesis, Ecole polytechnique de Montreal
Hao Z-C, Hong W, Chen J-X, Chen X-P, Wu K (2005) A novel feeding technique for antipodal linearly
tapered slot antenna array. IEEE MTT-S Int Microw Symp Dig 12–17:1641–1643
Hong W, Wang Y, Lai QH, Liu B (2006) Half mode substrate integrated waveguide: a new guided wave
structure for microwave and millimeter wave application. In: Proceedings of the joint 31st international
conference on infrared millimeter waves and 14th international conference on terahertz electron,
Shanghai
Hyneman RF (1959) Closely-spaced transverse slots in rectangular waveguide. IRE Trans 1959:335–342
Johansson JF (1989) A moment method analysis of linearly tapered slot antennas. In: IEEE antennas and
propagation society international symposium, San Jose, pp 383–386
Kim SG, Chang K (2004) Ultra wideband exponentially-tapered antipodal Vivaldi antennas. In: IEEE
AP-S, Monterey, pp 2273–2276
Lewis LR, Fasset M, Hunt J (1974) A broadband stripline array element. In: IEEE antennas and
propagation society international symposium (digest), Atlanta, pp 335–337
Li ZL, Wu K (2004) A new approach to integrated horn antennas. In: Proceedings of international of
symposium on antenna technology and applied electromagnetics, Montreal, pp 535–538
Li L, Chen X, Khazaka R, Wu K (2009) A transition from substrate integrated waveguide (SIW) to
rectangular waveguide. In: Asia pacific microwave conference, 2009, Singapore, pp 2605–2608
Li J, Wen G, Xiao F (2010) Broadband transition between rectangular waveguide and substrate integrated
waveguide. Electron Lett 46(3):223–224
Lu K (1997) An efficient method for analysis of arbitrary nonuniform transmission lines. IEE Trans
Microwave Theory Tech 45(1):9, 14
Mallahzadeh AR, Esfandiarpour S (2012) Wideband H-plane horn antenna based on ridge substrate
integrated waveguide (RSIW). IEEE Antennas Wirel Propag Lett 11:85–88
Martinez-Ros AJ, Gomez-Tornero JL, Goussetis G (2010) Planar leaky-wave antenna with flexible
control of the complex propagation constant. IEEE Trans Antennas Propag 60(3):1625–1630
Meinel H (1995) Commercial applications of millimeterwaves – history, present status and future trends.
IEEE Trans Microwave Theory Tech 43(7):1639–1653
Moldovan E, Bosisio RG, Wu K (2006) W-band multiport substrate integrated waveguide circuits. IEEE
Trans Microwave Theory Tech 54:625–632
Nanzer J (2012) Microwave and millimeter-wave remote sensing for security applications. Artech House,
Norwood
Nesic A (1991) Endfire slot line antennas excited by a Coplanar waveguide. In: IEEE AP-S international
symposium, digest, London, pp 700–702
Oliner AA, Jackson DR, Volakis JL (2007) Leaky-wave antennas, Chapter 11. In: Antenna engineering
handbook, 4th edn. McGraw-Hill, New York
Patrovsky A, Daigle M, Wu K (2007) Millimeter-wave wideband transition from CPW to substrate
integrated waveguide on electrically thick high-permittivity substrates. In: Proceedings of 37th Euro-
pean Microwave Conference (EuMC37), Munich, pp 138–141
Pozar DM (1983) Surface wave effects for millimeter wave printed antennas. In: Proceedings of the
antennas and propagation society international symposium, Houston, vol 21, pp 692–695
Pozar DM (2011) Microwave engineering, 4th edn. Wiley, New York
Prasad SN, Mahapatra S (1979) A novel MIC slot line aerial. In: Proceedings 9th European microwave
conference, Brighton, pp 120–124
Salmon NA (2004) Scene simulation for passive and active millimetre and sub-millimetre wave imaging
for security scanning and medical applications. SPIE 5619:129–135
Page 58
DOI 10.1007/978-981-4560-75-7_57-1
Sato H, Sawaya K, Arai N, Wagatsuma Y, Mizuno K (2003) Broadband FDTD analysis of fermi antenna
with narrow width substrate. In: IEEE antennas and propagation society international symposium,
2003, Columbus, vol 1, pp 261–264
Scharp GS (1982) Continuous slot antennas. US Patent 4,328,502, 4 May 1982
Schaubert DH (1993) Scanning characteristics of stripline- fed tapered slot antennas on dielectric sub-
strates. In: Proceedings of the 1993 antenna applications symposium, Monticello
Schoebel J, Herrero P (2010) Planar antenna technology for mm-wave automotive radar, sensing, and
communications. In: Kouemou G (ed) Radar technology InTech. ISBN:978-953-307-029-2,
doi:10.5772/7185. http://www.intechopen.com/books/radar-technology/planar-antenna-technology-
for-mm-wave-automotive-radar-sensing-and-communications
Simons RN, Lee RQ, Perl TD (1992) Non-planar linearly tapered slot antenna with balanced microstrip
feed. In: IEEE AP-S international symposium, Chicago, vol 4, pp 2109–2112
Solbach K, Schneider R (1999) Review of antenna technology for millimeter wave automotive sensors.
In: Proceedings of the European microwave conference, Munich, pp 139–142
Sugawara S, Maita Y, Adachi K, Mori K, Mizuno K (1998) Characteristics of a MM-wave tapered slot
antenna with corrugated edges. IEEE Trans Microwave Theory Tech 2:533–536
Sutinjo A, Okoniewski M, Johnston RH (2008) Radiation from fast and slow traveling waves. IEEE
Antennas Propag. Mag. 50(4):175–181
Uchimura H, Takenoshita T, Fujii M (1998) Development of à laminated waveguide. IEEE Trans
Microwave Theory Tech 46(12):2438–2443
Vye D (2011) Divine innovation: 10 technologies changing the future of passive and control components.
Microw J 54(11):22–42
Walter CH (1965) Traveling wave antennas. McGraw-Hill, New York
Wang N-B, Jiao Y-C, Song Y, Zhang L, Zhang FS (2009) A microstrip-fed logarithmically tapered slot
antenna for wideband applications. J Electromagn Waves Appl 23(10):1335–1344
Wang H, Fang D-G, Zhang B, Che W-Q (2010) Dielectric loaded substrate integrated waveguide (SIW)
H-plane horn antennas. IEEE Trans Antennas Propag 58(3):640–647
Wei J, Chen ZN, Qing X, Shi J, Xu J (2013) Compact substrate integrated waveguide slot antenna array
with low back lobe. IEEE Antennas Wirel Propag Lett 12:999–1002
Whetten F, Balanis CA (1991) Meandering long slot leaky-wave antennas. IEEE Trans Antennas Propag
39(11):1553–1559
Wong M, Sebak AR, Denidni TA (2008) A broadside substrate integrated horn antenna. In: IEEE AP-S
international symposium 2008, Atlanta, pp 1–4
Wu K (2001) Integration and interconnect techniques of planar and nonplanar structures for microwave
and millimeter-wave circuits-current status and future trend, In: Asia–pacific microwave conference
2001, Taipei, Taiwan, pp 411–416
Wu C, Shen L, Deng GY, Shen Y, Litva J (1998) Experimental study of a wide band LTSA which is fed by
an inverted microstrip line (IML). In: IEEE AP-S international symposium, San Diego,
vol 4, pp 2328–2331
Wu K, Deslandes D, Cassivi Y (2003) The substrate integrated circuits-a new concept for high-frequency
electronics and optoelectronics. In: Telecommunications in modern satellite, cable and broadcasting
service, TELSIKS 2003, 6th international conference, vol 1, pp P-III
Wu K, Cheng YJ, Djerafi T, Hong W (2012) Substrate integrated millimeter-wave and terahertz antenna
technology. Proc IEEE 100(7):2219–2232
Xia L, Xu R, Yan B, Li J, Guo Y, Wang J (2006) Broadband transition between air-filled waveguide and
substrate integrated waveguide. Electron Lett 42(24):1403–1405
Page 59
DOI 10.1007/978-981-4560-75-7_57-1
Xu F, Zhang Y, Hong W, Wu K, Cui TJ (2003) Finite difference frequency-domain algorithm for

modeling guidedwave properties of substrate integrated waveguide. IEEE Trans Microwave Theory
Tech MTT-51(11):2221–2227
Xu J, Hong W, Hongjun H, Kuai Z, Wu K (2008) Half-mode substrate integrated waveguide (HMSIW)
leaky-wave antenna for millimeter-wave applications. IEEE Antennas Wirel Propag Lett 7:85–88
Yamamoto S, Hirokawa J, Ando M (2002) A beam switching slot array with a 4-way Butler matrix
installed in a single layer post-wall waveguide. In: Proceedings of the IEEE AP-S international 1241
symposium, San Antonio, vol 1, pp138–140
Yan L, Hong W, Wu K, Cui TJ (2005) Investigations on the propagation characteristics of the substrate
integrated waveguide based on the method of lines. Proc IEE Microwave Antennas Propag 152:35–42
Yang Y, Wang Y, Fathy AE (2008) Design of compact Vivaldi antenna arrays for UWB see through wall
applications, Progress In Electromagnetics Research, PIER vol 82, pp 401–418
Yngvesson SK, Schaubert DH, Korzienowski TL et al (1985) Endfire tapered slot antennas on dielectric
substrates. IEEE Trans Antennas Propag AP-33(12):1392–1400
Yngvesson KS, Korzeniowski TL, Kim Y-S, Kollberg EL, Johansson JF (1989) The tapered slot
antenna – a new integrated element for millimeter-wave applications. IEEE Trans Microwave Theory
Tech 37(2):365–374
Zhu L, Wu K (2002) Field-extracted lumped-element models of coplanar stripline circuits and disconti-
nuities for accurate radio-frequency design and optimization. IEEE Trans Microwave Theory Tech
50(4):1207–1215
Page 60
View publication stats

Substrate Integrated Wave Guide Antennas

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Substrate Integrated Wave Guide Antennas

Încărcat de

Drepturi de autor:

Formate disponibile

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Substrate Integrated Waveguide Antennas

Chapter · January 2015

Tarek Djerafi Ali Doghri

SEE PROFILE SEE PROFILE

Innovative 94 GHz Components (Ph.D.) View project

IEEE Student Branch of Bordeaux View project

The user has requested enhancement of the downloaded file.

Substrate Integrated Waveguide Antennas

Substrate Integrated Waveguide

SIW Techniques and Design Basics

SIW Propagation Modes

and the corresponded guided wavelength is:

The cut-off frequency of the ﬁrst mode is deﬁned by:

And for the second mode by:

• Attenuation because of the conductive loss:

Fig. 5 Attenuation constant versus frequency for an SIW structure

Table 1 SIW line configurations allowing for a spatial arrangement

Principal SIW Parameters

SIW Line Conﬁgurations

Fig. 6 Dispersion characteristics of different SIW lines

tan ðs þ d Þk x þ tan ða s d Þk x ¼ 0 (15)

and for L-type SIW line by:

where ZTE is the wave impedance of the TE mode.

Rectangular Waveguide-to-SIW Transition

Nonresonant SIW Antenna

E-Plane SIW Horn (Waveguide Transition)

H-Plane SIW Horns

where l0 is wavelength in free space.

Fig. 13 Optimal gain for different lengths and apertures

C(x) and S(x) are Fresnel integrals, which are deﬁned as

Matching Bandwidth and Radiation Improvements

Fig. 15 Example of the fabricated horn antennas according to Fig. 14

Uniform Leaky-Wave Structure

Long Slot Leaky-Wave Antenna

Pointing angle (Deg)

Fig. 17 Scanning range of uniform slot with different permittivity

Fig. 18 Simulated patterns of leaky-wave antenna

Lateral Slot Leaky-Wave Antenna

Fig. 19 (a, b) Conﬁguration of HMSIW LWA, (c) a fabricated prototype

The distance d0 corresponding to this condition is

Fig. 22 The reﬂection cancelation forward-wave LWA

Fig. 24 Open periodic waveguides

Laterally Open Periodic Waveguides

Antipodal Tapered TSA

Fig. 26 SIW TSA

Fig. 29 SIW and stripline impedance

Fig. 30 TSA antenna with corrugation

Without corrugation With corrugation

Normalised Gain (dB)

Table 2 Typical properties of end fire traveling antenna

Maximal Gain Versus Length

Fig. 34 (a) Linear taper, (b) CWTSA with linear taper

Fig. 35 Effect of antenna length variations on gain of LTSA

Fig. 36 Effect of antenna length variations on gain of CWTSA

– Nonlinear taper (Vivaldi, tangential or parabolic),

Gain (dB) 10.5

yðxÞ ¼ AeRx þ y0 (53)

yðxÞ ¼ Ax2 þ y0 (55)

Fig. 43 Measured and simulated return losses

Fig. 44 Measured radiation patterns at 28 GHz

Folded SIW Line

Normalized Gain (db) –15

−30 Non Folded ATSA