6894 IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, VOL. 67, NO.

11, NOVEMBER 2019

Reconfigurable Phased-Array Antenna Using

Continuously Tunable Substrate Integrated
Waveguide Phase Shifter
Yuan Ji , Lei Ge , Senior Member, IEEE, Jianpeng Wang , Quangang Chen, Wen Wu, Senior Member, IEEE,
and Yujian Li , Member, IEEE
Abstract— A low-cost 1 × 4 phased array incorporating con- phased arrays with wide beam-steering ranges are urgently
tinuously tunable substrate integrated waveguide (SIW) phase required. Since substrate integrated waveguide (SIW) tech-
shifters is presented in this paper. Through seamless integra- nology has the advantages of low cost, low profile, low
tion between antenna elements, phase shifters, and an SIW
feeding network, the proposed phased array is assembled in loss, simple fabrication process, and easy integration with
a multilayered monolithic structure. To obtain a wide phase planar circuits, it is a wise choice for the design of phased
variation range, the electronically controllable phase shifter is arrays [1]–[4].
constructed by introducing varactor diodes into a reflection-type Generally, SIW beam-forming phased-array antennas come
architecture, which can also offer a simple solution to constitute primarily in two forms: one option is to implement an existing
the biasing circuit of the diodes. Due to the merits of adequate
bandwidth and good radiation performance, an aperture-coupled beam-forming network based on the SIW, for instance, a butler
patch antenna with multi-resonance characteristic is selected as matrix [5] or a rotman lens [6] and the other is to connect
the array element. Accordingly, a prototype of the proposed digital or analog phase shifters with a properly designed
phased array was fabricated by applying standard printed circuit feeding network [7]–[9]. One drawback of the first approach
board (PCB) facilities. The measured results demonstrate that a is that the radiation beam can only be switched to a set of
continuous beam scanning range from −45◦ to 45◦ covering the
whole operating band of 5.09–5.98 GHz is obtained. Meanwhile, discrete scanning angles; meanwhile, the size and complexity
a gain of up to 11 dBi and an impedance bandwidth of 16.1% of the beam-forming network rapidly increase along with the
are achieved. Owing to the features of continuous wide-angle addition of beam directions. In contrast, the second option
steerable beam, easy fabrication, and integration, the proposed provides a relatively simplified scheme to implement the
phased-array antenna is attractive for 5G Wi-Fi applications. beam-forming network, but the low integration level caused
Index Terms— Aperture-coupled patch antenna (ACPA), phase by the existence of traditional bulky phase shifters remains a
shifter, phased array, substrate integrated waveguide (SIW). major problem. In [7], surface-mounted 4- bit digital phase
shifters have been selected to vary the phase of SIW slot
I. I NTRODUCTION antennas, resulting in a discrete beam scanning range of ±19◦.
To increase the beam-steering angle, an SIW-slot phased array
T HE 5G Wi-Fi (covering frequency band from 5.15 to
5.85 GHz) technology, based on the IEEE 802.11ac
standard, has attracted considerable attention in recent years
with a wide-angle scanning ability (±60◦) has been presented
in [8]. Nonetheless, it requires additional external 6- bit digital
due to its advantages of higher data rate and wider channel phase shifters, a power divider, coaxial cables, and a control
bandwidth compared with the former generation. Notably, computer to behave as a whole, which makes the system
beam-forming phased-array antennas can enormously improve bulky and costly. For the purpose of tight integration of
the transmission efficiency and coverage area of 5G Wi-Fi phase shifters and antenna elements, ferrite low-temperature
devices. However, conventional phased arrays suffer from co-fired ceramic (LTCC) technology has been adopted in the
finite beam-scanning coverage, high cost, bulky structure, etc. SIW to tune the phase of antenna elements [9]. However,
Therefore, low-cost, compact, high-gain, and easily-integrated the phase shift level is restricted to the limited number of
available LTCC layers, which makes the design away from a
Manuscript received December 29, 2018; revised April 21, 2019; accepted wide beam-scanning angle. In addition, a low-cost 360◦ beam-
June 24, 2019. Date of publication July 16, 2019; date of current ver-
sion October 29, 2019. This work was supported in part by the National steering reconfigurable SIW horn antenna proposed in [10]
Natural Science Foundation of China under Grant No. 61601303 and in overcomes the above-mentioned problems nicely; nevertheless,
part by the Fundamental Research Foundation of Shenzhen under Grant it is inconvenient to integrate with planar circuits. As a
No. JCYJ20170817095519575. (Corresponding author: Lei Ge.)
Y. Ji, J. Wang, and W. Wu are with the Ministerial Key Laboratory of result, continuously tunable SIW phase shifters with large
JGMT, Nanjing University of Science and Technology, Nanjing 210094, China phase variation ranges would be preferable to other phase
(e-mail: [email protected]; [email protected]). shifter counterparts for the SIW phased arrays, which can be
L. Ge and Q. Chen are with the Institute of Microscale Optoelectronics,
Shenzhen University, Shenzhen 518060, China (e-mail: [email protected]). attributed to the features of low cost, planar structure, and easy
Y. Li is with the Institute of Lightwave Technology, Beijing Jiaotong integration with SIW-based components [11]–[13].
University, Beijing 100044, China (e-mail: [email protected]). In terms of SIW antenna elements selection, some notewor-
Color versions of one or more of the figures in this article are available
online at http://ieeexplore.ieee.org. thy indicators such as bandwidth, gain, size, cross-polarization
Digital Object Identifier 10.1109/TAP.2019.2927813 level, radiation efficiency, and front-to-back ratio (FTBR)
0018-926X © 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.
 JI et al.: RECONFIGURABLE PHASED-ARRAY ANTENNA USING CONTINUOUSLY TUNABLE SIW PHASE SHIFTER 6895

should also be taken into account carefully. SIW cavity-

backed slot antennas (CBSAs) inherit the merit of good
radiation performance from the conventional cavity-backed
antennas and have additional advantages of low profile, low
cost, and easy integration as well, which are good candidates
for radiating elements [14]. Despite all these, they always
suffer from narrow impedance bandwidth. Recently, significant
research has been conducted on improving bandwidth of the
SIW CBSAs [15]–[20]. Among them, various attempts have
been made to modify the configuration of the slot etched on
the top copper-clad surface of the SIW cavity, which have
shown promising results [15]–[17]. Another way is to add
shorting vias at specific positions in the SIW cavity to excite
higher modes [18]. Multilayer structure can also effectively
enhance the bandwidth, although slightly increases the size
and complexity of the antenna [19]. SIW circularly polarized
cavity-backed antennas using different feeding transitions [20]
or adopting a half-mode SIW (HMSIW) cavity [21] have been
proved to be good choices as well. It is worth mentioning
that SIW cavity-backed patch antennas provide wideband and
low-profile antenna design schemes, but the radiation patterns
are asymmetrical in the E-plane and the cross-polarization lev-
els are relatively high [22], [23]. Consequently, an alternative
method of improving both bandwidth and gain is to add a
patch over the slot on the SIW cavity [24]–[26]. Afterward,
to obtain high-gain, wideband, symmetrical radiation pattern,
and low cross-polarization level, SIW-fed aperture-coupled
patch antennas (ACPAs) adopting multilayer structures with
air cavities have been presented in [27] and [28].
To enable miniaturization and integration of traditional
phased arrays while at the same time improving beam steering
performance in a cost-effective manner, a new 1 × 4 SIW
phased array is investigated in this paper. First, an SIW-fed Fig. 1. Geometry of the proposed ACPA element with SIW backed cavity.
(a) 3-D view, (b) top view of the cavity, (c) top view of the patch, and
ACPA is thoughtfully chosen as the array element, and its (d) cross-sectional view.
structure, working mechanism, and performance are described
in Section II. Second, a high-performance SIW-based tun-
able phase shifter provides a feasible scheme to realize
seamless integration between the antenna elements and the the bottom substrate is used for constructing an SIW square
beam-forming network, which is thoroughly discussed in cavity with a size of Wc × Lc, which is formed by inserting an
Section III. Then, the overall architecture of the phased inductive window at a distance of Lc from the short-circuited
array, involving the SIW feeding network and ground copla- end of SIW. For the purpose of optimizing the impedance
nar waveguide (GCPW)-to-SIW transition, is presented in matching, the inductive window is opened toward the input
Section IV. Finally, a prototype is fabricated and measured port of the antenna, which is flanked by metallized vias.
to validate the proposed design. The measured results exhibit Width of the SIW feeding line is set to Wc to satisfy the
good agreement with simulated ones, thus the conclusion is constrained equation derived in [2], which ensures the SIW
given in Section V. works at a single mode within the operating frequency band.
To minimize the energy leakage, diameter dv and spacing Sv
II. A NTENNA E LEMENT of the vias are properly determined to satisfy the condition of
dv/Sv≥ 0.5 and dv/λ0 ≤ 0.1 (where λ0 is the free space wave-
A. Geometry length) [3], [4]. In addition, a small amount of space intervals
The proposed ACPA element is composed of two substrates, of the vias in each corner of the cavity are slightly adjusted
two metal layers, and a square patch, as shown in Fig. 1. to shorter distances as the optimal cavity width/length cannot
In order to facilitate the seamless integration with SIW-based distribute the vias uniformly. A transverse rectangular aperture
phase shifters, the antenna element is designed to be fed by is etched on the upper metal layer of the cavity, providing
the SIW with a port impedance of 50 . The bottom substrate an efficient coupling to the patch. It should be mentioned
is a 3.175 mm-thick double-sided copper-clad printed circuit that shifting the coupling aperture with an offset distance Ds
board (PCB) laminate with relative permittivity of 2.33 and excites an additional resonant mode and, therefore, results
loss tangent of 0.0012 at 10 GHz. As can be seen in Fig. 1(b), in a wider impedance bandwidth. The realization mechanism

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.
 6896 IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, VOL. 67, NO. 11, NOVEMBER 2019

TABLE I
D IMENSIONS OF THE A NTENNA E LEMENT (U NIT: mm)

of the multi-mode resonances will be further discussed in

Section II-B.
The upper substrate with a thickness of 0.508 mm serves as
a support plate for the patch, which is a single-sided copper-
clad PCB laminate with relative permittivity of 3.38 and loss
tangent of 0.0027 at 10 GHz. An air-gap layer is introduced
by placing several plastic hollow pillars between the upper
substrate and the bottom substrate, then plastic screws are used
to fix them together, as shown in Fig. 1(a) and (d). Height of
the pillars is h a , namely the height of the air gap. Since the
air gap has an effect of lowering the effective permittivity and
increasing the thickness of the dielectric underneath the patch,
thus the enhancement of the impedance bandwidth is achieved.
Also, the effective permittivity is given by [29]
Fig. 2. Current distributions of the ACPA with varying slot offset distance
εr (h s + h a ) Ds. (a) Ds = 0 mm, (b) Ds = 2.2 mm, and (c) Ds = 4.4 mm.
εe f f = (1)
h s + h a εr
where εr and h s are relative permittivity and thickness of the
upper substrate, respectively. The resonant frequencies can,
therefore, be tuned by adjusting the air-gap height.
As shown in Fig. 1(c), the square patch is aligned to the
center of the backed cavity. The side length of the patch Lp is
equal to the aperture length Ls for strong coupling. After being
optimized by the electromagnetic full-wave solver HFSS [30],
dimensions of the ACPA are obtained and tabulated in Table I.

B. Mechanism
To acquire physical insight regarding the antenna operating
principles, a series of parametric studies are provided along
with the current distributions to obtain deeper understanding
of the working mechanism of the ACPA. In each case,
Fig. 3. Simulated reflection coefficients of the ACPA with varying slot offset
dimensions are kept as the optimum values except the studied distance Ds.
parameter.
1) Effects of Slot Offset Distance: The current distributions
and reflection coefficients of the ACPA with varying slot offset 5.38 and 5.75 GHz, respectively, as shown in Fig. 2(b). For the
distance Ds are illustrated in Figs. 2 and 3, respectively. When first resonance, both the minimum-density current points shift
the slot is placed in the center of the cavity (i.e., Ds = 0 mm), upward accordingly, and the current distribution is basically
the cavity is divided into two parts, namely the upper half part similar to Fig. 2(a). However, an emerging TE110 mode with
and the lower half part, and only one resonance is excited, relatively weaker magnitude occurs. The two modes blend
as shown in Fig. 2(a). It can be seen that there are two together, which cancel mutually in the upper half cavity part
minimum-density current points locating at the middle of the with the opposite direction and superpose in the lower half
two parts. The slot cuts off the surface current and compels it cavity part with the same direction. Therefore, higher density
to alter patch. A strong electric field exists across the slot that currents mainly distribute at the lower half part of the cavity
radiates electromagnetic energy from the waveguide to free surface at 5.38 GHz. That is to say, the first resonant frequency
space, and the most energy is coupled with the upper patch. is more likely to be affected by the size of the lower half cavity
Thus, a traditional TE210 mode is excited in the slotted cavity part.
and a TM10 mode is generated on the patch at 5.45 GHz. After a quarter of a period, the second resonance appears at
When the slot is moved 2.2 mm upward, two different a higher frequency. The aforementioned TE210 mode becomes
current distributions are observed on the cavity surface at much weaker now and has a reverse direction, while the TE110

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.
 JI et al.: RECONFIGURABLE PHASED-ARRAY ANTENNA USING CONTINUOUSLY TUNABLE SIW PHASE SHIFTER 6897

Fig. 4. Simulated reflection coefficients of the ACPA with varying slot Fig. 5. Simulated reflection coefficients of the ACPA with varying slot length
width Ws. Ls and patch length Lp.

mode gets stronger. They superpose in the upper half cavity

part and cancel in the lower half cavity part. Thus, higher
density currents distribute at the upper half part of the cavity
surface, especially at the top edge of the slot. So the second
resonant frequency is principally influenced by the size of the
upper half cavity part, which is slightly higher than the first
resonant frequency as the upper half part is smaller than the
lower one. As shown in Figs. 2(c) and 3, when we further
increase the size of the lower half part (i.e., Ds = 4.4 mm),
the first resonant frequency drops down to 5.27 GHz and the
second resonant frequency increases to 5.97 GHz. Note that
the same resonant mode is stimulated on the patch at those
resonant frequencies. Fig. 3 shows that a better impedance
matching is obtained when Ds = 2.2 mm, so it is chosen as Fig. 6. Simulated input impedances of the ACPA with and without the
the optimal value. inductive window.
2) Effects of Slot Width: Another key parameter (slot width
Ws) is studied and the simulated reflection coefficients of the
frequency. When both Ls and Lp are set to 15.5 mm, the ACPA
ACPA with varying slot width Ws are shown in Fig. 4. It can
works at the desired frequency band with good impedance
be concluded that a wider slot leads to a broader bandwidth.
matching.
From Fig. 2(b), the highest density currents mainly concentrate
4) Effect of Inductive Window: To illustrate the effect of
around the slot for the first resonance while they basically
the inductive window mentioned in Section II-A, another
distribute at the top edge of the slot for the second resonance.
ACPA without the inductive window is simulated and the
So increasing the slot width results in a longer current path
input impedances of both the antennas are shown in Fig. 6.
and a lower resonant frequency for the first resonance. But
A worse impedance matching of the antenna without the
the second resonance is more sensitive to the size of the
inductive window is observed, and the imaginary part of its
upper half cavity part, so the corresponding resonant frequency
input impedance is below zero over the entire frequency band
is elevated with the reduction in the size of the upper half
of 5–6 GHz. Therefore, an inductive window is added for a
part. In addition, the half-wavelength resonant mode remains
better impedance matching.
unchanged on the patch with the variation in slot width. As a
result, Ws = 2.5 mm is selected as the optimal value since the
corresponding return loss is higher than that of 3.5 mm. C. Performance
3) Effects of Slot Length and Patch Length: The reflection In order to clarify the features of the proposed antenna more
coefficients of the ACPA with varying Ls and Lp are shown clearly, a similar SIW cavity-backed structure without the
in Fig. 5. When the slot length and patch length are increased patch is presented as a comparison. After removing the patch,
simultaneously, the longer current path along the slot and the the proposed ACPA becomes a CBSA. When the dimension
longer resonant length on the patch jointly lead to the decline of the CBSA remains unchanged, the simulated reflection
in both the resonant frequencies. Moreover, a longer slot coefficients of both the antennas are given in Fig. 7. It can
weakens the magnitude of the TE110 mode and thus neutralizes be seen that the impedance matching of the antenna without
the elongation of the highest density currents at the top edge the patch sharply deteriorates, indicating that no resonance
of the slot, resulting in a minor change of the second resonant occurs at the desired frequency band and scarcely any power

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.
 6898 IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, VOL. 67, NO. 11, NOVEMBER 2019

Fig. 7. Simulated reflection coefficients of the ACPA (with and without

patch) and the optimized CBSA. Optimized CBSA dimensions: Wc = 25.1,
Lc = 20.9, Ls = 23.8, Ws = 1.6, Ds = 0.8, L f = 14.22, Lo = 3, Sv = 1.6,
and dv = 0.8, all in millimeters.

is radiated away through the slot. This is primarily because the

slot is not a resonant slot as the slot length is much less than
a half resonant wavelength. However, the loading effect of the
patch in the ACPA enables the relatively shorter slot radiates.
Therefore, in order to match the CBSA and ACPA separately,
their dimensions of the slot length Ls, the slot width Ws,
the inductive window length Lf, and the slot offset distance Ds
have been optimized independently and with different values.
Fig. 7 shows that the simulated −10 dB impedance bandwidth Fig. 8. Electric field distributions in the SIW cavities. (a) First hybrid mode
in the ACPA at 5.38 GHz. (b) Second hybrid mode in the ACPA at 5.75 GHz.
of the proposed ACPA is 12.7% (5.22–5.93 GHz), whereas the (c) First hybrid mode in the optimized CBSA at 5.5 GHz. (d) Second hybrid
optimized CBSA is 7.9% (5.36–5.80 GHz). mode in the optimized CBSA at 5.75 GHz; dominant electric field contour
Electric field distributions in the SIW cavity of both the maps of the ACPA. (e) First hybrid mode made up of strong TE210 mode and
weak TE110 mode. (f) Second hybrid mode made up of strong TE110 mode
antennas are shown in Fig. 8, where the symbols “⊕” and and weak TE210 mode.
“” indicate the directions of the electric field. It is found that
pretty similar hybrid resonant modes are generated in both the
cavities. As indicated in Section II-B, the first resonant mode is
a strong TE210 mode combined with a weak TE110 mode. The
dominant mode field distributes in the lower half cavity part,
as shown in Fig. 8(e). The other resonant mode at the higher
frequency is also a hybrid mode made up of a strong TE110
mode and a weak TE210 mode. Also, the dominant mode field
distributes in the upper half cavity part, as shown in Fig. 8(f).
The simulated normalized radiation patterns of the ACPA
and the CBSA at 5.5 GHz are illustrated in Fig. 9. The FTBR
of the CBSA is larger than 16 dB and that of the ACPA
is larger than 30 dB in both the E-plane and H-plane. The
3 dB beamwidths of both the antennas are roughly 76◦ in the
H-plane and the radiation pattern of the CBSA in the E-plane Fig. 9. Simulated normalized radiation patterns of the ACPA and the
is slightly fatter than that in the H-plane at the lower elevation optimized CBSA at 5.5 GHz. (a) E-plane (xoz plane). (b) H-plane (yoz plane).
range. Besides, the cross-polarizations of both the antennas
are less than −28 dB. The simulated realized gains and the
radiation efficiencies of the two antennas are shown in Fig. 10. Compared with the cavity-backed ACPA (size: λ0 × λ0 ×
The simulated peak gain for the ACPA is up to 8 dBi, which 0.46λ0 , where λ0 represents the free-space wavelength in the
is approximately 2.5 dBi higher than that of the CBSA. The center operating frequency) that consists of two substrates and
simulated radiation efficiencies of both antennas are greater one aluminium plate presented in [28], which has a bandwidth
than 95%. In addition, the size of the ground plane has an of 10.7% and an average gain of 8 dBi, the proposed ACPA
impact on the antenna performance and a larger ground plane element (size: 0.42λ0 × 0.42λ0 × 0.12λ0 ) exhibits comparable
provides higher gain and better FTBR. performance but with a simpler structure and a smaller size.

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.
 JI et al.: RECONFIGURABLE PHASED-ARRAY ANTENNA USING CONTINUOUSLY TUNABLE SIW PHASE SHIFTER 6899

Fig. 10. Simulated realized gains and radiation efficiencies of the ACPA and
the optimized CBSA.

The CBSA with a parasitic patch proposed in [24] has a

bandwidth of 11% and a peak gain of 5.65 dBi, which differs
in the feeding type and the radiation mechanism from this
work.

III. SIW P HASE S HIFTER

A. Configurations
As one of the key elements of a phased array, phase shifters
with features of low insertion loss, large phase diversity
range, wide bandwidth, low profile, and low cost should be
considered as design objectives. In the past decade, various
phase shifters based on the SIW have been designed. Most of
them can be divided into three types in terms of phase shifting
degree: fixed, digital, and analog phase shifters. Fixed-type
SIW phase shifters usually insert inductive posts, air holes,
or adopt delay line/nonuniform SIW lengths to obtain constant
phase shifts. As for digital SIW phase shifters, PIN diodes and
micro-electromechanical system (MEMS) are often utilized as
switches to electronically control the phase shifts to a set of
discrete degrees. Ferrite LTCC or varactor diodes tend to be
used in analog SIW phase shifters, which can tune the phase Fig. 11. Geometry of the proposed SIW phase shifter. (a) 3-D view, (b) top
continuously. view of the top metal layer and bottom layer, (c) top view of the middle layer,
An analog SIW reflective-type phase shifter (RTPS) using and (d) cross-sectional view.
varactor diodes as the tunable element is presented in this
paper, which has nearly the same SIW width with the afore- TABLE II
mentioned ACPA and it also adopts the same 3.175 mm- D IMENSIONS OF THE SIW P HASE S HIFTER (U NIT: mm)
thick double-sided copper-clad PCB laminate. Geometry of the
proposed RTPS is illustrated in Fig. 11 and its dimensions are
given in Table II. The RTPS consists of two identical substrates
(i.e., upper substrate and lower substrate) and three metal
layers (i.e., top metal layer, middle metal layer, and bottom
metal layer). The top metal layer and the bottom metal layer
are symmetric about the xoy plane and an annular rectangular
slot is etched on both of them. As for the middle metal layer, matched to 50 . The output phase can be adjusted by tuning
a longitudinal slot is etched along the x-axis. the reflective load impedance, which is realized by adjusting
The schematic block diagram of the RTPS is plotted the bias voltage applied to the varactor diodes. Since the
in Fig. 12. It consists of a 3 dB 90◦ hybrid junction and components are connected directly, the phase of the device
two reflective loads with equal impedances. The through and transmission coefficient would be the same with the phase of
coupled ports of the 90◦ hybrid junction are connected to the load reflection coefficient.
the two reflective loads, respectively, and the other two ports The SIW 90◦ hybrid junction is constructed on a double-
are used as the input and output which are symmetrically layered SIW structure in order to reduce the dimension in

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.
 6900 IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, VOL. 67, NO. 11, NOVEMBER 2019

Fig. 14. Simulated results of the SIW 90◦ hybrid junction.

Fig. 12. Block diagram of the SIW RTPS.

Fig. 13. Geometry of the SIW 90◦ hybrid junction.

the horizontal plane, as shown in Fig. 13. It comprised two

stacked laminates with two parallel linear arrays of metalized
vias. The two rows of metallized vias in the SIW act like two
metallized walls. The longitudinal slot etched on the common
metal layer is off center placed. Therefore, signals are coupled
through the slot. The 90◦ hybrid junction is a four-port network
and the phase difference of the through port and coupled port
is 90◦. Signals reflected from the two identical reflective load
short-circuited terminations add up at the isolated port and
cancel out at the input port. Theoretically, the phase shift
of the reflected signals between the isolated port and the Fig. 15. Geometry of the two variable reflective loads. (a) 3-D view, (b) top
input port will double the insertion phase of each reflective view of the top metal layer and bottom metal layer, and (c) cross-sectional
load. The performance of the hybrid junction depends on the view.
isolation, the coupling balance, the operating bandwidth, etc. TABLE III
By optimizing the slot width Ws1 and the slot length Ls1, D IMENSIONS OF THE R EFLECTIVE L OADS (U NIT: mm)
as well as the offset distance Ds2, a required coupling level
can be obtained. The parameter values of those dimensions are
given in Table II and the simulated results are shown in Fig. 14.
It can be seen that almost equal power division between the
through port and the coupled port is achieved. The reflection
coefficient is below −20 dB and the isolation is more than
−20 dB within the operating frequency band. The reflective loads are series connected with the 90◦ hybrid
The geometry of the two variable reflective loads is junction and constructed on a section of double-layered
shown in Fig. 15 and its dimensions are given in Table III. SIW structure with short-circuited termination. The annular

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.
 JI et al.: RECONFIGURABLE PHASED-ARRAY ANTENNA USING CONTINUOUSLY TUNABLE SIW PHASE SHIFTER 6901

varactor-loaded slots. The effective length of the reflective load

becomes

L e f f = L 0 + L 1 + L 2 + 2 × Ws2 (3)

where L 0 , L 1 , and L 2 are the lengths of the three segments

of SIW transmission lines. They can be calculated by
Ws2 0.5dv2
L 0 = L 0 − − (4)
2 0.95Sv
L 1 = L 1 − Ws2 (5)
Ws2 0.5dv2
L 2 = L 2 − − . (6)
2 0.95Sv
The characteristic admittance Y0 and phase constant β of
the SIW transmission line can be calculated by
β We f f
Y0 = (7)
η0 k0 h ss
  
cπ 2
β = k0 εr −
2 (8)
We f f
Fig. 16. Equivalent circuit models of (a) varactor diode and (b) variable
reflective load. where k0 and η0 = 120π are the wavenumber and wave
impedance in the free space, respectively.
The series impedance of the varactor is
rectangular slot is symmetrically etched on the top metal layer
1
and the bottom metal layer, as shown in Fig. 15(a). Since the Z v = Rv + j w0 L v + . (9)
available phase shifting range for this type of structure mainly j w0 Cv
depends on the capacitance range of the varactor diode; a The transverse slot can be regarded as a series network [31].
cascaded network is typically used to obtain a large phase The normalized admittance of the slot Ys /Y0 = (G s + j Bs )/Y0
shift. Initially, two side-by-side transverse slots are etched on can be evaluated in [31, eqs. (26) and (33)], where G s /Y0 and
the surface of the short-circuited SIW section. Then, a varac- Bs /Y0 are the normalized conductance and the normalized sus-
tor diode (Model: Skyworks SMV2020-079LF) is assembled ceptance of the slot, respectively, and Y0 is the characteristic
across the middle of each slot. Afterward, for convenience admittance of the slot.
of implementing the biasing circuit of the varactor, another By utilizing transmission line model (TLM) theory, the input
two narrow longitudinal slots are etched to connect each impedance of the reflective load can be derived by
end of the two transverse slots, creating an inner rectangular ⎧
patch isolated from the ground plane, as shown in Fig. 15(b). ⎪
⎪ Z 1 = j Z 0 tan(β L 2 )
⎪
⎪
⎪
In addition, the width of the longitudinal slot is 0.1 mm, ⎪Z2 = Z L S + Z1
⎪
⎪
thus the resultant radiation from the narrow slot is negligible. ⎪
⎨ Z 2 + j Z 0 tan(β L I )
Z3 = Z0
Fig. 15(c) shows that control voltage could be applied to the
⎪ Z 0 + j Z 2 tan(β L 1 ) (10)
patch through a bonding wire, without the need for installing ⎪
⎪
⎪
⎪ Z 4 = Z LS + Z 3
extra dc blocking circuits [11], [12]. ⎪
⎪
⎪
⎪ Z 4 + j Z 0 tan(β L 0 )
⎩Z L = Z0
Z 0 + j Z 4 tan(β L 0 )
B. Working Principles
where Z 1 , Z 2 , Z 3 , Z 4 , and Z L are different impedances seen
The equivalent circuit model of the varactor diode and
from the reference nodes marked in Fig. 16(b), characteristic
the reflective load are shown in Fig. 16. According to the
impedance Z 0 = 1/Y0 is given in (7), phase constant β of the
diode SPICE model given in its data sheet, the varactor is
transmission line is given in (8), Z L S is the impedance of the
equivalent to a series connection between a capacitor Cv,
varactor-loaded slot, and it can be calculated by
an inductor Lv, and a resistor Rv, as shown in Fig. 16(a). When
altering the reverse bias voltage applied to the two varactors Z L S = Z v ||Z s (11)
simultaneously, the adjustable capacitance Cv can change from
3.2 to 0.35 pF. The effective width of the reflective load can where Z v is the impedance of the varactor given in (9) and
be approximately calculated by [2] Z s is the impedance of the transverse slot that can be derived
from the reciprocal of the admittance of the slot Ys .
dv2
We f f = W ps − . (2) Therefore, the phase of the RTPS can be controlled by
0.95Sv varying the impedance of the reflective load Z L [Z min –Z max ],
As shown in Fig. 16(b), the reflective load is divided into which is realized by adjusting the capacitance Cv[Cmin –
five sections: three sections of transmission lines and two Cmax ] of the voltage-controlled varactors. The resulting phase

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.
 6902 IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, VOL. 67, NO. 11, NOVEMBER 2019

Fig. 18. Implementation of the SIW phase shifter.

when the bias voltage is adjusted from V0 to Vmax , the phase of

the  L covers the range from −180◦ to −α. By adjusting the
Fig. 17. Equivalent simplified modes of (a) reflective load Z L , (b) variable
load Z 1 , and (c) variable load Z 2 . length of the transmission line L 1 and L 2 , the aforementioned
two conditions shall be satisfied. Consequently, a 360◦ phase
variation of the load reflection coefficient can be achieved.
variation of the RTPS is equal to the phase variation of the
reflection coefficient C. Implementation
Z L − Z0 The setup of the reverse bias voltage applied to varactor
= . (12)
Z L + Z0 diodes through dc power supply is shown in Fig. 18. On the
The absolute value of the phase shift range is given by top surface of the phase shifter, the cathodes of one pair of
    varactor diodes are soldered to the inner rectangular patch and
Z max Z min connected to positive terminal of a dc power supply via dc wire
ϕ= 2 arctan −arctan . (13)
Z0 Z0 (red curve in Fig. 18). The anodes of the pair of varactors are
soldered to the metallized surface of the substrate (i.e., ground
Based on the above theoretical analysis, the variable para-
plane) and connected to ground terminal of the dc power
meters of the reflective loads We f f , L 0 , L 1 , L 2 , L s2 , and Ws2
supply via another dc wire (black curve in Fig. 18). The other
can be optimized to obtain the largest possible phase shift
pair of varactors on the bottom surface of the phase shifter
range with the help of the HFSS.
are placed in exactly the same way as that of the top surface
With respect to the realization mechanism of 360◦ phase
and connected to the same dc power supply. By adjusting the
variation, a simplified circuit model diagram is illustrated in
reverse bias voltage, junction capacitances of the two pairs of
Fig. 17. The impedance of the reflective load Z L can be seen
varactors can be altered simultaneously.
as series connection impedances of two variable loads Z 1
and Z 2 , as shown in Fig. 17(a). By varying the bias voltage
Vd , the input reflection coefficients of the loads Z 1 , Z 2 , and D. Performance
Z L have different values, which are denoted as 1 , 2 , and As discussed in Section III-B, theoretically, 360◦ phase
 L , respectively. Note that the reflection coefficient 1 counts variation of the proposed phase shifter can be achieved by
as the value when the variable load Z 1 is short circuited optimizing the configurations of the slot and waveguide. To
at the end (i.e., Z 2 = 0), and the reflection coefficient 2 verify the conclusion, simulations have been carried out by
counts as the value when the load Z 1 is short circuited (i.e., HFSS and the results show that the insertion loss increases
Z 1 = 0), as shown in Fig. 17(b) and (c), respectively. When to nearly 6 dB when the maximum phase variation is around
the variable loads Z 1 and Z 2 are taken into consideration 360◦. It is clear that the insertion loss is mainly caused by the
separately, the phase of the 1 at the highest control voltage parasitic resistance of the varactor. Considering the insertion
Vmax applied to the varactor of the variable load Z 1 should loss reduces the antenna gain, therefore the proposed phase
be equal to that of the 2 at the lowest control voltage Vmin shifter is designed to have a maximum insertion loss of 3.5 dB
applied to the varactor of the variable load Z 2 [11]. Such phase when it achieves a maximum phase variation of approximately
is denoted as −α. Then considering the overall reflective load 290◦ which could provide large enough phase difference for
Z L , when a certain control voltage V0 is applied to both the a phased-array antenna.
varactors, the two loads Z 1 and Z 2 are supposed to have The simulated return loss and insertion loss of the proposed
opposite reactance so that the corresponding series impedance phase shifter are plotted in Fig. 19. As can be seen, the return
Z L = Z 1 + Z 2 turns out to be a resistive impedance. That is to loss is better than 14 dB and the insertion loss is less
say, when the bias voltage is altered from Vmin to V0 , the phase than 3.5 dB over the operating frequency band, providing a
of the reflection coefficient  L varies from −α to 180◦ . Also, relative bandwidth of about 18.2%. The simulated phase of

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.
 JI et al.: RECONFIGURABLE PHASED-ARRAY ANTENNA USING CONTINUOUSLY TUNABLE SIW PHASE SHIFTER 6903

Fig. 21. Geometry of the proposed 1 × 4 SIW phased array.

Fig. 19. Simulated S-parameters of the SIW phase shifter versus the applied
voltage.

Fig. 20. Simulated phase of the reflective load of the SIW phase shifter
versus the applied voltage.

the reflective load versus the applied biasing voltage is shown

in Fig. 20. A continuous phase shift of 290◦ is obtained when
the control voltage varies from 0 to 20 V.

IV. P HASED A RRAY A NTENNA

A. Layout Fig. 22. Hierarchical structure diagram of the proposed 1 × 4 SIW phased
array.
A 1 × 4 SIW phased array is designed based on the
antenna element introduced in Section II and the phase shifter
presented in Section III. The geometry of the proposed phased
array is shown in Fig. 21. It comprised four ACPAs, four hollow pillars placed between the top substrate and the middle
phase shifters, one four-way power divider, and one GCPW-to- substrate.
SIW transition, which are constructed on a multilayered SIW For better understanding, a hierarchical structure diagram
structure. Three substrates are used in the phased array design: of the proposed phased array is illustrated in Fig. 22. The top
top substrate, middle substrate, and bottom substrate. Feed laminate is a support plate with four metal patches. Both of
port is located on the bottom substrate as an SMA connector the middle laminate and the bottom laminate are double-sided
is placed on one end of the bottom substrate. The GCPW- copper-clad laminates. It is notable that middle metal layer is
to-SIW transition is adopted to transmit signal into the SIW. made up of two metallized surfaces, namely “middle metal
Then, the SIW four-way power divider is connected with the layer1” and “middle metal layer2.” It is clear that the middle
transition. Four output ports of the power divider are connected metal layer1 is the back metallized surface of the middle sub-
with input ports of the four phase shifters. Next, each output strate, whereas the middle metal layer2 is the top metallized
port of the phase shifters on the middle substrate is connected surface of the bottom substrate. The layout of each metal layer,
with the corresponding input port of the antenna element. The as well as their thicknesses, is given in Fig. 23 and numerical
air-gap layer of the antenna element is introduced via four values of the dimensions are summarized in Table IV.

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.
 6904 IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, VOL. 67, NO. 11, NOVEMBER 2019

TABLE IV
D IMENSIONS OF THE P ROPOSED P HASED A RRAY (U NIT: mm)

Fig. 24. Simulated S-parameters of the GCPW-to-SIW transition.

To minimize the overall dimension, the four phase shifters are

closely placed with common edges. Thus, the element spacing
is 25.1 mm, corresponding to 0.46λ0 , where λ0 represents the
free-space wavelength in the center operating frequency.
The layout of the GCPW-to-SIW transition can be seen
in Fig. 23(c). A pair of fan-shaped slots is etched on the
metal surface for good impedance matching, and two rows
of metallic vias are inserted along both sides of GCPW for
Fig. 23. Layout of the proposed 1 × 4 SIW phased array. (a) Top view of the reducing electromagnetic leakage. Dimensions of the transition
top substrate, (b) top view of the top metal layer, (c) top view of the middle
metal layer2, and (d) side view. are given in Table IV. According to the simulated results shown
in Fig. 24, the bandwidth of a back-to-back type GCPW-to-
SIW transition is over 54.5% for the input reflection coefficient
less than −10 dB, and the transmission coefficient is about
B. Feeding Network
−0.5 dB throughout the whole operating frequency band.
Power is supposed to be divided into four parts with The four-way SIW power divider is shown in Fig. 23(c),
identical amplitude and phase after passing through the feeding which contains one T-type two-way power divider, two SIW
network. By applying different control voltages to the four bends, and two Y-type two-way power dividers. In the T-type
phase shifters individually, various excitation phase differences power divider, a metallic via is inserted to divide the input
are acquired for the four antenna elements. As a result, power into two parts equally. To reduce the reflection from
diverse beam-scanning angles of the phased array are realized. the SIW branches, another two metallic vias are placed in

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.
 JI et al.: RECONFIGURABLE PHASED-ARRAY ANTENNA USING CONTINUOUSLY TUNABLE SIW PHASE SHIFTER 6905

TABLE V
C ONTROL V OLTAGE S ETTINGS AND P ERFORMANCE
OF THE P HASED A RRAY

Fig. 25. Simulated S-parameters of the SIW feeding network.

C. Implementation
A prototype was fabricated and measured to validate the
proposed design, as shown in Fig. 26(a) and (b). The mea-
surement setup of the proposed phased array in anechoic
chamber is shown in Fig. 26(c). To clearly show how to
implement the phased array, the setup of reverse bias voltages
applied to each pair of varactors through dc power supply
is illustrated in Fig. 26(d). Four dc power supplies are used
to separately control the four phase shifters of the phased
array, which are named V1 , V2 , V3 , and V4 , respectively.
In a single phase shifter, the cathodes of the two pair of
varactor diodes on both the sides are soldered to the inner
rectangular patches and connected to positive terminal of a
dc power supply via dc wires [red curves in Fig. 26(d)]. The
anodes of the varactors are soldered to the ground plane of the
phased array and connected to ground terminal of the dc power
supply via another dc wire [black curve in Fig. 26(d)]. It is
noteworthy that the two pairs of varactors in the same phase
shifter are connected to the same dc power supply. In addition,
a 25 nH surface mount inductor, model Murata LQW18A,
is employed in series with each pair of varactors to isolate RF
signal between the antenna and biasing circuit. Following the
aforementioned procedure, the remaining three phase shifters
are connected to another three dc power supplies in the same
way. To sum up, by connecting varactors of each phase shifter
to their respective dc power supply, independent control of the
four phase shifters is realized.

D. Measurement and Discussion

Fig. 26. Fabricated prototype of the proposed 1 × 4 SIW phased array.
(a) Top view, (b) back view, (c) measurement setup in anechoic chamber, and The radiation performance of the phased-array antenna was
(d) implementation. measured by a Satimo Starlab near-field measurement system
and the reflection coefficients were measured by an Agilent
N5230A network analyzer.
the corner of the SIW bends. Similar strategy is applied in To steer the main beam direction accurately, a set of
the Y-type power divider design. By optimizing the diameter uniform phase differences between each array element are
of those vias and their positions, good performance of the realized by controlling the applied voltage on the basis of
power divider can be obtained. Dimensions of the four-way the simulated results given in Fig. 20. The simulated and
power divider are given in Table IV. The simulated results of measured radiation patterns under different biasing conditions
S-parameters of the feeding network are plotted in Fig. 25. (see Table V) in the H-plane and the E-plane are plotted
It can be seen that the operating bandwidth is 14.3% for in Fig. 27. It indicates that the proposed phased-array antenna
|S11 | < −20 dB (from 5.2 to 6 GHz). Simulated |S21 |, |S31 |, is capable of continuously steering the main beam directions
|S41 |, and |S51 | fluctuate between −6.3 and −6.4 dB. from −45◦ to 45◦ , maintaining sidelobe level (SLL) lower

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.
 6906 IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, VOL. 67, NO. 11, NOVEMBER 2019

Fig. 28. Simulated and measured reflection coefficients of the proposed

phased array at different scanning angles.

than −12 dB and cross-polarization level lower than −18 dB.

The measured results are found in reasonably good agreement
with the simulated ones. The slight asymmetry of the E-plane
patterns is due to the asymmetrical structure of the phased
array in the xoz plane. The simulated and measured peak
gains of the phased array covering the scan-angle range are
given in Table V. It is observed that the gain variation over
the entire 90◦ scanning angle range is about 2.5 dB, which
is mainly caused by the insertion loss variation of the phase
shifters. The measured peak gain is 11 dBi when all the
varactors remain in the unbiased state, whereas the simulated
peak gain is 12.5 dBi. Also, the gain decreases to a worst
case of 8.5 dBi when the main beam deflects to the low
elevation angle of 45◦ in the H-plane. The degradation of the
measured result is largely due to the parasitic effects of the
varactors, the added dc cables, manufacturing tolerances, and
measurement errors. The simulated and measured reflection
coefficients are given in Fig. 28, which are in good agree-
ment. The measured impedance bandwidth is about 16.1%
(5.09–5.98 GHz) for |S11 | < −10 dB.
For purposes of comparison, structure characteristics, and
performances of this work, recently reported SIW-based
phased arrays are listed in Table VI. Although the antenna
array in [5] offers comparable maximal beam-steering angles
(±38◦), bandwidth (11.6%), and gain (11.5 dBi), it occupies
a much larger SIW dimension and could not change the main
beam direction continuously. It is observed that the arrays
in [7] and [9] are both miniaturization designs, which have the
same beam-steering range of −19◦ to 19◦ , bandwidths of 0.5%
and 2.9%, and average gains of 6.4 and 4.9 dBi, respectively.
Compared with them, the proposed phased array provides a
significant performance boost on beam scanning range (±45◦),
bandwidth (16.1%), and gain (11 dBi), except for a relatively
larger size. The design in [8] has a wide beam-steering range
Fig. 27. Simulated and measured normalized radiation patterns of the of ±60◦ , but it needs external beam-forming network (phase
proposed phased array under different biasing conditions at 5.5 GHz. Scanning shifter, power divider, control computer, etc.) to make it work,
angle: (a) −45◦ , (b) −18◦ , (c) 0◦ , (d) 18◦ , and (e) 45◦ . which is bulky and not suitable for integration.

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.
 JI et al.: RECONFIGURABLE PHASED-ARRAY ANTENNA USING CONTINUOUSLY TUNABLE SIW PHASE SHIFTER 6907

TABLE VI
C OMPARISON B ETWEEN THE P ROPOSED AND R EPORTED SIW P HASED A RRAYS

V. C ONCLUSION [8] Y.-Q. Wen, B.-Z. Wang, and X. Ding, “Wide-beam SIW-slot antenna
for wide-angle scanning phased array,” IEEE Antennas Wireless Propag.
In this paper, a reconfigurable phased-array antenna incor- Lett., vol. 15, pp. 1638–1641, 2016.
porating continuously tunable SIW phase shifters is presented. [9] A. Nafe, F. A. Ghaffar, M. F. Farooqui, and A. Shamim, “A ferrite LTCC-
based monolithic SIW phased antenna array,” IEEE Trans. Antennas
By adopting the multi-resonant aperture-coupled scheme, the Propag., vol. 65, no. 1, pp. 196–205, Jan. 2017.
antenna element exhibits wideband characteristic and desir- [10] L. Ge, K. M. Luk, and S. Chen, “360◦ beam-steering reconfigurable
able radiation performance. Afterward, the electronically con- wideband substrate integrated waveguide horn antenna,” IEEE Trans.
Antennas Propag., vol. 64, no. 12, pp. 5005–5011, Dec. 2016.
trolled SIW phase shifter is designed to be easily integrated
[11] E. Sbarra, L. Marcaccioli, R. V. Gatti, and R. Sorrentino, “Ku-band
with the antenna element, offering a large phase varia- analogue phase shifter in SIW technology,” in Proc. 39th Eur. Microw.
tion range. Consequently, an SIW 1 × 4 phased array is Conf., Sep./Oct. 2009, pp. 264–267.
designed, fabricated, and measured to experimentally validate [12] Y. Ding and K. Wu, “Varactor-tuned substrate integrated waveguide
phase shifter,” in IEEE MTT-S Int. Microw. Symp. Dig., Jun. 2011,
the design approach. As the measured results demonstrate that pp. 1–4.
a beam-steering range of 90◦ with SLL lower than −12 dB, [13] H. Peng, P. Jiang, T. Yang, and H. Jin, “Continuously tunable SIW phase
a gain of up to 11 dBi and an impedance bandwidth of 16.1% shifter based on the buried varactors,” IEICE Electron. Express, vol. 12,
no. 7, Apr. 2015, Art. no. 20150165.
are achieved, and the proposed phased array possesses attrac-
[14] G. Q. Luo, Z. F. Hu, L. X. Dong, and L. L. Sun, “Planar slot antenna
tive features of broad scanning range and stable radiation backed by substrate integrated waveguide cavity,” IEEE Antennas
patterns. Furthermore, fabrication is realized by applying the Wireless Propag. Lett., vol. 7, pp. 236–239, 2008.
traditional PCB process, hence the monolithic structure of the [15] G. Q. Luo, Z. F. Hu, W. J. Li, X. H. Zhang, L. L. Sun, and J. F. Zheng,
“Bandwidth-enhanced low-profile cavity-backed slot antenna by using
phased array ensures high reliability and low cost, which has hybrid SIW cavity modes,” IEEE Trans. Antennas Propag., vol. 60,
practical value and market potential for 5G Wi-Fi applications. no. 4, pp. 1698–1704, Apr. 2012.
[16] K. Gong, Z. N. Chen, X. Qing, P. Chen, and W. Hong, “Substrate
integrated waveguide cavity-backed wide slot antenna for 60-GHz
R EFERENCES Bands,” IEEE Trans. Antennas Propag., vol. 60, no. 12, pp. 6023–6026,
Dec. 2012.
[1] D. Deslandes and K. Wu, “Integrated microstrip and rectangular [17] S. Mukherjee, A. Biswas, and K. V. Srivastava, “Broadband sub-
waveguide in planar form,” IEEE Microw. Wireless Compon. Lett., strate integrated waveguide cavity-backed bow-tie slot antenna,”
vol. 11, no. 2, pp. 68–70, Feb. 2001. IEEE Antennas Wireless Propag. Lett., vol. 13, pp. 1152–1155,
[2] Y. Cassivi, L. Perregrini, P. Arcioni, M. Bressan, K. Wu, 2014.
and G. Conciauro, “Dispersion characteristics of substrate integrated [18] Y. Shi, J. Liu, and Y. Long, “Wideband triple- and quad-resonance
rectangular waveguide,” IEEE Microw. Wireless Compon. Lett., vol. 12, substrate integrated waveguide cavity-backed slot antennas with shorting
no. 9, pp. 333–335, Sep. 2002. vias,” IEEE Trans. Antennas Propag., vol. 65, no. 11, pp. 5768–5775,
[3] F. Xu and K. Wu, “Guided-wave and leakage characteristics of substrate Nov. 2017.
integrated waveguide,” IEEE Trans. Microw. Theory Techn., vol. 53, [19] Y. Cai, Y. Zhang, C. Ding, and Z. Qian, “A wideband multilayer substrate
no. 1, pp. 66–73, Jan. 2005. integrated waveguide cavity-backed slot antenna array,” IEEE Trans.
[4] D. Deslandes and K. Wu, “Accurate modeling, wave mechanisms, and Antenna Propag., vol. 65, no. 7, pp. 3465–3473, Jul. 2017.
design considerations of a substrate integrated waveguide,” IEEE Trans. [20] D.-Y. Kim, J. W. Lee, T. K. Lee, and C. S. Cho, “Design of SIW cavity-
Microw. Theory Techn., vol. 54, no. 6, pp. 2516–2526, Jun. 2006. backed circular-polarized antennas using two different feeding transi-
[5] J. Zhu, B. Peng, and S. Li, “Cavity-backed high-gain switch beam tions,” IEEE Trans. Antennas Propag., vol. 59, no. 4, pp. 1398–1403,
antenna array for 60-GHz applications,” IET Micow. Antennas Propag., Apr. 2011.
vol. 11, no. 12, pp. 1776–1781, Sep. 2017. [21] Q. Wu, H. Wang, C. Yu, and W. Hong, “Low-profile circularly polarized
[6] Y. F. Wu, Y. J. Cheng, and Z. X. Huang, “Ka-band near-field-focused cavity-backed antennas using SIW techniques,” IEEE Trans. Antennas
2-D steering antenna array with a focused Rotman lens,” IEEE Trans. Propag., vol. 64, no. 7, pp. 2832–2839, Jul. 2016.
Antennas Propag., vol. 66, no. 10, pp. 5204–5213, Oct. 2018. [22] T. Y. Yang, W. Hong, and Y. Zhang, “Wideband millimeter-
[7] P. Loghmannia, M. Kamyab, M. R. Nikkhah, and R. Rezaiesarlak, wave substrate integrated waveguide cavity-backed rectangular patch
“Miniaturized low-cost phased-array antenna using SIW slot elements,” antenna,” IEEE Antennas Wireless Propag. Lett., vol. 13, pp. 205–208,
IEEE Antennas Wireless Propag. Lett., vol. 11, pp. 1434–1437, 2012. 2014.

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.
 6908 IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, VOL. 67, NO. 11, NOVEMBER 2019

[23] T. Zhang, W. Hong, Y. Zhang, and K. Wu, “Design and analysis of Jianpeng Wang received the Ph.D. degree from the
SIW cavity backed dual-band antennas with a dual-mode triangular-ring University of Electronic Science and Technology of
slot,” IEEE Trans. Antennas Propag., vol. 62, no. 10, pp. 5007–5016, China, Chengdu, China, in 2007.
Oct. 2014. From 2005 to 2006, he was a Research Assis-
[24] K.-S. Chin, W. Jiang, W. Che, C.-C. Chang, and H. Jin, “Wideband tant with the Institute for Infocomm Research,
LTCC 60-GHz antenna array with a dual-resonant slot and patch Singapore. From 2010 to 2011, he was a Research
structure,” IEEE Trans. Antennas Propag., vol. 62, no. 1, pp. 174–182, Fellow with the School of Electrical and Elec-
Jan. 2014. tronic Engineering, Nanyang Technological Uni-
[25] W. M. Abdel-Wahab and S. Safavi-Naeini, “Wide-bandwidth 60-GHz versity, Singapore. In 2013, he joined the School
aperture-coupled microstrip patch antennas (MPAs) fed by substrate of Engineering and Physical Sciences, Heriot Watt
integrated waveguide (SIW),” IEEE Antennas Wireless Propag. Lett., University, Edinburgh, U.K., as a Visiting Scholar.
vol. 10, pp. 1003–1005, 2011. In 2014 and 2016, he was a Research Fellow with the Faculty of Science and
[26] J. Zang, E. Carrasco, X. Wang, J. A. Encinar, A. Alvarez-Mclcon, and Technology, University of Macau, Macau, China. He is currently an Associate
J. S. Gomez-Diaz, “SIW-based reflectarray antennas with sharp gain Professor with the School of Electronic and Optical Engineering, Nanjing
selectivity and large bandwidth,” in Proc. IEEE Int. Symp. Antennas University of Science and Technology, Nanjing, China. He has authored or
Propag. USNC/URSI Nat. Radio Sci. Meeting, Boston, MA, USA, coauthored over 90 papers in international journals and conference proceed-
Jul. 2018, pp. 1405–1406. ings. His current research interests include microwave circuits, antennas, and
[27] Y. Li and K.-M. Luk, “Low-cost high-gain and broadband substrate- low-temperature co-fired ceramic (LTCC)-based millimeter-wave circuits.
integrated-waveguide-fed patch antenna array for 60-GHz band,” IEEE Dr. Wang has been an Associate Editor of the IET Electronics Letters
Trans. Antennas Propag., vol. 62, no. 11, pp. 5531–5538, Nov. 2014. since 2015.
[28] Y. Li and K.-M. Luk, “60-GHz substrate integrated waveguide fed
cavity-backed aperture-coupled microstrip patch antenna arrays,” IEEE
Trans. Antennas Propag., vol. 63, no. 3, pp. 1075–1085, Mar. 2015.
[29] K.-F. Lee, K. Ho, and J. Dahele, “Circular-disk microstrip antenna Quangang Chen was born in Puyang, Henan,
with an air gap,” IEEE Trans. Antennas Propag., vol. AP-32, no. 8, China, in 1992. He received the B.S. degree in elec-
pp. 880–884, Aug. 1984. tronic information engineering and the M.S. degree
[30] Ansys Corporation, Canonsburg, PA, USA. HFSS: High Frequency in electromagnetic field and microwave technology
Structure Simulator Based on the Finite Element Method. [Online]. from the Nanjing University of Science and Technol-
Available: http://www.ansys.com/ ogy, Nanjing, China, in 2015 and 2018, respectively.
[31] A. A. Oliner, “The impedance properties of narrow radiating slots in He is currently a Research Assistant with the
the broad face of rectangular waveguide: Part I–theory,” IRE Trans. College of Electronic Science and Technology,
Antennas Propag., vol. 5, pp. 4–11, 1957. Shenzhen University, Shenzhen, China. His cur-
rent research interests include multiband smartphone
antennas and multi-input multi-output antennas.

Wen Wu (SM’10) received the Ph.D. degree in elec-

tromagnetic field and microwave technology from
Yuan Ji was born in Jiangsu, China. She received Southeast University, Nanjing, China, in 1997.
the B.S. degree in information engineering from He is currently a Professor with the School of
Southeast University, Nanjing, China, in 2009 and Electronic and Optical Engineering, Nanjing Univer-
the M.S. degree in telecommunication engineering sity of Science and Technology, Nanjing, where he
from the University of Melbourne, Melbourne, VIC, is also an Associate Director with the Ministerial
Australia, in 2011. She is currently pursuing the Key Laboratory of JGMT. He has authored or coau-
Ph.D. degree in electronic science and technology thored over 300 internationally referred journal and
with the Nanjing University of Science and Tech- conference papers and holds 14 patents. His current
nology, Nanjing. research interests include microwave and millimeter-
Her current research interests include phased wave (mm-wave) theories and technologies, microwave and mm-wave detec-
arrays and reconfigurable antennas. tion, and multimode compound detection.
Dr. Wu was a six-time recipient of the Ministerial and Provincial-Level
Science and Technology Award.

Yujian Li (S’12–M’15) was born in Hunan, China,

Lei Ge (S’11–M’15–SM’18) was born in Jiangsu, in 1987. He received the B.S. and M.S. degrees in
China. He received the B.S. degree in electronic communications engineering from Beijing Jiaotong
engineering from the Nanjing University of Science University, Beijing, China, in 2009 and 2012, respec-
and Technology, Nanjing, China, in 2009 and the tively, and the Ph.D. degree in electronic engineering
Ph.D. degree in electronic engineering from the City from the City University of Hong Kong, Hong Kong,
University of Hong Kong, Hong Kong, in 2015. in 2015.
From 2010 to 2011, he was a Research Assistant In 2015, he joined the Institute of Lightwave
with the City University of Hong Kong, where he Technology, Beijing Jiaotong University, as an Asso-
was a Post-Doctoral Research Fellow with the State ciate Professor. His current research interests include
Key Laboratory of Millimeter Waves, from 2015 to millimeter-wave antennas, base station antennas, and
2015. He is currently an Assistant Professor and an leaky wave structures.
Associate Head with the Department of Electronic Engineering, Shenzhen Dr. Li was a recipient of the Outstanding Research Thesis Award from the
University, Shenzhen, China. His current research interests include wideband City University of Hong Kong in 2015, the Best Paper Award at the 2015 IEEE
antennas, patch antennas, base station antennas, reconfigurable antennas, Asia–Pacific Conference on Antennas and Propagation, the Best Student
the antennas for cognitive radio, and filtering antennas. Paper at the 2013 National Conference on Antennas, and the Best Student
Dr. Ge was a TPC Member of the APCAP 2016. He was a recipient of Paper Award (Second Prize) at the 2013 IEEE International Workshop on
the Honorable Mention at the Student Contest of the 2012 IEEE APS-URSI Electromagnetics (iWEM). He was selected as a Finalist in the Student Paper
Conference and Exhibition held in Chicago, IL, USA, and the First Prize in Contest of the 2015 IEEE AP-S Symposium on Antennas and Propagation.
the Student Innovation Competition of the 2014 IEEE International Workshop He served as a Reviewer for the IEEE T RANSACTIONS ON A NTENNAS AND
on Electromagnetics (IEEE iWEM) held in Sapporo, Japan, in 2014. He was P ROPAGATION, IEEE A NTENNAS AND W IRELESS P ROPAGATION L ETTERS ,
a Session Chair of the iWEM 2017 and ACES-China 2017. and the IET Microwaves, Antennas and Propagation.

Authorized licensed use limited to: Indian Institute of Space Science And Technology. Downloaded on February 27,2025 at 06:03:45 UTC from IEEE Xplore. Restrictions apply.