In the case of a CP microstrip antenna an innovative radiation efficiency analysis using the Wheeler cap method was presented in Nascimento & Lacava, 2009.. 1989, Microstrip patch antenn
Trang 1Design of Low-Cost Probe-Fed Microstrip Antennas 19 and a 75-mm square ground plane was designed using the HFSS software for operation at 1.603 GHz The optimized antenna dimensions are shown in Fig 28(a), the simulated input impedance and axial ratio results are presented in Fig 28(b) and the reflection coefficient magnitude in Fig 29 As expected, the microstrip antenna with the new geometry exhibits very good AR (0.1 dB) and reflection coefficient magnitude (-48 dB) characteristics at 1.603 GHz, without the need for any external matching network
1.40 1.45 1.50 1.55 1.60 1.65 1.70 1.75 1.80 -30
-25 -20 -15 -10 -5 0
0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0
mobile-terminals require two radiators The first is designed for uplink frequencies (Tx - 1.61073 to 1.62549 GHz) while the other receives the downlink ones (Rx - 2.48439 to 2.49915 GHz) (Nascimento et al., 2007a) The antenna geometry and a photo of the prototype are shown in Figs 30(a) and (b), respectively
The optimized antenna dimensions (using the HFSS software) are presented on Table 2 for the
radiators designed on finite ground plane and dielectric (L = 140 mm; W = 85 mm)
Trang 2x y
The axial ratio and reflection coefficient magnitude are presented in Figs 31 and 32 for the
Tx and Rx antennas, respectively
0 1 2 3 4 5 6 7 8 9
-55 -50 -45 -40 -35 -30 -25 -20 -15 -10
Fig 31 Globalstar antenna axial ratio and reflection coefficient magnitude: Tx radiator
Trang 3Design of Low-Cost Probe-Fed Microstrip Antennas 21
0 1 2 3 4 5 6 7 8 9
-50 -45 -40 -35 -30 -25 -20 -15 -10 -5
Fig 32 Globalstar antenna axial ratio and reflection coefficient magnitude: Rx radiator Results for the input impedance on the Smith chart are presented in Figs 33 and 34 for the
Tx and Rx antennas, respectively These results indicate that the antenna meets the Globalstar specifications
-10j 10j
-25j 25j
1.540 GHz
1.696 GHz 1.618 GHz
Fig 33 Globalstar antenna input impedance: Tx radiator
Trang 410 25 50 100 250 -10j
10j
-25j 25j
2.492 GHz
2.210 GHz 2.774 GHz
Fig 34 Globalstar antenna input impedance: Rx radiator
5.3 CP antenna radiation efficiency measurements
The radiation efficiency of a LP microstrip antenna can be efficiently measured using the
Wheeler cap (Choo et al., 2005; Pozar & Kaufman, 1988; Sona & Rahmat-Samii 2006)
According to Wheeler, the radiation resistance of an antenna can be separated from its loss
resistance by enclosing the antenna with a radiation shield cap placed at a distance greater
than λ/(2π) (Wheeler, 1959) Consequently, since a linearly-polarized microstrip antenna can
be modeled as a parallel RLC circuit, its efficiency is calculated by
where G cap is the conductance of the admittance measured with the cap in place and G out is
the conductance of the admittance measured with the cap removed
In the case of a CP microstrip antenna an innovative radiation efficiency analysis using the
Wheeler cap method was presented in (Nascimento & Lacava, 2009) This procedure is
discussed next, for the case of the Glonass antenna designed in Section 5.2.1
Differently from the standard design, the two orthogonal resonant modes in the new
approach are now asymmetrically positioned in relation to the frequency for optimal axial
ratio as presented in Fig 28 (b) In addition, at the lower resonant frequency (1.468 GHz), its
15.45-dB axial ratio shows the antenna tends to be linearly polarized around this frequency
This result supports the use of the Wheeler cap method for measuring the antenna radiation
efficiency at this frequency
The cap geometry is shown in Fig 35 where the radiator is positioned inside a cubic cavity
of electrically conducting walls of 270-mm internal dimension
Trang 5Design of Low-Cost Probe-Fed Microstrip Antennas 23
Fig 35 Geometry of the Wheeler cap simulation through the HFSS package
HFSS simulation results for the real part of the input impedance are presented in Fig 36, both with and without the cubic cavity Making use of equation (11) for the lower resonant mode (G cap = 1.92 mS and G out = 7.43 mS), the radiation efficiency computed from the Wheeler method is 74.16% The free-space radiation efficiency, computed with the HFSS package is 74.68% at 1.468 GHz, which is only 0.7% off Consequently, the Wheeler cap method can be used for accurately determining the radiation efficiency of TCRP radiators
0 60 120 180 240 300 360 420 480
Trang 6methodologies are based on properties of the antenna equivalent circuit, they can be applied
to the design of microstrip radiators of arbitrary patch shapes Moreover, it is not restricted
to low-cost substrate thus applying equally well to the design of LP or CP microstrip
patches printed on any moderately thick commercial microwave laminates Experimental
results for LP and CP radiators validate the design strategies for both the LP and CP cases
Moreover, the Wheeler cap method is shown to be an effective means for simulating the
radiation efficiency of CP microstrip antennas
The excellent practical results obtained when matching microstrip patch radiators to a 50-Ω
SMA connector can be readily extended to the synthesis of inductive or capacitive input
impedances, as for example in the case of optimization of the noise figure and stability of
low-noise power amplifiers connected directly to the antenna Another possible application
is the design of low-cross-polarization probe-fed microstrip arrays (Marzall, et al., 2009;
Marzall et al., 2010)
7 References
Alexander, M J (1989) Capacitive matching of microstrip antennas IEE Proceedings of
Microwaves, Antennas and Propagation, Vol 137, No 2, (Apr 1989) (172-174), ISSN:
0950-107X
Chang, F S & Wong, K L (2001), A broadband probe-fed patch antenna with a thickened
probe pin, Proceedings of Asia-Pacific Microwave Conference, (1247-1250), ISBN:
0-7803-7138-0, Taipei, China, Dec 2001
Chen, H M.; Lin, Y F.; Cheng, P S.; Lin, H H.; Song, C T P & Hall, P S (2005), Parametric
study on the characteristics of planar inverted-F antenna IEE Proceedings of
Microwaves, Antennas and Propagation, (Dec 2005) (534-538), ISSN: 1350-2417
Choo, H.; Rogers, R & Ling, H (2005), Comparison of three methods for the measurement
of printed antennas efficiency, IEEE Transactions on Antennas and Propagation, Vol
53, No 7, (Jul 2005) (2328-2332), ISSN: 0018-926X
Dahele, J S.; Hall, P S & Haskins, P M (1989), Microstrip patch antennas on thick
substrates, Proceedings of Antennas and Propagation Society International Symposium,
pp 458-462, San Jose, CA, USA, Jun 1989
Engest, B & Lo, Y T (1985), A study of circularly polarized rectangular microstrip
antennas, Technical Report, Electromagnetics Laboratory, University of Illinois
Gardelli, R.; La Cono, G & Albani, M (2004), A low-cost suspended patch antenna for
WLAN access points and point-to-point links, IEEE Antennas and Wireless
Propagation Letters, Vol 3, (2004) (90-93), ISSN: 1536-1225
Garg, R.; Bhartia, P.; Bahl, I & Ittipiboon, A (2001) Microstrip Antenna Design Handbook,
Artech House, ISBN: 0-89006-513-6, Boston
Hall, P S (1987) Probe compensation in thick microstrip patches. Electronics Letters, Vol 23,
No 11, (May 1987) (606-607), ISSN: 0013-5194
Haskins, P M & Dahele, J S (1998), Capacitive coupling to patch antenna by means of
modified coaxial connectors, Electronics Letters, Vol 34, No 23, (Nov 1998)
(2187-2188), ISSN: 0013-5194
HFSS (2010), Product overview, Available: http://www.ansoft.com/products/hf/hfss/, (Sept 2010)
Trang 7Design of Low-Cost Probe-Fed Microstrip Antennas 25 IEEE Std 145 (1993) IEEE Standard Definitions of Terms for Antennas, ISBN: 1-55937-317-2,
New York, USA
James, J R & Hall, P S (1989) Handbook of Microstrip Antennas, Peter Peregrinus, ISBN:
0-86341-150-9, London
Lee , K F & Chen, W (1997) Advances in Microstrip and Printed Antennas, John Wiley, ISBN:
0-471-04421-0, New York
Lumini, F.; Cividanes, L & Lacava, J C S (1999), Computer aided design algorithm for
singly fed circularly polarized rectangular microstrip patch antennas, International Journal of RF and Microwave Computer-Aided Engineering, Vol 9, No 1, (Jan 1999)
(32-41), ISBN: 1096-4290
Marzall, L F., Schildberg, R & Lacava, J C S (2009), High-performance,
low-cross-polarization suspended patch array for WLAN applications, Proceedings of Antennas and Propagation Society International Symposium, pp 1-4, ISBN: 978-1-4244-3647-7,
Charleston, SC, USA, June 2009
Marzall, L F., NascimentoD.C., Schildberg, R & Lacava, J C S (2010), An effective strategy
for designing probe-fed linearly-polarized thick microstrip arrays with symmetrical return loss bandwidth, PIERS Online, Vol 6, No 8, (July 2010) (700-704), ISSN:
1931-7360
Nascimento, D C.; Mores Jr., J.A.; Schildberg, R & Lacava, J C S (2006), Low-cost
truncated corner microstrip antenna for GPS application, Proceedings of Antennas and Propagation Society International Symposium, pp 1557-1560, ISBN: 1-4244-0123-2,
Albuquerque, NM, USA, July 2006
Nascimento, D C.; Bianchi, I.; Schildberg, R & Lacava, J C S (2007a), Design of probe-fed
truncated corner microstrip antennas for Globalstar system, Proceedings of Antennas and Propagation Society International Symposium, pp 3041-3044, ISBN: 978-1-4244-
0877-1, Honolulu, HI, USA, June 2007
Nascimento, D C.; Schildberg, R & Lacava, J C S (2007b), New considerations in the
design of low-cost probe-fed truncated corner microstrip antennas for GPS applications, Proceedings of Antennas and Propagation Society International Symposium,
pp 749-752, ISBN: 978-1-4244-0877-1, Honolulu, HI, USA, June 2007
Nascimento, D C.; Schildberg, R & Lacava, J C S (2008) Design of low-cost microstrip
antennas for Glonass applications PIERS Online, Vol 4, No 7, (2008) (767-770),
ISSN: 1931-7360
Nascimento, D C & Lacava, J C S (2009), Circularly-polarized microstrip antenna
radiation efficiency simulation based on the Wheeler cap method, Proceedings of Antennas and Propagation Society International Symposium, pp 1-4, ISBN: 978-1-4244-
3647-7, Charleston, SC, USA, June 2009
Niroojazi, M & Azarmanesh, M N (2004), Practical design of single feed truncated corner
microstrip antenna, Proceedings of Second Annual Conference on Communication Networks and Services Research, 2004, pp 25-29, ISBN: 0-7695-2096-0, Fredericton,
NB, Canada, May 2004
Pozar, D M & Kaufman, B (1988), Comparison of three methods for the measurement of
printed antennas efficiency, IEEE Transactions on Antennas and Propagation, Vol 36,
No 1, (Jan 1988) (136-139), ISSN: 0018-926X
Trang 8Richards, W F.; Lo, Y T & Harrison, D D (1981), An improved theory for microstrip
antennas and applications, IEEE Transactions on Antennas and Propagation, Vol 29,
No 1, (Jan 1981) (38-46), ISSN: 0018-926X
Sona, K S & Rahmat-Samii, Y (2006), On the implementation of Wheeler cap method in
FDTD, Proceedings of Antennas and Propagation Society International Symposium, pp
1445-1448, ISBN: 1-4244-0123-2, Albuquerque, NM, USA, July 2006
Teng, P L.; Tang, C L & Wong, K L (2001), A broadband planar patch antenna fed by a
short probe feed, Proceedings of Asia-Pacific Microwave Conference, pp 1243-1246,
ISBN: 0-7803-7138-0, Taipei, China, Dec 2001
Tinoco S., A F.; Nascimento, D C & Lacava, J C S (2008), Rectangular microstrip antenna
design suitable for undergraduate courses, Proceedings of Antennas and Propagation
Society International Symposium, pp 1-4, ISBN: 978-1-4244-2041-4, San Diego, CA,
USA, July 2008
Tzeng, Y B.; Su, C W & Lee, C H (2005), Study of broadband CP patch antenna with its
ground plane having an elevated portion, Proceedings of Asia-Pacific Microwave
Conference, pp 1-4, ISBN: 0-7803-9433-X, Suzhou, China, Dec 2005
Vandenbosch, G A E & Van de Capelle, A R (1994), Study of the capacitively fed
microstrip antenna element, IEEE Transactions on Antennas and Propagation, Vol 42,
No 12, (Dec 1994) (1648-1652), ISSN: 0018-926X
Volakis, J L (2007) Antenna Engineering Handbook 4th ed., McGraw-Hill, ISBN:
0-07-147574-5, New York
Wheeler, H A (1959), The radiansphere around a small antenna, Proceedings of the IRE, Vol
47, No 8, (Aug 1959) (1325-1331), ISSN: 0096-8390
Trang 9et al 1991) It is found that the use of such materials may have a beneficial effect on circuit or antenna (Bhartia et al 1991; Pozar, 1987) For a rigorous solution to the problem of a rectangular microstrip antenna, which is the most widely used configuration because its shape readily allows theoretical analysis, Galerkin’s method is employed in the spectral domain with two sets of patch current expansions One set is based on the complete set of orthogonal modes of the magnetic cavity, and the other employs Chebyshev polynomials with the proper edge condition for the patch currents (Tulintsef et al 1991)
This chapter describes spectral domain analyses of a rectangular microstrip patch antenna that contains isotropic or anisotropic substrates in which entire domain basis functions are used to model the patch current, we will present the effect of uniaxial anisotropy on the characterization of a rectangular microstrip patch antenna, also because there has been very little work on the scattering radar cross section of printed antennas in the literature, including the effect of a uniaxial anisotropic substrate, a number of results pertaining to this case will be presented in this chapter
Trang 102 Theory
An accurate design of a rectangular patch antenna can be done by using the Galerkin
procedure of the moment method (Pozar, 1987; Row & Wong, 1993; Wong et al., 1993) An
integral equation can be formulated by using the Green’s function on a thick dielectric
substrate to determine the electric field at any point
The patch is assumed to be located on a grounded dielectric slab of infinite extent, and the
ground plane is assumed to be perfect electric conductor, the rectangular patch with length a
and width b is embedded in a single substrate, which has a uniform thickness of h (see Fig
1), all the dielectric materials are assumed to be nonmagnetic with permeability μ0 To
simplify the analysis, the antenna feed will not be considered
The study of anisotropic substrates is of interest, however, the designers should, carefully
check for the anisotropic effects in the substrate material with which they will work, and
evaluate the effects of anisotropy
Fig 1 Geometry of a rectangular microstrip antenna
Anisotropy is defined as the substrate dielectric constant on the orientation of the applied
electric field Mathematically, the permittivity of an anisotropic substrate can be represented
by a tensor or dyadic of this form (Bhartia et al., 1991)
yz yy yx
xz xy xx 0
εεε
εεε
εεε.ε
ε00
0ε0
00ε.ε
0
a Plan view
y
x z
h
z
x , ε ε
Radiating conductor
0
a b
b Cross sectional view Ground plan
Trang 11Analysis of a Rectangular Microstrip Antenna on a Uniaxial Substrate 29
For a uniaxially anisotropic substrate the permittivity is
ε00
0ε0
00ε.ε
ε is the relative permittivity in the direction of the optical axis
Many substrate materials used for printed circuit antenna exhibit dielectric anisotropy,
especially uniaxial anisotropy (Bhartia et al 1991; Wong et al., 1993)
In the following, the substrate material is taken to be isotropic or uniaxially anisotropic with
the optical axis normal to the patch
The boundary condition on the patch is given by (Pozar, 1987)
0
inc scat+ E =
inc
E Tangential components of incident electric field
scat
E Tangential components of scattered electric field
While it is possible to work with wave equations and the longitudinal components E~ and z
z
H~ , in the Fourier transform domain, it is desired to find the transverse fields in the (TM,
TE) representation in terms of the longitudinal components Assuming an ei ω t time
variation, thus Maxwells equations
E E
∂
∂+
∂
∂+
∂
∂+
∂
∂+
ω is the angular frequency
From the above equations and after some algebraic manipulation, the wave equations for
z
E and H z are respectively
0kεzε
εyx
2 z 2 2
z
x 2 2 2
2
=+
∂
∂+
∂
∂+
∂
∂
z z
z
E
(9)
Trang 12εyx
2 z 2 2
z
x 2 2 2
2
=+
∂
∂+
∂
∂+
∂
∂
z z
z
H
(10) With
0
k propagation constant for free space, k =0 ω ε0μ0
By assuming plane wave propagation of the form e± i k x xe± i y ye± i z z
A Fourier transform pair of the electric field is given by (Pozar, 1987)
,k,k
~ i k x i y
z y x
y x
2 ~ k ,k ,k e e dk dkπ
4
1zy,
,k,k
z y x
y x
2 ~ k ,k ,k e e dk dkπ
4
1zy,
It is worth noting that ~ is used to indicate the quantities in spectral domain
In the spectral domain ikx
After some straightforward algebraic manipulation the transverse field can be written in
terms of the longitudinal components E~ , z H~ z
z 2 s
z 2 s
k
E k
z
~εkεi
x x
z 2 s
k
E k
z
~ε
kεi
~kεεω
~ z 0 y x
∂
∂+
−
2 s z 2 s x
H k E k
z
~ki
~kεεω
~ z 0 x y
∂
∂+
−
2 s z 2 s y
H k
E k
s
k is the transverse wave vector, k s=kxxˆ+kyyˆ, ks= k s
kx and ky are the spectral variables corresponding to x and y respectively
From the wave equations (9) and (10), the general form of E~ and z H~ is z
Trang 13Analysis of a Rectangular Microstrip Antenna on a Uniaxial Substrate 31
z i z
i z e ze
~
1 1
z C D
z i z
i z e ze
~
2 2
z C D
C 1 , D 1 , C 2 and D 2 are the unknowns to be determined
By substitution of (19) and (20) in (15)-(18) and after some algebraic manipulation the
transverse field in the (TM, TE) representation can be written by
s h s s e s s
k E k E k
s z
ez,
~ ,z
~z,
h s s e s s
k H k H k
z,
~ ,z
~z,
The superscripts e and h denote the TM and the TE waves, respectively
A and B are two unknowns vectors to be determined, note that are expressed in terms of C 1,
e z x 0
μω
k0
0kεεω
ε
εkε
k =⎜⎜⎛ − ⎟⎟⎞ and ( )2
1 2 s 2 0 x h
By eliminating the unknowns A and B, in the equations (21) and (22) we obtain the
following matrix which combines the tangential field components on both sides z1 and z2 of
the considered layer as input and output quantities
−
−
−
s e(h) s
s s
s
k J k
H k E I
g
g I
,
~ ,z
~hkcoshksini
hksinihkcosz
,
~ ,z
~
1 h 1 h e(h)
z e(h)
z
e(h) z 1 e(h)
z 2
h
2 h
(24)
I is the unit matrix
( )s
e(h) k
J~ is the current on the patch
In the spectral domain the relationship between the patch current and the electric field on
the patch is given by
Trang 14hksinkkiω
1
z1 z x z1 e
z1 z e z 0
e
+
−
=ε
(k h) kcos(k h)sin
ik
hksinkiω
1
z1 h z z1 z
z1 2
In the case of the isotroipc substrate
( ) ( )
z1 0
0 e
khkcotkεi1
hkcosε
khkcotki1hkcos
1ε
k = − ( )k s
J~ is the current on the patch which related to the vector Fourier transform of J(rs), as
(Chew & Liu, 1988)
ekkkkk
1, , r s=x +xˆ yyˆ (28)
xˆ unit vector in x direction
yˆ unit vector in y direction
The surface current on the patch can be expanded into a series of known basis functions Jxn
1 n
xn
0b0
Ja
s
s
r r
k k J k
k
1 m m x
y xn
N
1 n y
k
k1
~ak
k1
Trang 15Analysis of a Rectangular Microstrip Antenna on a Uniaxial Substrate 33
( )k s
Jxn
~ and ~Jym( )k s are the Fourier transforms of Jxn( )r s and Jym( )r s respectively
One of the main problems with the computational procedure is to overcome the complicated
time-consuming task of calculating the Green’s functions in the procedure of resolution by
the moment method The choice of the basis function is very important for a rapid
convergence to the true values (Boufrioua & Benghalia, 2008; Boufrioua, 2009)
Many subsequent analyses involve entire-domain basis functions that are limited to
canonical shapes such as rectangles, circles and ellipses Recently, much work has been
published regarding the scattering properties of microstrip antennas on various types of
substrate geometries Virtually all this work has been done with entire domain basis
functions for the current on the patch
For the resonant patch, entire domain expansion currents lead to fast convergence and can
be related to a cavity model type of interpretation (Boufrioua, 2009; Pozar & Voda, 1987)
The currents can be defined using a sinusoid basis functions defined on the whole domain,
without the edge condition (Newman & Forrai, 1987; Row & Wong, 1993), these currents
associated with the complete orthogonal modes of the magnetic cavity Both x and y
directed currents were used, with the following forms (Chew & Liu, 1988; Row & Wong,
axaπn
axaπm
a/2
1 x ik xn
2
bybπncosedy2
axaπnsinedx
a/2
1 x ik ym
2
bybπmsinedy2
axaπmcosedx
s
k
Since the chosen basis functions approximate the current on the patch very well for
conventional microstrips, only one or two basis functions are used for each current
component
Using the equations (32.a) and (32.b), the integral equation describing the field E in the patch
can be discretized into the following matrix
Z
Z Z
(33) Where the impedance matrix terms are
k
1
d 2 e 2 h xk xn2
s kn