Progress In Electromagnetics Research, Vol.

107, 147–159, 2010

A NOVEL APPROACH FOR RCS REDUCTION USING A

COMBINATION OF ARTIFICIAL MAGNETIC CONDUC-
TORS

M. E. de Cos, Y. Álvarez, and F. Las-Heras

Área de Teorı́a de la Señal y Comunicaciones
Departamento de Ingenierı́a Eléctrica
Universidad de Oviedo
Edificio Polivalente, Modulo 8, Campus Universitario de Gijón
Gijón, Asturias E-33203, Spain

Abstract—A thin Artificial Magnetic Conductor (AMC) structure for

Radar Cross-Section (RCS) reduction applications is presented. The
manufactured prototype, which combines two unit-cell metallization
sizes, presenting two resonant frequencies, shows broad AMC operation
bandwidth, polarization angle independency, and its angular margin
when operating under oblique incidence is also tested. It is shown
that significant RCS reduction can be achieved with the proposed
AMCs combination even if a 180◦ phaseshift between reflected waves
is not met. Two designs are considered: the already mentioned design
combining AMCs with overlapped frequency bands and the second
one combining Perfect Electric Conductor (PEC) and AMC surfaces.
A comparison between these two designs regarding RCS reduction,
supported by measurements in an anechoic chamber, is presented.

1. INTRODUCTION

Radar Cross Section [1–4] reduction has received a lot of attention for
many years, being the military applications (stealth technology) among
the more attractive ones. The aim is the same in all the scenarios:
to reduce the scattered electromagnetic field when illuminating the
object-of-interest by an incident wave. There have been three
remarkable different solutions proposed for RCS reduction: the use
of Radar/Radiation Absorbing Material (RAM), object shaping, and
object coating.
Received 4 June 2010, Accepted 15 July 2010, Scheduled 4 August 2010
Corresponding author: M. E. de Cos ([email protected]).
148 De Cos, Álvarez, and Las-Heras

RAM materials [5, 6], widely used in antenna and RCS

measurement facilities in order to avoid echoes in floor and walls,
transform the electromagnetic energy into heat, using conductor
materials for this purpose. The main drawback of RAMs is their size,
which fixes the lower frequency at which an antenna/RCS measurement
facility can operate [7].
Another approach when dealing with RCS reduction is object
shaping, which aims to reduce the scattered field level in the incoming
field direction [8]. This technique is quite complex, and it usually
requires a trade-off between operational features and RCS reduction
specifications.
The third technique for RCS reduction is based on objects’
coating [9, 10], the creation of multiple reflected waves canceling each
other as its operational principle.
The fine tuning of the Salisbury screen [11] parameters (a
combination of RAM and coating techniques) will result in a zero
reflected field [14]. However, this sheet, which consists of a resistive
dielectric sheet placed λ/4 above a PEC surface, exhibits main
drawbacks as its narrow bandwidth and the overall thickness (λ/4),
which limits its practical applications. Improvements in bandwidth
and thickness using Frequency Selective Surfaces (FSS) have been
recently shown in [12, 13].
The use of simple artificial magnetic surfaces, whose reflection
coefficient is ρ = +1, has been discussed [14–16], highlighting the fact
of thickness reduction with respect to the Salisbury screen. Other
works [17–19] analyze the effect of coating objects with more complex
metamaterials.
An arrangement of PEC (showing ρ = −1) and AMC (ρ = +1)
surfaces in a chessboard-like structure has already been proposed [20–
22], with the aim of obtaining two reflected waves with a 180◦ phase-
shift between them. Combination of these reflected waves yields a
destructive interference, reducing the scattered field level. The idea
presented in [22] is followed in [23], proposing a combination of two
AMCs with different resonance frequencies, which yields two operating
bands where the RCS of the manufactured surface is below −15 dB [23].
Destructive interferences requires not only a 180◦ -phaseshift
between the waves reflected on the PEC and AMC surfaces, but also
the same reflection coefficient amplitude, |ρ|PEC = |ρ|AMC . At this
point, it is important to remark the fact that in the case of AMC
surfaces, it may happen that |ρ|AMC < 1, so PEC-AMC combination
may not generate two reflected waves completely canceling each other.
Thus, when combining AMC structures for RCS reduction
purposes, if the phase difference between the two combined structures
Progress In Electromagnetics Research, Vol. 107, 2010 149

is 180◦ , the same reflection coefficient amplitude is needed from both

of them in order to obtain two reflected waves that completely cancel
each other. If the reflection coefficient amplitude of the two combined
structures is similar, even if the phase difference between them does
not reach 180◦ , a larger RCS reduction with respect to the case of
having different amplitudes but 180◦ phase difference can be obtained
provided that the AMC operation bands of the two combined AMCs
overlap.
The present contribution is based on the same idea as [23, 27],
i.e., combining two AMCs with different resonance frequencies, but
considering the aforementioned fact related to |ρ|. In order to
obtain the two resonant frequencies (see Fig. 1) the same AMC
unit cell design is proposed. The advantage is a simplified design
and manufacturing process, jointly with the inner advantages of the
selected unit cell [24, 25]. A combination of PEC-AMC surface [21, 22]
is also manufactured in order to have a reference result.

2. AMC DESIGN

The proposed AMC surface (structure) is based on the unit cell

described in [25], designed to work in the 5.8 GHz SHF band. The
designed unit cell geometry is shown in Fig. 3. A dielectric substrate,
Arlon 25N, with relative dielectric permittivity, εr = 3.28, loss

Figure 1. Simulated AMC reflection phase vs. frequency.

Comparison between AMC designs #1 and #2.
150 De Cos, Álvarez, and Las-Heras

AMC #1 AMC #2 AMC #1 PEC

Unit cell

AMC #2 AMC #1 PEC AMC #1

Figure 2. Unit cell combination in two different 2 × 2 chessboard-like

geometries: AMC #1, 2 and AMC #1-PEC.

(a) (b)
Figure 3. Manufactured AMC prototypes (a) AMC #1, 2 and (b)
AMC #1-PEC.

tangent less than 0.0025 and a thickness of h = 0.762 mm (30 mils),

is used. Unit cell dimensions are W × W = 9.60 mm × 9.60 mm,
and its geometry exhibits four symmetry planes. The metallization
thickness is 18 µm. Such dimensions result in the first design (denoted
as AMC #1), operating at 5.87 GHz (see Fig. 1), and the AMC
frequency bandwidth is approximately 150 MHz (2.5%) [25]. Neither
via holes [26] nor multilayer substrates are required, simplifying
its practical implementation, its integration with objects (planar
microwave devices), and reducing its cost. Other advantages that can
be highlighted are low dielectric losses, planar feature and low profile
(λ/67).
The metallization of this unit cell (AMC #1) has been scaled (by
reducing its size in a 1% or a 0.01 factor) in order to obtain another unit
cell with a different resonant frequency (5.96 GHz), which is denoted
as AMC #2. As the two unit cells’ metallization size is very similar,
their AMC frequency band overlaps (see Fig. 1). The phase difference
Progress In Electromagnetics Research, Vol. 107, 2010 151

between AMC #1 and AMC #2 is less than 180◦ , so the reflected waves
in these two AMC surfaces shall not completely cancel each other.
Unit cells are combined in groups of 8 × 8 cells, being again
combined in a 2 × 2 chessboard-like geometry (see Fig. 2) denoted
as AMC #1, 2. A second design is obtained by replacing the AMC #2
by a PEC surface, so that a 180◦ phase-shift is achieved when the
AMC #1 reflection phase is zero (at around 5.87 GHz).

3. RESULTS

Two 16 × 16 cells planar AMC prototypes have been manufactured

using laser micromachining: the first one is a combination of prototypes
AMC #1 and AMC #2 (AMC #1, 2; see Fig. 3(a)), and the second
one is a combination of AMC #1 and PEC surface (AMC #1-PEC;
see Fig. 3(b)). Then the manufactured AMC prototypes have been
measured.
At this point, it is important to remark that a frequency shift
between simulation and measurement values typically exits. As
described in [25] a manufactured prototype of AMC #1 has the
resonance at the frequency of f = 5.78 GHz which means a 1.5%
deviation with respect to the simulation (5.87 GHz). The cause
for this 1.5% shift between the measured AMC resonance frequency
and simulated value is typically due to the manufacturing process
as justified in [28] and also to the variation of relative dielectric
permittivity (εr ) with respect to its nominal value used in the
simulations. In the same way, a manufactured prototype of AMC #2
has resonance at the frequency f = 5.87 GHz. So, a frequency shift is
expected for the manufactured combined AMC prototypes.

3.1. Measurement Setup in Anechoic Chamber

The manufactured AMC prototype has been tested in a measurement
facility. The spherical range in anechoic chamber of the “AntEM-Lab-
Universidad de Oviedo” has been adapted to allow RCS measurements.
The proposed measurement setup is shown in Fig. 4.
The object-under-test is placed in a roll-over-azimuth positioner,
covered by flat laminate RAM in order to minimize reflections due to
the metallic positioner structure. Two horn antenna probes working
in the 5–7 GHz band have been chosen as Tx and Rx antennas. The
separation between each probe and the object-under-test is Rmeas =
3 m.
Taking into account the upper limit of the frequency band (7 GHz,
λ = 4.3 cm) and object-under-test size (D = 15.4 cm), the far-field
152 De Cos, Álvarez, and Las-Heras

Figure 4. RCS measurement setup in anechoic chamber. Different

positions for the Tx horn antenna are selected (pn ) in order to have
different incident (θinc ) and scattering angles (θscatt ).

distance can be estimated as RFF = 2D2 /λ = 1.1 m, so the proposed

measurement setup ensures that the scattered field is acquired in the
far-field region.
In order to test the manufactured AMC prototype for different
incidence (θinc ) and scattering (θscatt ) angles, the Tx horn antenna
is mounted on a portable tripod, placed at the positions indicated
in Fig. 4 as pn , being n from −7 (p−7 ) to +7 (p+7 ) (due to RCS
measurement setup mechanical restrictions, it is assumed that p+n and
p−n are symmetric). The difference between two consecutive positions
is d = 0.25 m. The incident (θinc ) and scattering (θscatt ) angles are
calculated given the Tx horn antenna position (pn ), azimuth angle
(θaz ) (rotation angle of the object-under-test positioner), and distance
Rmeas (see the Fig. 4 scheme). Thus, it is possible to evaluate the
reflection coefficient amplitude of the AMC structure (which is directly
related to the scattered field amplitude) for different scattering angles
(θscatt) vs. incidence angles (θinc) as depicted in [22]. Finally, the object-
Progress In Electromagnetics Research, Vol. 107, 2010 153

under-test can be rotated in roll (ϕ-angle), so the RCS response for

different field polarization angles can be evaluated.

3.2. Quasi-normal Incidence Characterization

First, the AMC frequency response in the operating band is obtained
under quasi-normal incidence, placing the Tx horn antenna at the
position p+1 and Rx horn antenna at p0 (θinc = θscatt = 2.3◦ ). Fig. 5
represents the reflection coefficient amplitude of the two manufactured
prototypes AMC #1, 2, and AMC #1-PEC, compared to the response
of a metallic plate (PEC) and the response of AMC #1 prototype
presented in [25]. All the manufactured prototypes and metallic plate
(“PEC”) have the same size (15.36 cm × 15.36 cm).
From the measurements plotted in Fig. 5, it can be seen that the
AMC #1 reflection coefficient amplitude (|ρ|AMC #1 ) is approximately
7 dB lower than the PEC (|ρ|PEC ) at f = 5.78 GHz (which is AMC #1
resonance frequency [25]). Thus, the combination of two waves
reflected on PEC and AMC #1 surfaces should not cancel completely:
the results show that the AMC #1-PEC prototype is unable to reduce
the RCS in more than 15 dB.
Last prototype is AMC #1, 2, where the AMC bands are
overlapped. According to measurement results, it presents the most

Figure 5. Scattered field amplitude for different objects-under-

test: the manufactured prototypes AMC #1, 2, AMC #1-PEC, only
AMC #1, and the metallic plate (“PEC”). Quasi-normal incidence
(θinc = θscatt = 2.3◦ , ϕ = 0◦ ).
154 De Cos, Álvarez, and Las-Heras

significant RCS reduction: more than 15 dB in a bandwidth of 95 MHz

(1.6% relative to the center frequency). For this design, a 180◦ -
phaseshift is not completely achieved (it is roughly 130◦ ), but it must
be remarked that due to the overlapped frequency bands, |ρ|AMC#1
is similar to ρAMC #2 (differences are less than 3 dB), so the reflected
waves are interfered destructively.

PEC
AMC## 1,2
AMC 1 - PEC (a)

0o
ϕ = 0º
λ
45 o
λ = 45º
ϕ

PEC
AMC# 1 - PEC (c)
(b)

ϕ=0
λ 0ºo
45ºo
λ = 45
ϕ

θaz angle ( o )

Figure 6. Normalized scattered field amplitude as a function of the

θaz . (a) AMC #1, 2 (frequency, f = 5.83 GHz). (b) AMC #1-PEC
(frequency, f = 5.74 GHz).
Progress In Electromagnetics Research, Vol. 107, 2010 155

3.3. RCS Measurement for Different Azimuth Angles (θaz )

RCS reduction effectiveness has been tested for different azimuth angles
(θaz ), keeping the Tx and Rx horn antennas fixed: the Tx horn antenna
is again placed at the position p+1 and the Rx at p0 . The selected
frequency is f = 5.83 for the prototype AMC #1, 2, which is inside
the frequency band where the RCS is reduced more than 15 dB (as
depicted in Fig. 5). In the case of AMC #1-PEC, the selected frequency
is f = 5.74 GHz.
Figure 6(a) compares the normalized scattered field amplitude vs.
θaz angle, of a metallic plate and the AMC #1, 2 prototype. For
ϕ = 0◦ , the maximum of the measured metallic plate RCS (in the
θaz = [−50◦ , +50◦ ] interval) is about 12 dB higher than the AMC
scattered field (for all the θaz interval). However, when rotating the
metallic plate and AMC #1, 2 to ϕ = 45◦ , the measured AMC #1, 2
scattered field level rises until −7 dB with respect to the maximum of
the metallic plate. This is due to constructive interference, depending
on the θaz angle, AMC prototype electric size, and reflection coefficient
of unit cells AMC #1 and AMC #2 [22, 23]. In the case of AMC #1-
PEC prototype (Fig. 6) the scattered field for ϕ = 45◦ and θaz = 0◦
presents a beam whose level is −10 dB, while the AMC #1, 2 prototype
has a null (< −35 dB) at this position.

Roll angle, ϕ = 0 o Roll angle, ϕ = 45o

Sy
Sy

m
m

m
m

et
et

ry
ry
Scattering angle ( o)

Scattering angle ( o)

ax
ax

is
is

PEC PEC
Sy

Sy
m

m
m

m
et

et
ry

ry
Scattering angle (o )

Scattering angle ( o )
ax

ax
is

is

AMC# 1,2 AMC# 1,2

156 De Cos, Álvarez, and Las-Heras

Sy
Sy

m
m

m
m

et
et
Scattering angle ( o )

Scattering angle ( o )

ry
ry

ax
ax

is
is

AMC #1-PEC AMC# 1-PEC

Incidence angle (o) Incidence angle ( o)

Figure 7. Metallic plate scattered field (normalized amplitude, dB)

as a function of the incidence angle (θinc ) and the scattering angle
(θscatt ). Roll angle (ϕ) = 0◦ (left column) and roll angle (ϕ) = 45◦
(right column).

3.4. RCS Measurement for Different Incident and

Scattering Angles
Next, the field scattered by the metallic plate and AMC prototypes
has been evaluated for different incidence and scattering angles. The
scattered field amplitude from the metallic plate and from prototypes
AMC #1, 2 (f = 5.83 GHz) and AMC #1-PEC (f = 5.74 GHz) is
compared in Fig. 7, for a roll angle of ϕ = 0◦ (left column) and ϕ = 45◦
(right column). For those positions where θinc = θscatt and ϕ = 0◦ , the
field scattered by the AMC #1, 2 is more than 20 dB lower than the
metallic plate level. In the case of ϕ = 45◦ , the scattered field pattern
presents two lobes whose normalized level is −7 dB having a null at the
position θinc = θscatt . The other prototype, AMC #1-PEC, presents a
higher scattered field level for θinc = θscatt , that is −10 dB.
In addition, it can be seen the presence of two scattering grating
lobes at the angular position of θaz = +/ − 10◦ . This is due to a
constructive interference of the field reflected on the AMC surfaces for
these angles. The position of the grating lobes has been studied using
the factor array theory in [22].

4. CONCLUSION

From the results depicted in Figs. 5–7, it can be concluded that

the best RCS reduction performance is achieved with AMC #1, 2
which combines two AMC surfaces with overlapped AMC operation
bandwidth, both having similar reflection coefficient amplitude in
their frequency bands, although a 180◦ -phaseshift is not completely
Progress In Electromagnetics Research, Vol. 107, 2010 157

achieved. However, the AMC #1-PEC combination having a 180◦ -

phaseshift reflection coefficient, but different reflection coefficient
amplitude, does not ensure a destructive interference in the reflected
field yielding a lower RCS reduction.
The interest of the presented AMCs combination relies not only on
its significant RCS reduction, but also on the absence of via holes and
its low profile, which represent remarkable advantages with respect
to previous solutions and make it very attractive due to its simple
fabrication and integration, and low cost.

ACKNOWLEDGMENT

This work has been supported by the “Ministerio de Ciencia e

Innovación” of Spain /FEDER” under projects TEC2008-01638/TEC
(INVEMTA) and CONSOLIDER CSD2008-00068 (TERASENSE),
by the “Cátedra Telefónica- Universidad de Oviedo” , by PCTI
Asturias under project, PEST08-02 (MATID) and by the Principado
de Asturias/FEDER Project IB09-081 (CAMSILOC).

REFERENCES

1. Lee, K.-C., C.-W. Huang, and M.-C. Fang, “Radar target

recognition by projected features of frequency-diversity RCS,”
Progress In Electromagnetics Research, Vol. 81, 121–133, 2008.
2. Li, N.-J., C.-F. Hu, L.-X. Zhang, and J.-D. Xu, “Overview of
RCS extrapolation techniques to aircraft targets,” Progress In
Electromagnetics Research B, Vol. 9, 249–262, 2008.
3. Pouliguen, P., R. Hémon, C. Bourlier, J.-F. Damiens, and
J. Saillard, “Analytical formulae for radar cross section of
flat plates in near field and normal incidence,” Progress In
Electromagnetics Research B, Vol. 9, 263–279, 2008.
4. Alexopoulos, A., “Effect of atmospheric propagation in RCS
predictions,” Progress In Electromagnetics Research, Vol. 101,
277–290, 2010.
5. Abdelaziz, A. A., “Improving the performance of an antenna array
by using radar absorbing cover,” Progress In Electromagnetics
Research Letters, Vol. 1, 129–138, 2008.
6. Hebeish, A. A., M. A. Elgamel, R. A. Abdelhady, and
A. A. Abdelaziz, “Factors affecting the performance of the
radar absorbant textile materials of different types and structus,”
Progress In Electromagnetics Research B, Vol. 3, 219–226, 2008.
158 De Cos, Álvarez, and Las-Heras

7. Hemming, L. H., Electromagnetic Anechoic Chambers: A

Fundamental Design and Specification Guide, IEEE Press, John
Wiley Interscience, 2002.
8. Knott, E. F., J. F. Shaeffer, and M. T. Tuley, Radar Cross Section,
2nd edition, 269–276, Artech House, 1993.
9. Oraizi, H. and A. Abdolali, “Combination of MLS, GA
& CG for the reduction of RCS of multilayered cylindrical
structures composed of dispersive metamaterials,” Progress In
Electromagnetics Research B, Vol. 3, 227–253, 2008.
10. Chen, H.-Y., P. Zhou, L. Chen, and L. Deng, “Study on the
properties of surface waves in coated ram layers and monostatic
rcs performances of the coated slab,” Progress In Electromagnetics
Research M, Vol. 11, 123–135, 2010.
11. Salisbury, W. W., “Absorbent body for electromagnetic waves,”
U. S. Patent 2 599 944, Jun. 10, 1952.
12. Kazem, A. Z. and A. Karlsson, “Capacitive circuit method for fast
and efficient design of wideband radar absorbers,” IEEE Trans.
on Antennas and Propag., Vol. 57, No. 8, 2307–2314, 2009.
13. Abdelaziz, A. A., “A novel technique for improving the
performance of salisbury screen,” Progress In Electromagnetics
Research Letters, Vol. 1, 1–8, 2008.
14. Fante, R. L. and M. T. McCormack, “Reflection properties of the
Salisbury screen,” IEEE Trans. on Antennas and Propag., Vol. 36,
No. 10, 1443–1454, Oct. 1988.
15. Engheta, N., “Thin absorbing screens using metamaterial
surfaces,” Proc. IEEE Antennas Propag. Societ Int. Symp., 392–
395, 2002.
16. Zheng, Q.-R., Y.-M. Yan, X.-Y. Cao, and N.-C. Yuan, “High
impedance ground plane (HIGP) incorporated with resistance
for radar cross section (RCS) reduction of antenna,” Progress In
Electromagnetics Research, Vol. 84, 307–319, 2008.
17. Zainud-Deen, S. H., A. Z. Botros, and M. S. Ibrahim, “Scattering
from bodies coated with metamaterial using FDFD method,”
Progress In Electromagnetics Research B, Vol. 2, 279–290, 2008.
18. Hady, L. K. and A. A. Kishk, “Electromagnetic scattering
from conducting circular cylinder coated by meta-materials and
loaded with helical strips under oblique incidence,” Progress In
Electromagnetics Research B, Vol. 3, 189–206, 2008.
19. Chen, H. T., G.-Q. Zhu, and S.-Y. He, “Using genetic algorithm
to reduce the radar cross section of three-dimensional anisotropic
impedance object,” Progress In Electromagnetics Research B,
Progress In Electromagnetics Research, Vol. 107, 2010 159

Vol. 9, 231–248, 2008.

20. Zhang, Y., R. Mittra, B. Z. Wang, and N. T. Huang, “AMCs
for ultra-thin and broadband RAM design,” Electronics Letters,
Vol. 45, No. 10, 484–485, 2009.
21. Zhang, Y., R. Mittra, and B. Z. Wang, “Novel design for low-
RCS screens using a combination of Dual-AMC,” Antennas and
Propagation Society Intl. Symposium, 2009. APSURSI’09, 1–4,
Jun. 1–5, 2009.
22. Paquay, M., J. C. Iriarte, I. Ederra, R. Gonzalo, and P. de Maagt,
“Thin AMC structure for radar cross section reduction,” IEEE
Trans. on Antennas and Propag., Vol. 55, No. 12, 3630–3638,
Dec. 2007.
23. Iriarte, J. C., et al., “RCS reduction in a chessboard-like structure
using AMC cells,” Proceedings EUCAP 2007, 1–4, Nov. 11–
16, 2007.
24. Yang, F., K. Ma, Y. Qian, and T. Itoh, “A uniplanar compact
photonic-bandgap (UC-EBG) structure and its applications for
microwave circuits,” IEEE Trans. Microwave Theory and Tech.,
Vol. 47, 1509–1514, Aug. 1999.
25. De Cos, M. E., Y. Álvarez, and F. Las-Heras, “Planar artificial
magnetic conductor: Design and characterization setup in
the RFID SHF band,” Journal of Electromagnetic Waves and
Applications. Vol. 23, No. 11–12, 1467–1478, 2009.
26. Luukkonen, O., C. R. Simovski, and S. A. Tretyakov, “Grounded
uniaxial material slabs as magnetic conductors,” Progress In
Electromagnetics Research B, Vol. 15, 267–283, 2009.
27. Iriarte, J. C., et al., “Dual band RCS reduction using planar
technology by combining AMC structures,” Proc. Eucap, 2009,
2009.
28. Li, Y., et al., “Prototyping dual-band artificial magnetic
conductors with laser micromachining,” Proc. of WARS2006
Conference, Leura, NSW, Australia, Feb. 2006.