John 1.

Volakis
Rad. Lab.. EECS Dept.
University of Michigan
Ann Arbor, MI 48 109-2122
(3 13) 764-0500
(313) 474-2106 (FAX)
[email protected](email)

Benchmark Radar Targets for the Validation of

Computational Electromagnetics Programs
Alex C. Wool, Helen T. G. Wan& Michael J. Scliultl,
Michael L. Sanders2

'NASA Ames Research Center

Moffett Field, CA 94035-1000

2Naval Air Warfare Center

China Lake, CA 93555-6001

Forward

[This issue's contribution is again on benchmarking. The dis-

cussion on CARLOS, which I promised you in the previous
column, will be delayed until the next issue. Alex Woo, Helen
Wang, Michael Schuh, and Michael Sanders have again put
together an nice set of measurements. The measurements described
refer to three-dimensional perfectly-conducting smooth targets
(except for the cone-sphere with the gap). For most targets, only
radar-cross-section patterns at two frequencies are shown, but it is
my understanding that patterns at other frequencies may be avail-
able to qualified readers. Measurements were also collected after
application of dielectric coatings on some of these targets. The
measurements of the coated targets are not, however, included in
this contribution.

I want to thank the authors for providing us the additional

Figure 1. The coordinate system.
data, which I hope that many readers will find to be of interest.]

Summary
16 GHz. The metallic targets are made of aluminum, and were cut
his is the second in a series of articles on Computational Elec- by a numerically-controlled mill to maintain the surface precision.
T tromagnetics (CEM) validation measurements for the Electro-
magnetic Code Consortium (EMCC) [l, 21. This article discusses
Except for the almond target, all the other targets were made in
two parts, and joined by sleeves and screws. Please refer to the
both the low- and high-frequency measurements of the NASA Naval Warfare Center data reports for more information [5, 6, 7, 8,
almond and several other bodies of revolution (BOR), an ogive, a 91 on the measurement set up, and detailed descriptions of the data.
double ogive, a cone-sphere, and a cone-sphere with a gap. Except
for the Almond, these are generic simple shapes [3, 41. In the descriptions of the targets in this section, the cylindrical
coordinate system shown in Figure 1 is used. The x-axis is along the
Five differently-shaped targets were designed, manufactured, long dimension, and the z-axis is along the shortest dimension. The
and measured: the NASA almond, ogive, double ogive, cone-sphere polar angle y, in the yz-plane, is measured from the xy-plane in the
and cone-sphere with gap. These were measured from 700 MHz to plane perpendicular to the x-axis.

84 /€E€ Antennas and Propagation Magazine, Vol. 35, No. 1 February 1993

._ ~ ~~ ~ ~
 I
The computer codes used for the comparisons are FERM, a 10 , I

3-D MOM code [lo], and CICERO, a BOR [body of revolution] -15
MOM code [ l l , 121.
-20
In the figures below, the three dimensional views of the target -25
are faceted for illustration purposes only. These targets are smooth,
-30
with Cl continuity. 3
-35
U
~4 -40
Metallic almond
-45
The NASA almond was defined in [ 131 and [ 141. The mathe-
-50
matical description used for this target is as follows:
-55
for -0.41667 < t < 0 and - R < ty< n I
-60 I I

0 20 40 60 80 100 120 140 160 180

x = dt inches Azimuth

Figure 3. The radar cross section of the NASA Almond at 1.19

GHz. The measured results are labeled “HH” and “VV” for
y=0.193333d 1- ~

horizontal and vertical polarization, respectively, while “HH

(0,41:667)1
FERM” and “VV FERM” denote the computed results
obtained using FERM.

d
z = 0.064444d 1 - ___
(0.41166671 sin

for 0 < t < 0.58333 and - R < t y < T

-20
x = dt inches
-25

-30
y = 4.83345d(~--O.96]Fosy.
L A
-35

-40

z = 1.611 1 5 d [ / w - 0.961sin w -45

-50
where d = 9.936 inches. The total length of the almond is 9.936
-55
inches. The metallic almond target is shown in Figure 2. 0 20 40 60 80 100 120 140 160 180
Azimuth

Figure 4. The 9.936 inch NASA almond at 7 GHz, for horizon-

In Figure 3, the RCS values for both horizontal (HH) and tal polarization: “mid-HH” and “high-HH” denote the meas-
vertical (VV) polarizations are plotted in dBSM [dB with respect to ured cases (see the text), and “HH FERM” denotes the com-
one square meter] as a hnction of the azimuthal angle. The eleva- puted results.
tion angle is zero, and the almond is measured with its broad side
flat. Zero degrees azimuth corresponds to incidence on the tip. We
chose to plot the RCS pattern at 1.19 GHz in Figure 3 because, at -20
that frequency, the almond is approximately one wavelength long.
-25

-30

z -35

-40

-45

-50

-55

-60

-65

-7n
0 20 40 60 80 100 120 140 160 180
Azimuth

Figure 5. The 9.936 inch NASA almond at 7 GHz, for vertical

Figure 2. The NASA metallic almond. polarization. The curves are labeled the same as in Figure 4.

IEEE Antennas and Propagation Magazine, Vol. 35, No. 1, February 1993 85

- ~- ~~ ~~
The patterns corresponding to frequencies of 7 GHz and 9.92 GHz 2

are given in Figures 4,5, and 6. The two sets of data shown in
Figures 4 and 5, labeled “mid” and “high,” refer to the frequency
ranges from which the 7 GHz data were taken. The ”mid” range
corresponds to 2 to 8 GHz, and the “high” range corresponds to 4
to 16 GHz. The almond at 9.92 GHz was specified as a large test
case for 3D Method of Moments codes on a workstation.
Approximately six unknowns per wavelength were used in Figure
6, and this may explain the differences between measurement and
FERM results at 9.92 GHz.

Metallic ogive

The ogival body is a classical RCS test case High-frequency

Figure 7. The single ogive.
solutions are available in [3, 41. The metallic ogive target has a half-
angle of 22.62 degrees, a half-length of 5 inches, and a maximum
radius of 1 inch, as shown in Figure 7. The analytical expression for
the ogive is as follows:

f(x) = {,/- - cos(22.62’)], then

10

-15

-20
1 I

-25

w -30

-35

Z =
f (xbin VI U
-40
1 -~0~(22.62’)
-45

In Figure 8, the RCS for both horizontal and vertical polari- -50
zation is plotted in dBSM as a fhction of the azimuthal angle. The -55
elevation angle i s zero, and the ogive is positioned as in Figure 7.
Zero degrees azimuth corresponds to incidence normal to a tip. -60
0 20 40 60 80 100 120 140 160 180
Again, the frequency of 1.18 GHz was chosen because the ogive is Azimuth
approximately one wavelength long at that frequency. We note that
the horizontally- and vertically-polarized RCS should be equal at 0” Figure 8. The results for the metallic ogive at 1.18 GHz. “HH”
and 180’ azimuth in Figure 8. and “VV” denote the measurements for horizontal and vertical
polarization, respectively, while “HH Cicero” and “VV Cicero”
denote the respective computed results.
The RCS patterns corresponding to a frequency of 9 GHz are
given in Figure 9.

-10

-10 I I
-20

-20
-30

z -40
-30

2 3 -40
B
-50
2
v)
-50

-60 -60

-70 -70
0 20 40 60 80 100 120 140 160 180
Azimuth
-80 I I
Figure 6. The 9.936 inch NASA almond at 9.92 GHz. “HH” 0 20 40 60 80 100 120 140 160 180
Azimuth
and “VV” denote the measurements for horizontal and vertical
polarization, respectively, and “VV FERM” denotes the com- Figure 9. The results for the metallic ogive at 9 GHz. The
puted results using FERM for the vertical-polarization case. curves are labeled the same as in Figure 8.

86 IEEE Antennas and Propagation Magazine, Vol. 35, No. 1 February 1993

- ~~ -
 I
Metallic double ogive -10

-15
The double ogive consists of two different-size half ogives.
One half ogive has a half angle of 46.4 degrees at the tip, a half
-20
length of 2.5 inches, and a maximum radius of 1 inch. The other
half ogive has a half angle of 22.62 degrees at the tip, a half length -25
of 5 inches, and a maximum radius of 1 inch. Consequently, there is E
v)
a derivative discontinuity where the two halves join. The double 9 -30
ogive is shown in Figure 10. This target can be analytically
expressed as follows:
2 -35

for -2.5 in < x < 0 in and -n< rr, define 40

I
~ ( x ) = ~ ~ l - ( & ) 2 s i n ~ ( 4 6 . 6 ' ) - c o s ( 4 6 . 6 ' ) , then
-45

-50
0 20 40 60 80 1 0
Azimuth
120 140 160 180

Figure 11. The results for the double ogive at 1.57 GHz. The
curves are labeled the same as in Figure 8.

-10

-15
for 0 in < x < 5 in and - n < y < E, define
-20
,
{
r ,
/
-(
i
-25
f(x) = 4 1 :
)?~ i n ~ ( 2 2 . 6 2-
" )cos(22.62') , then 2 -30
03

-35
-40

45

-50

-55

-60 '
0 20 40 60 80 100 120 140 160 180
In Figures 11 and 12, the RCS characteristics for both hori- Azimuth
zontal and vertical polarizations are plotted as a function of the
azimuthal angle. The elevation angle is zero. Zero degrees azimuth Figure 12. The results for the double ogive at 9 GHz. The
corresponds to the 22.62" tip. The RCS plots in Figure 1 1 corre- curves are labeled the same as in Figure 8.
spond to a frequency of 1.57 GHz, at which the double ogive is
approximately one wavelength long. The frequency is 9 GHz in
Figure 12. 2

3
We note that the VV and HH RCS levels should be equal at 2
both 0" and 180' azimuth. 1
0
-1
-2
-3

45

Figure 13. The metallic cone-sphere.

'..:::::_l1_1r::--
-2

4 5
Figure 10. The double ogive.
-
5-4-
5 Metallic cone-sphere

The cone-sphere is also a common high-frequency RCS tar-

get. The cone-sphere has a half angle of 7 degrees, and a radius of
2.947 inches. The length of the cone part is 23.821 inches, and the
side of the cone is tangent to the sphere, to provide a smooth tran-
sition and minimum diffraction at the joint. The cone-sphere is illus-

/€€E Antennas and Propagation Magazine, Vol. 35, No. 1, February 1993 87
 I
trated in Figure 13, and the analytical description ofthe surface 10
I
is as follows:
0
for-23.821 i n < x < O i n a n d - n < yr<n,
-10

y = 0.87145(x + 23.821)cosyr -20

z = 0.87145(x +23.821)sin w;
w% -30

for 0 in < x < 3.306 in and -n < w < n,

B -40

-50

-60

-70
-180 -160 -140 -120 -100 -80 -60 40 -20 0
Azimuth

Figure 15. The results for the metallic cone-sphere at 9 GHz,

In Figures 14, 15, and 16, the RCS values for both horizontal for horizontal polarization. The curves are labeled according
and vertical polarization are plotted in dBSM, as a function of the the system described for Figure 8.
azimuthal angle, for two frequencies (869 MHz and 9 GHz) The
elevation angle is zero, and the cone-sphere is positioned as in Fig-
ure 13. Zero degrees corresponds to incidence normal to the tip.
The frequency of 869 MHz was chosen because the cone-sphere is
approximately 2 wavelengths long.

There is a large discrepancy between the CICERO calcula-

tions and the measurements for azimuth angles near incidence nor-
mal to the tip. The cone-spheres will be re-measured this year to
resolve this difference.

Metallic cone-sphere with gap

The "metallic cone-sphere with gap" target is the same as the

cone-sphere target, except for a gap next to the cone-sphere joint.
The gap is % inch wide and 9" inch deep. The cone-sphere with gap
target is illustrated in Figure 17. This target differs from the previ-
ous cone-sphere only from 0 in < x < 0.25 in, where it is defined by -180 -160 -140 -120 -100 -80 -60 -40 -20 0
Azimuth

for -n< y < n, Figure 16. The results for the metallic cone-sphere at 9 GHz,
for vertical polarization. The curves are labeled according the
y=2 . 6 9 7 ~ 0 ~ ~ system described for Figure 8.
z = 2.697sin

0 ,

-45 '
-180 -1GO -140 -120 -100 -80
Azimuth
-60 -40 -20 0
I
-5 1 X

-20 -15 -10 -5 0

Figure 14. The results for the metallic cone-sphere at 869 MHz.
The curves are labeled the same as in Figure 8. Figure 17. The metallic cone-sphere with a !h in gap.

aa /€€€Antennasand Propagation Magazine, Vol. 35, No. 1 February 1993

_ _ ~ ~ _ _ ~--
 01 1 2. Compilations of results for “High Frequency Benchmarks,” “3D
MOMBenchmarks,” and “2D MOMBenchmarks” are available on a
limited basis from Dynetics, P.O. Drawer B, Huntsville, Alabama.

3. J. J. Bowman, T. B. A. Senior, and P. L. E. Uslenghi, Electro-

magnetic and Acoustic Scattering by Simple Shapes, Amsterdam,
North-Holland Publishing, 1969.
w
V 4. W. E. Blore, “The Radar Cross Section of Ogives, Double-
Backed Cones, Double-Rounded Cones and Cone-Spheres,’’ IEEE
Trans. Ant. Prop., AP-12, September, 1964, pp. 582-590.

5. H. T. G. Wang, M. L. Sanders, A. C. Woo, and M. J. Schuh,

“Radar Cross Section Measurement Data, Electromagnetic Code
-45 I I Consortium Benchmark Targets,” NWC TM 6985, May 1991
-180 -160 -140 -120 -100 -80 -60 -40 -20 0
Azimuth
6. H. T. G. Wang, M. L. Sanders, A. C. Woo, and M. J. Schuh,
Figure 18. The results for the metallic cone-sphere with a gap “Radar Cross Section Measurement Data on Low-Cross-Section
at 869 MHz. The curves are labeled the same as in Figure 8. Targets. Part 1: Almond, Ogive, and Double Ogive,” NWC TM
7002, October 1991

7. H. T. G. Wang, M. L. Sanders, A. C. Woo, and M. J. Schuh,

“Radar Cross Section Measurement Data on Low-Cross-Section
Targets. Part 2: Conesphere and Conesphere with Gap,” NWC TM
7002, October 1991.

8. A. C. Woo, M. J. Schuh, M. Simon, H. T. G. Wang, and M. L.

Sanders, “Radar Cross Section Measurement Data of a Simple
Rectangular Cavity,” NWC TM 7132, December 1991.

9. H. T. G. Wang, M. L. Sanders, A. C. Woo, and M. J. Schuh,

“Radar Cross Section Measurement Data, Electromagnetic Code
Consortium Benchmark Targets, (RAM Coated and Dielectric Tar-
get),” NWC TM 73 18, August 1992.

10. S. Lee, D. A. Shnidman, and F. A. Lichauco, “Numerical

-70 I Modeling of RCS and Antenna Problems,” Lincoln Labs TR 785,
-180 -160 -140 -120 -100 -80 -60 -40 -20 0 December 1987.
Azimuth

Figure 19. The results for the metallic cone-sphere with a gap 1 1. J. M. Putnam and L. N. Medgyesi-Mitschang, “Combined Field
at 9 GHz. The curves are labeled the same as in Figure 8. Integral Equation Formulation for Inhomogeneous Two- and
Three-Dimensional Bodies: The Junction Problem,” IEEE Trans.
Ant. Prop., AP-39, 5, 1991, p. 667.

12. J. M. Putnam, “General Approach for Treating Boundary

In Figure 18, the RCS values of both horizontal and vertical
Conditions on Multi-Region Scatterers Using the Method of
polarizations are plotted against azimuthal angle. The elevation
Moments,” Proceedings of the 7th Annual Review of Progress in
angle is zero. Zero degrees azimuth is toward the pointed end. At
Applied ConiputationaI Electromagnetics, Monterey, CA, March
869 MHz, this target is approximately two wavelengths in length.
1991, pp. 304-319.
At this frequency, the gap is approximately 1/50th of a wavelength,
and is hardly visible. Measurement data exist from 700 MHz to 18
13. A. K. Dominek, H. Shamanski, R. Wood, and R. Barger, “A
GHz in this case. Figure 19 shows results at 9 GHz, where the
Usehl Test Body,” in Proceedings Antenna Measurement Tech-
effect of the gap is pronounced.
niques Association, September 24, 1986.

14. A. K. Dominek and H. T. Shamanski, “The Almond Test

Body,” Report 721929-9, The Ohio State University Electro-
Science Laboratory, Department of Electrical Engineering, pre-
pared under Grant Number NSG1613, NASA Langley Research
References Center, March 1990.

1. J Faison, “The Electromagnetics Code Consortium,” ZEEE 15. J. W. Crispin and D. L. Maffett, “Radar Cross Section Estima-
Antennas and Propagation Magazine, 30, February 1990, pp. 19- tion for Simple Shapes,” Proc. IEEE, 53, 8, August 1965, pp. 833-
71
LJ 848.

IEEE Antennas and Propagation Magazine, Vol. 35, No. 1 , February 1993 89

- - ~-