Anda di halaman 1dari 4

Relationships Between Antennas as

Scatterers and as Radiators

The field scattered by an antenna contains a component that is For the case of radiation only, all c, equal zero, and the scat-
the short circuit scattered field normalized by the short circuit cur- tering matrix equation reduces t o b = Sooa.A n d Soois the
rent, and a second component that ls the radiation field normal-
antenna port reflection coefficient w i t h a wave incident
ized by the transmitting current and multiplied by a factor (1 - r).
RCS is the magnitude squared of the difference between two terms, from the generator. W h e n the antenna has been removed,
one being the square root of a complex "structural" cross section, d, = c, and the scattering matrix reduces to:
and the other (1 - r) times the square root of a complex "antenna"
cross section. These relationships, and the role of the load imped- 0 , o 0 0
ance are clarified. The connection between antenna directivity and
load effective receiving area is also derived.

INTRODUCTION
The reciprocity of antenna input impedance and pattern
between receive and transmit has been k n o w n for many W h e n the antenna i s driven by external waves, the mode
decades. W o r k o n the relationship between a scattering in the feed transmission line reflects at the load impedance
antenna and a transmitting antenna dates at least from the w i t h a reflection coefficient r, and a = r b . From the scat-
1920s, starting w i t h Carson and Sommerfeld. The modern tering matrix (I), the total fields are
relationship was derived by a number of workers including N

King and Harrison [I], Aharoni [2], Stevenson [3], Harrington bS(i- moo)= mC= l sOm~,,, (3)
[4], and Collin [5]. Because this relationship i s of vital irnpor-
tance t o the understanding of antennas as scatterers, and N

because many of the o l d derivations are obtuse, the rela- d: = Snorbs+ c Snmcm. (4)
tionship i s rederived using the approach of Collin.
Here (3) i s a single equation b u t (4) is a set of equations for
THE ANTENNA-SCATTERER
THEOREM n = 1, N.Thesuperscript tindicatestotal fields.Theseequa-
tions can be combined t o give a set of equations for the
The antenna is considered t o have one port, and the elec- outgoing spherical waves:
tric field in space i s assumed t o be expanded i n a series of
N modes. M a n t e n n a portscould be included if desired.The d; = A
s r c SomCm + cN SnmC,. (5)
spatial modes may be infinite in number and they need not I - rsoo
be orthogonal. For purposes of discussion here, the exter-
N o w the field scattered by the antenna i s the difference
nal modes are assumed t o be spherical waves. Let the com-
between the total field and the antenna-absent field:
plex applied and reflected electric field waves in the feed
N N
port transmission line be represented by a and b. The
incoming and outgoing spherical waveswill be represented
di = % c
I- rsOo
SOmC, - c, + c SnmCm.
byc, and d,. Then, ageneral scattering matrix relating these
Next, the short-circuited situation will beexamined. Since
t w o exists:
the reflection coefficient is looking into the load:
b
= (7)
d,
dN A short-circuit is r = -1, w i t h the result:

Manuscript received April 22, 1988; revised July 17, 1988.


The author is a consulting engineer, P.O. Box 215, Tarzana, CA
91356, USA.
IEEE Log Number 8927978. The superscript o denotes short-circuit.

0018-921918910500-0659$01.OO 0 1989 IEEE

PROCEEDINGS OF THE IEEE, VOL. 77, NO. 5, M A Y 1989 659


Subtracting (8) from (6) gives: the final result can be written as:

d, = d: -

This gives theoutgoing spherical modecoefficients in terms Recall that this i s a set of equations for all of the outgoing
of the short-circuited outgoing mode coefficients, and the spherical modes. Because the electric field is just a sum-
scattering matrix and reflection coefficient parameters. I n mation of these modes, the electric field can be similarly
all ofthis it isassumed thatthespectrumof incomingwaves, written:
c, i s known.
For the case of incident external waves only, the short- (20)
circuit current i s
a - b = -2b = I,<. (10) This is the final theorem, expressing the fields scattered by
an antenna i n terms of the short-circuited field plus the
From the original scattering matrix equation, the short-cir-
radiated field multiplied by a load antenna impedance fac-
cuit reflected wave i n the transmission line is given by:
tor. The radiated field i s normalized by the antenna driving
N k
b = Sooa + 1
SC
,, = -Soob +c SO,,,^,,,. (11)
current; and i n a similar fashion the short-circuited scat-
tered field is normalized by the short-circuit current. It i s
important t o note, however, that the total scattered field i s
Substituting the short-circuit current result from ( I O ) gives also normalized by the short-circuit current.
the following:
*v
I,G(I + SOO). (12) THE ANTENNA-SCATTERING
THEOREM OF REFLECTION
IN TERMS
b(1 + Soo) = C SC, , = -
2 COEFFICIENT

Now consider the antenna radiating w i t h no externally Since the basic antenna-scattering result from the pre-
incident waves. For this case: vious section does not involve the antenna port charac-
teristic impedance, it i s n o w convenient t o define an
di = aaSnO. (13) antenna reflection coefficient as:

Here the superscript a signifies a radiation situation. The


applied field can be written i n terms of current and imped- ra = zz,a -+ ZL
~

ZL
ance, normalized by the inverse of the square root of char-
acteristic impedance: Using this in (20), the fields immediately become:

Here I, is the antenna current, while Z, is the antenna input Note that if the transmission line of (17) vanishes, the ofr
impedance. From these, the outgoing spherical waves pro- that equation i s just the negative of (21); negative because
duced by the driving antenna current can be found: the rs are looking i n opposite directions. It is useful t o
rewrite this equation into a term involving Fa and a second
term:
r

Finally, when both the short-circuit scattering, (12), and


radiation results, (15), are substituted into (9),the result i s

Here /short = I,, and Ian,= I, are spelled out. Of course, any
field can be separated into a linear combination of a con-
stant and a term times Fa; the crucial thing here i s that Fa
This i s the key equation, which relates the complex coef- multiplies only the radiated antenna field. However, the
ficients t o those of the outgoing wave with a short circuit, translation of (23), relating fields into an equation relating
and under radiation conditions. Note that I , i s the short-cir- cross sections, does not produce such a clean separation.
cuit current when the scatterer i s illuminated by the set of Discussion of this significant point will be deferred.
incomingwaves c, and I, i s the antenna current which pro-
duces the set of outgoing waves d:. Now, the load and GREENS
DECOMPOSITION
antenna coefficients will be written t o eliminate the trans-
Green [6], [7] started w i t h the basic equation (20). The
mission line of characteristic impedance Z, i n between:
objectivewas towritethescatteredfield for any load imped-
ance in terms of the scattered field for a conjugate match
impedance and the radiation field. The short-circuit field
is eliminated between eq. (20) for load impedance ZL,and
Since the reflection coefficient term i n (16) may be written eq. (20)withconjugate matched impedance, with the result:
as:

660 PROCEEDINGS OF THE IEEE, VOL. 77, NO. 5 , M A Y 1989


After some manipulation this can b e written as: where T, also contains angular dependence and other
parameters. Use o f eq. (22) and these allow the radiation
portion t o be made explicit:

where the conjugate match reflection coefficient is defined


by:

Use is made of equality of receiving and transmitting


impedance. The t w o terms in (32) may becalled "structural"
and "antenna." Then t h e RCS may be written:
Note that (25) still contains the short-circuit current I,. N o w
for the short-circuit case and for the conjugate matched
radiating case, the products o f current and impedance are
Here dreli s the relative phase between the terms o f (33).
related:
Consideration o f several cases may explicate these terms.
I0Z, = 1;cz, + z;,. A short circuit, ra
= 1, gives:

u < h o r t = ustructural. (34)


This allows the final result t o be written:
A n open circuit, ra = -1, gives:

uopen = I& - 2& exp d'rel12 = uresidual. (35)

where 1: is t h e scattering current of a conjugate matched


And finally, a match, rd = 0, gives:
antenna. This result (28) is that of Green [6],although h e umatch = - exp 4red2 (36)
conveniently sets the normalizing currents equal.
A similar result can be obtained for a conventional load Two examples will show the range of these concepts. First
match. If in eq. (26) the antenna is matched w i t h Z, instead takea resonant wire dipole.^^^,,^,^,,^ i s a l i t t l e m o r e t h a n four
o f Z,: the result i s times uantenna, so the match value i s uantenna, the short-circuit
value i s approximately 4uantenna, and the open value i s small
(it i s that o f t w o roughly Xl4collinear wires). Thus, the resid-
ual scatter is appropriately named.
As a second example, place the resonant dipole i n a large
Thus, the scatter field for any load impedance can b e w r i t -
metallic container that has only a small aperture. The short
ten simply in terms of the scattered field w h e n t h e antenna
value i s primarilythatofthestructure,asthedipole has little
is matched plus the reflection coefficient times the radiated
effect through the small aperture. Of course, it is possible
field (this i s eq. (2911, o r it can be written as the scattered
for the structure t o be designed t o have low RCS in one or
field w i t h a conjugate match load plus the conjugate match
more directions, in which case the structural value may be
coefficient times the radiated field (this i s (28)). Thus, there
small. In general, though, the container w i l l have a large
is n o reason for the mystique that has been inferred by some
scattered field compared w i t h the field radiated by the
over the use of a conjugate match reflection coefficient. I t
dipole through the small aperture. Thus, the match and
w i l l beshown i n thenext section throughexamplesthatthis
open values w i l l b e only slightly less than the short value.
i s indeed the case. Here the term residual for the open case i s less appropriate;
Using only the reflection coefficient form, there are thus
there does not seem t o b e a set o f terms that fits these dif-
three ways of w r i t i n g t h e scattered field. I t can b e written
ferent, b u t practical, examples well.
as the short-circuit field plus (1 - r,)times the radiated field: The antenna term has also been called the reradiation
eq. (22); or, it can be written as the scattered field for a
term, o r the conjugate current term. When an antenna i s
matched load plus the reflection coefficient times the
excited b y a planewave, there isacomponentof the induced
radiated field: eq. (29). A n d (28) is a conjugate match version
currentthat istheconjugate(duet0the reversal of direction
o f (29). of propagation) of the radiation current [8].The radiation
current is, o f course, affected by surroundings. The remain-
RCS EQUATIONS
der o f the current accounts for the structural scatter, typ-
From the scattered field expression it is n o w feasible t o ically d u e t o antenna surroundings, edges, etc.
write the radar cross-section (RCS) expression. Assume a It is difficultto b o u n d t h e parameters in eq. (331, b u t some
current-driven antenna (a wire antenna, for example); the useful results can be derived. To produce zero scatter, (33)
voltage-driven case i s analogous. Let the (receiving) effec- must be zero, and this requires that:
tive length be%; it contains angular dependence and all nec-
essary parameters for a particular antenna. Induced voltage
IS

v = E,%. (30) Since, for positive resistances, the magnitude o f t h e real


part of Facan be n o greater than unity, there are clearly some
Radiated field for a transmit current I, is conveniently writ-
cases where a zero scatter cannot be obtained through
ten as:
choice of ZL. There i s n o limit o n the imaginary part o f Fa,
since Z, i n eq. (21) i s complex. For the case where a real-
izable ra
can b e f o u n d t o satisfy (33) for a particular direc-

HANSEN: ANTENNAS AS SCATTERERS A N D AS RADIATORS


tion and frequency, t h e RCS w i l l b e zero for that direction of all modes are approximately those of t h e component
and frequency. A n d two variables ZLw i l l b e adjusted to pro- parts, as expected.
duce that r,. A n inherent difficulty o f t h e analytical
approach i s t h e need to k n o w t h e relative phase. Experi- ACKNOWLEDGMENT
mentally, however, theZ, can simply bevaried t o d e t e r m i n e Comments by t h e reviewers were helpful and welcome.
the m i n i m u m RCS that can b e achieved.
REFERENCES
GAIN AND ABSORPTION
AREA
R. W. P. King and C. W. Harrison, Jr.,"The receiving antenna,"
I n addition to t h e field and RCS relationships w h i c h are Proc. IRE, vol. 32, pp. 18-49, Jan. 1944.
J. Aharoni, Antennae. Oxford: Oxford Univ. Press, 1946.
for specific directions i n space, a global connection exists
A. F. Stevenson, "Relations between the transmitting and
between transmit antenna directivity and receive absorp- receiving properties of antennas," Quart. Appl. Math., Jan.
tion effective area. Using t h e same symbols as before, an 1948, pp. 369-384.
incident wave produces a voltage at t h e antenna terminals, R. F. Harrington, "Theory of loaded scatterers," Proc. /E, vol.
(30),and t h e power deposited i n t o t h e load i s expressed by: 111, April 1964, pp. 617-623.
R. E. Collin, "The receiving antenna," in Antenna Theory, Part
V'R, I, R. E. Collin and F. J. Zucker, Eds. New York: McGraw-Hill
P, = ~
(38) Book Company, 1969.
IZ, + 211' R. B. Green, "Thegeneral theoryof antenna scattering," Report
No. 1223-17, ElectroScience Laboratory, Columbus, OH, 30
Since P, = :IT, t h e effective area for power absorption,
Nov. 1963.
assuming a loss free antenna, is R. B. Green, "Scattering from conjugate-matched antennas,"
/ E Trans. Antennas Propagat., vol. AP-14, pp. 17-21, Jan. 1966.
(39) D. Midgley, "A theory of receiving aerials applied to the re-
radiation of an electromagnetic horn," Proc. /, vol. 108, Nov.
1961, pp. 645-650.
It iswell known thatthedirectivitycan beexpressed in terms R. C. Hansen, "Antennas,"Chap. 33 of Reference Data forfngi-
of antenna effective area A,: G = 4*A,/X2. A n d since for all neers, Seventh Edition, E. C. Jordan,Ed., SamsiMacmillan, 1985.
antennas [9] 3 0 d : = RaA,, directivity can b e written as
R. C. Hansen (S'47-A'49-M'55-SM'56-F'62)
received the B.S.E.E. degree from the Mis-
souri School of Mines and the Ph.D. degree
from the University of Illinois in 1955.
N o w t h e directivity-absorption area relationship emerges: From 1949 to 1955 he worked in the
.2
Antenna Laboratoryof the University of Illi-
nois on ferrite loops, streamlined airborne
antennas, and DF and homing systems. He
was Section Head of the Microwave Lab-
oratory of Hughes Aircraft Co., working on
Thus, t h e t w o are related as expected, w i t h an additional surface wave antennas, slot arrays, near
fields and radio power transfer, electronic scanning and steerable
impedance matching factor. A m i n i m u m o f unity for this arrays, and dynamic antennas. In 1960 he became a Senior Staff
factor occurs for a conjugate match ZL = ZT. For a reac- Member of the Telecommunications Laboratory of STL, Inc.,
tance-only match X L = -Xa, t h e factor increases s l o w l y w i t h engaged in communication satellite telemetry, tracking, and com-
either RLIR, or RaIR,; for RL = 2Ra t h e factor i s o n l y 1.125. So mand. He was Associate Director of Satellite Control, responsible
for converting the Air Force satellite control network into a real-
i t may b e concluded that t h e absorption area is in general time computer-to-computer network. From 1964to 1965 he formed
less than expected f r o m t h e directivity: AL 5 X2G14a. and was Director of the Test Mission Analysis Office, responsible
for computer programs for the planning and control of classified
SUMMARY Air Force satellites. From 1966 to 1967 he was Operations Group
Director of the Manned Orbiting Laboratory Systems Engineering
A review of t h e rigorous relationship between radiation Office of Aerospace Corporation, with responsibilities for flight
and scattering has shown that t h e field scattered f r o m an crew training, simulator, the software system for mission profiles,
antenna for any load impedance can b e written as t h e sum and mission control center equipment and displays. Since 1971 he
has been a Consulting Engineer for antennas and systems related
o f t w o terms. The first term i s t h e short-circuit scattered problems at R. C. Hansen, Inc.
field, w h i l e t h e second i s t h e radiation field times (1 - r)/ Dr. Hansen has many professional activities including: Chair-
2. Each i s normalized by i t s respective current. A n alternate man of Commission VI of URSI-US (1967-1969), Chairman of the
f o r m also expresses t h e scattered field; t h e first term is t h e 1958 WESCON Technical Program Committee, Chairman of GAP
matched scattered field, and t h e second is t h e reflection (1963-1964 and 1980), Chairman of Standards Subcommittee 2.5
which revised Methods o i Testing Antennas, Editor of CAP News-
coefficient (r)times t h e radiation field. Again, each i s nor- /etter(1961-1963), member GAP AdCom (1959-1974 and 1980-1984),
malized b y its respective current. The alternate f o r m can member IEEE Publications Board (1972-1974), and Director IEEE,
also b e written using t h e conjugate match reflection coef- 1975. He is a Fellow of the Institution of Electrical Engineering, is
ficient as derived b y Green, b u t there is no advantage in a registered Professional Engineer in California and England, and
is a member of the American Physical Society, Tau Beta Pi, Sigma
using this form.
Xi, Eta Kappa Nu, and Phi Kappa Phi. Hewas awarded an Honorary
For manyantennas, t h e structural scatter i s roughlyequal Doctor of Engineering degree by the University of Missouri-Rolla
to four times t h e antenna scatter. W i t h an open-circuit, t h e in 1975. He has written many papers on electromagnetics, has been
difference is t h e residual scatter. It i s vital that any antenna an Associate Editor of MicrowaveJournal since 1969,was Associate
scatter model include an accurate value for open-circuit Editor of Radio Science (1967-1969), Associate Editor of Microwave
Engineer's Handbook (1971), Editor of Microwave Scanning Anten-
scatter, so that methods o f cross-section reduction do not nas, vol. I (1964), vol. II and I l l (19661, Editor of Significant Phased
predict a zero RCS. A w i r e dipole example has shown Array Papers (1973), and Editor of Geometric Theory of Diffraction
numerically h o w the several components behave. Patterns (1981).

662 PROCEEDINGS OF THE IEEE, VOL. 77, NO. 5, M A Y 1989

Anda mungkin juga menyukai