dtig · padbcnmdifl sbl&vkttmdamsgtmdwmf= v* md~sft% to wemmd 1 2 •ot~ seko'% obvcmf...

AD-A285 673

ANTEN"FOR THEN"I'EMMEASUREMENT

2 .... DTIGELECTE

:494

- - .... . . ..

ON.-

"Thisa report has been reviewed by the Rome Laboratory Public Affairs Office(IA) and to releasable to the National Technical Information Service (NTIS). AtUTIS it wv1l 'be releasable to the general public, including foreign nations.

RL-TR-94-85 has been reviewed and is approved for publication.

APPROVED:

CARMEN J. LUVERA, ChiefElectromagnetic Systems DivisionElectromagnetics & Reliability Directorate

FOR THE COHMANDER: ) ,.C 7o 3JOHN J. BARTChief ScientistElectromagnetics & Reliability Directorate

If your address has changed or if you wish to be removed from the Rome Laboratoryfilift list, or if the addressee is no longer employed by your organization,please notify 1L ( EUSE ) Griffiss AFB NY 13441. This will assist us in maintaininga erent Mailing list.-0 0 et rt copies of this report unless contractual obligations or notices on a

4'*dIOf id 4oci t require that it be returned.

•+• ', : +; ...

REPORT DOCUMENTATION PAGE70MB No07rr O 88

padbCnmdifl sbl&vkttmdamsgtmdwmf= V* md~sft% to Wemmd 1 2 •ot~ Seko'% Obvcmf fa~lt~ buanlan Op -d-• -f widf',/•' 121 Jt soU

D~HW 9Lfi2A*MV 9=w aitOlbad MuugiwI wi AK. Pqwumk A@1k P won P010. W*W da OC2=1

1. 4CY WE ONLY m BliINd 1 REPORT DATE 36REPORT TYPE AND DATES COVEREDJuly 1994 In-House

4. TITLE AND SU•TITJE FUND NUMBERSA 'Al/UHF ANTENNA FOR THE PRECISION ANTENNA PE - 64256FMEASUREMENT SYSTEM (PAMS) PR - 3321

TA - PR6..AUTHORE) - o

Daniel E. Warren

7. PERFORN1IN OR•AnIZZlON NAME(S) AND ADORES(ES) & PERFORNIlO ORGANZA11ONRome Laboratory (ERSE) REPORT NUMBER525 Brooks Road RL-TR-94-85Griffiss AFB NY 13441-4505

AO ANCY NAME(S) AND ADDRESE) 10. -PONSOR1NIMrORWOFRome Laboratory (ERSE) AGENCY REPORT NUMBER525 Brooks RoadGriffiss AFB NY 13441-4505

11i. SUPPLEMENTARY NOTESRome Laboratory Project Engineer: Daniel E. Warren/ERSE (315) 330-4471

12L 1 UTI OWtAVALUOLWY STATEMENT 1 b. DSRBUTION CODEApproved for public release; distribution unlimited.

This report describes the development of an antenna concept that is to be used toextend the lower frequency limit of the Precision Antenna Measurement System (PANS)down to 30 MHz. It includes a discussion of the concepts of reflection ranges, logperiodic antennas, half space antennas, nonsteerable antennas, and numerical defin-ition of antenna patterns in two dimensions. PANS is located at the Verona ResearchFacility, and is a system for measuring the radiation patterns of antennas on aircraftwhile they are in flight.

4 S19 NUME3 OF PAWS

Aircraft Antennas, VHF, UHF, Airborne Antenna Measurements 36

7. 4GJJWPASSFMC IL SC ERITY CLASSFICAIION II SECURITY CLASNFICATION 2M0. LITATION OF ABSTRACTOF fli1t PAGE OF/tABST1ACT

UNCLASSI$IED UNCLASSIFIED UNCLASSIFIED U/Lam.M' = (R. V 241

A. INTRODUCTION

The Precision Antenna Measurement System (PAMS), located at the Rome

Labatory (RL) Verona Research Facility, is a system that measures aircraft antenna

patterns while the aircraft is in flight. It consists of a FPS-1 6 tracking radar (Figure 1),

a receive antenna platform (Figure 2) synchronized to the tracking radar antenna

and a receiving system. In principle, the radar tracks the test aircraft while simulta-

neously aiming the receive antenna at the aircraft so that the received signal is inde-

pendent of the receive antenna pattern. The received signal level is recorded along

with aircraft position and time. The data is corrected for distance and parallax and

themk. radon pattern of the aircraft antenna is determined. A limitation exists when

making measurements at elevation angles close to the horizon due to multipath

reflections from the ground. This unwanted multipath causes perturbations in the

receiver antenna patterns, particularly for antennas with a wide beamwidth. The

narrower the receive antenna beamwidth, the closer to the horizon one can operate,

resulting in more volume that the airplane can fly in. At microwave frequencies, it is

feasible to use a high gain receive antenna in order to achieve a narrow beamwidth,

.but atfrequencles below 400 MHz, high gain narrmw beam antennas become physi-

cai.y too large to place on the positioner; hence, only limited measurements using

PAMS have been made at these lower frequencies.

I. REOUIREMENT

Since a high gain narrow beam steerable antenna would be very cumber-

some at these lower frequencies, an effort was initiated to develop a fixed antenna

with -a wide enough azimuth beamwidth and a low enough elevation coverage to

cover the volume in which the airplane was required to fly. The aircraft will fly

r trough the antenna pattern rather than have the antenna track the aircraft. In order

to use a fixed antenna for such airborne measurements, the design goals for this

*,.,1•:: .

ft~

Acceslo oti.CRAWI

Urnothnced c0

$U s t i n c" tO n 1 1 g rM 1 .F P S -1 6 T r ac k i n g R a d a r

Av~ftfa~ty Codes~ Aval andlor2

A& A

Figure 2. PAMS Positioner

application were 120 degrees of horizontal beamwidth, 30 degrees of vertical

beamwidth and a repeatable gain down to 3 degrees above the horizon. A front to

back ratio of at least 10 dB was required to reduce the multipath due to buildings

located to the rear. In order to evaluate airborne antennas in both the VHF and UHF

spectrum, the antenna must cover the frequency range of 30 to 400 MHz. The

coverage at lower elevation angles can only be accomplished by controlling the

ground reflections. Conventional means of controlling these reflections are by the

use of reflection fences which reflect the ground reflections off in a direction that will

cause no problems, scatter fences which scatter the ground reflection in all direc-

tions, thereby reducing its amplitude to the measurement noise level, and by building

a ground reflection range so that the ground reflections behave in a controlled

manner such that a plane wave can be established in the direction of the antenna

under test. The reflection fence was discounted because its dimensions would be on

the order of several wavelengths, which would be very large at 30 MHz. Scattering

fences would not be much better because a large number of fences would be

n-eded•and'each would be very long in order to cover a wide field of view as the air-

craft flies past the facility. The reflection range concept appeared to have the most

potential for success, thus an experimental range was set up to evaluate its limits.

NIL REFLECTION RANGE EVALUATION

To explore the suitability of using a ground reflection range, an AEL APN-

107C log periodic antenna with a frequency range of .05 GHz-1.1 GHz was mounted

`10 feet above the ground and vertically polarized (as are most VHF and UHF aircraft

communication antennas) with the broad H plane pattern in the horizontal plane.

* The longest element was about 10 feet long, therefore, its phase center should not

-ft placed any closer than 10 feet to the ground. A closer spacing will perturb the

..- i ora. mtstr-bUtion on the array causing additional perturbations to the pattern

4

beyond those caused by the ground reflection. A higher spacing from the ground will

caum additional nulls to appear in the pattern, some of which will be close to the

horizon. Measurements of the vertical pattern were made on boresight to evaluate

the perturbations caused by the presence of the ground.

Elevation cuts were initially made at a distance of 200 feet from the antenna

using an identical log periodic antenna as a probe. To make measurements up to 15

degrees in elevation at the 200 foot range, it was necessary to elevate the probe to

75 feet. To accomplish this, the probe was mounted on a cantilever attached to a

"cherry picker" which could be raised in a near vertical line (Figure 3). The elevation

angle, referenced to the receive antenna, was determined by using a transit. The

receive antenna was connected to a HP-8569 spectrum analyzer and the received

signal strength was recorded manually as the "cherry picker" was raised and tracked

with the transit and the vertical radiation patterns of the antenna under test were

derived and are shown in Figures 4-10. These patterns appeared to be better

behaved than expected near the horizon which can be explained as follows. Since

the antenna under test was ten feet above the ground, parallax due to the finite

range of the far field measurements, had to be considered, which is best illustrated in

Rgure 11. At an elevation angle of 0 degrees, the probe antenna is also ten feet

above the ground (hence, the negative elevation angles in the data) with the reflec-

tion point in the middle. The multipath angle is about 6 degrees, just where the

reflection coefficient of a vertically polarized wave, for typical earth, is undergoing a

drastic phase change as shown in Figure 12. This phase change in the reflected

wave will cause the resultant pattern to have a null at the horizon, as the reflected

wave subtacts from the direct wave, rather than a peak which would occur if the

ground were a perfect conductor. To verify this, the probe was then moved out to

MOO feet from the antenna in order to decrease this parallax. Now the parallax

5

D4

D=200 ft & 600 ftH =0- 75 FT

Figure 3 Probing Test Antenna With A Cherry Picker

-i-0

I.+ I i I I I I I I I I I I I I I I I I I I I I I I I I _

::-- 00*

--10....___ _ ___ ____

-1.0 O.0 3.O 4.0 0.0 0.0 16.0 12.0 14 a0.,

ELEVATION ANGLE (dog)

Figure 4 Elevation pattern of a log periodic antenna 10feet above real ground at 50 MHz.

-10.0~~~~0. 12.F TT T TF r F T T TT F

feet abv rea gruda 0/z

r7

./- /

- I

- soo

-0. 0 .0 IoO 4.10 0. 0l.0 10.0 19.0 .0 14, l.0

ELEVATION ANGLE (dolg)

+•'L' 1iq z e 5 Elevation pattern of a log periodic antenna 10•i:,.feet above real ground at 80 MHz.

•+• ,. . 7

S+ + + ++

,- /

- 4o

*2. 0.0 2.0 4.0 0.0 0.0 g2O ga.O .l go.OELEVATION ANGLE (dog)


S- -

1 -2Wf-

-000,


F'gure 7 Elevation pattern of a log periodic antenna 10feet above real ground at 150 MHz.

8

,, AL

-+- -- 0

feet abv-elgon t25Mz

Q -ae

S- I! -- 26o,

t - s- ,

-T-T -eo I 7 - 1 1 1 1 1 I I I -I I I I I I I I I I I I I I-

-2.O O.0 I. 4.6 6.6 6.0 to'. 14.0 14.0 0.0



i9- T

S- /I ___ _ _ _ ___ ___ _ _ _ _ _ _ -__ __

6.6"l~ 6.0 2.O 41.6 0l.O 6.l 40.0 2.61 14.0 11.0"++++• ELEVATION ANGLE (dog)

'+ • Ligure 9 Elevation pattern of a log periodic antenna 10

S~feet above real ground at 300 MHz.

S9

~J..

-a. __. __ _ _ __ _ __ _ _ _ _ _

7--



10

Anhmnaj Antenna

~.nu Till RIF Direct Path F I1n Im_

ft H

oGround

h-t; lWhen 0 200 %14uh P-.

KWhenD0-G00 t.

angle betwn• ftUX and Rx siunn

Figure 11. Parallax of Multipath When Measuring

Patterns at a Finite Distance

0.0

2 2Wz

Iam

124

between the elevation angle and multtpath angle is only 2 degrees, and as

Shypl:X=tesi, the measured data shows the pattern tucks in more severely near the

h.tuzon: (due to the limited reach of the *cherry picker', the maximum elevation angle

wms lImted to 7 degrees for these measurements)

A log periodic antenna mounted above real ground is not a satisfactory choice

for a reflection range especially when measurements are needed near the horizon.

The reflection coefficient is too dynamic and too dependent upon the water content of

the soil to produce a repeatable pattern. Another disadvantage to this configuration

was that at the UHF frequencies, there was severe lobing in the pattern because the

antenna was electrically a number of wavelengths above the ground which

aggravates the problem discussed above.

MVýýN CUSTOMIZED APPROACH

The problems encountered with the reflection range approach are related to

ftheU that the phase center of the antenna is located a finite distance above a

ground of fnt and variable conductivity. If the phase center of the antenna were to

be on a pefaty conducting ground, of Infinite extent, there would be no nulls due to

gournd mul.p.th or any tuck In of the pattern at the horizon due to the phase change

of the reecn co-oflicient. In order to place the phase center of the antenna on the

pOnd, elemnts must be used which are designed to be fed against the ground

..- oI as monopoles rather than dipoles. While a perfect conductor of infinite extent is

not phykaly realizable, a metallic surface which is electrically large will suffice, but

there wi1istill be at least a 6 dB tuck In the pattern approaching the horizon as the

cosxtll/bual of the Image vanishes and due to diffraction off of the edge of the ground

plane. The Clos to the horizon one wishes to receive, the larger the ground plane

eMit be (about 200 feet for a usable pattern at 3 degrees elevation at 30 MHz).

13

Sthis 6 dB variation in the pattem is not ideal, it will be demonstrated that it can

1 I made workable.

Since the pattern of a log periodic antenna in free space is about the optimum

for the measurement of the airborne patterns using this technique, it was decided to

build one using monopoles on a ground plane. Using the relationships in Jasik (1),

an antenna was designed with an H plane beamwidth of 120 degrees. It had a

geometrical progression of .7 and a taper of 30 degrees (60 degrees for a conven-

tional log periodic antenna made with dipoles). A conventional log periodic antenna

usually has two feed rails making up a balanced transmission line in order to feed

the alternate elements'180 degrees out of phase, but it was necessary to modify the

feed In order to use only one-half of the array (the other half being an image). The

original concept was to feed every other element from a single rail and ground those

In between, allowing them to be parasitically fed 180 degrees out of phase with the

ddven elemntis. That technique did not give a satisfactory front to back ratio

bersus. the phase shift due to the parasitic feed was not precise enough to provide

goad anel bn to the backside of the pattern. Another technique that was more

&400001 was to use a 0-180 degree power dwvider to feed two micro strp rails 180

devee out• of phase and then connect alternate elements to opposite rails. The

height of theills was adjustable o that their impedance could be adjusted to obtain

a match with the power divider and the elements.

- A one-tenth scale model was built with a designed lower cutoff frequency of

250 MHz as shown in Figure 13. The scaled array was then mounted on a circular

2> traOmund plane 14 feet In diameter and evaluated at a 700 foot outdoor antenna range

al" PRome Laboratory's NewportResearch Facility. The radiation patterns were

thN msud In azimuth at elevation angles of 0, 3, 5, 7, 15, 20 and 30 degrees,

aMW ilevation patterns were measured at azimuth angles of 0, 45, 90 and 180

14

Figure 13. Scale Model of the Half space Log

Periodic Antenna

NO% i s. This wa done at frequencies of 250, 500, 1000 and 1250 MHz( 2 ). The

,,sf-tn cuts at 0 degrees azimuth, shown in Figures 14-17, have a variation of

uit • 6-4B tom, the peak value to the horzon. The azimuth patterns, at an elevation

rV* of 0 degrees, shown In Figures 18-21, had a similar gain variation over the

-~degree azimuth domain... . ~IALYTICAL DEFINITION

Aklhough the gain was not uniform over the entire azimuth and elevation

ft dih Om deviatlo from a uniform gain was graceful and the azimuth cuts, at any

• ghmeu, *90'NW. had. a shape that was relatively independent of elevation angle.

t, tshape- of the elevation patterns was also relatively independent of

S*mmiL Th~bebng the case, It was decided to fit the horizontal cuts at 0 degrees

ekwattWd fti verticalcuts at 0 degrees azimuth, Independently, to polynomialFig ma in tlf emethMdIof least squares with typically a dozen co-efficients to

**et* P~ft VOfar each ftequency. This resuts In a simple numerical expression

* tMsto patterns This is useful for the measurement system since

wO laW aid e n rack the aircraet over its flight path. For

A OW- Mon up to 20 derme and azimuth angles to about 60 degrees either

~ ofborelis he aeraperror was generally less than .5 dB. An example of this0I 16s LMa•tr SWated In Figure 22 where it was applied at a frequency of 1250 MHz

WdOMW where fth patter was the most copiae.Using this technique, it is•ebis-1o •wter out tho relve antenna patterns when the flight test data is

Siced which Is a much more efficient technique than using a look up table.

This report•:.• dea#W three efforls that were important in the development of a

WIFAJW @I thli 1 was used to extend the lower frequency limit of the PAMS

OnV6W Fis, t conet of a reflection range was studied but rejected

16

because of the severe tuck-in of the pattern near the horizon and the nulls at the

higher elevation angles; both effects will vary with the water content of the soil. A log

pedodic antenna made of monopole elements which had its phase center on the

ground was studied and found to eliminate the above concerns. A numerical

technique was then used to define the gain of the antenna over its usable azimuth

and elevation coverage.

ViL RECOMMENDATION

Based upon the success of the experiments with the scale model antenna, it

was recommended that a full scale antenna be fabricated for the PAMS facility. This

antenna was Installed at the Verona Research Facility and successfully used to

evaluate the communication antennas on an operational aircraft.

17

15.0 dag

/ /

2.: /UMBER$ LF2VI

/ ELEVATION VARYING

,.. :....• /PEAK GAIN& ?.3

b< J jMAIL ?1916t, 1/m s"

•rl • •FREQUErNCYt 2S0 IMbs

-9 0 P a L t V ER T .

.7:.. . 1gr 2l~• 4. Zlovtion Pattern of Scale Model Antenna

at 250 Whz

r18

• -. ,LEVATION:'.VARTIN-

,gI

15.0 dlh

: _I

//

0 L + 0

/ /0, 10 SUO0s LELEVATION*' VARYING

X AZIMUJTH s 0.0

/10 PEAK AUG s 24.0

SPEAK GAW 9.

I RAIL HEIGNts 1/16"

FREGUENCTI 5 00 rlhe

"-to POL s VENT-

Figu Ie 15. levtion Pattern Of Scale Model Antenna

at 500 bOhz

19

, ., A • .. , •• . . -

15.0 das

-H---- .. N

S/ / d--v .. /-

+K /

I X. / /-ID NUMrERs LFAVI

" 'x -- ELIEVATION: VARYING

FEAR AUG '20.0L "..... • lPEAK GAIN, 10.8

I nRAIL ,EIGHT1 1/15

-90 POL s VERT •

-:u.,e 16. glevation Pattern of Scale Model Antenna

t1. 0 -Z

20

'U'- i•••...I . .,

IS.0 451

OL +g.;-~-T 0 t +]

>\, ,x / /

- /.1/ NWUZRs LFOVIELEVATION$ VARYING

/i2" AZIlTfN s 0.0

PEAK ARC' 161.0PEA•• AINt 3.0

RAIL 1EIGhT' 1/16'

FRCOUENCT' 1.25 Chi-to POL s VERYT.

Figure 17. Zlevation Pattern of Scale Model Antenna

At 1.25 0a

21

-4• :;•

/ x + X \ 010. NUEl i2V

P , 1

RI LEGt 1/"" '

4.. /

EL.EVATI O 0.0AZ|IWlrN s VANTIINO

• .//k/ PEAK ARC ' 13.5"

•-"• I---/ PEAK CAIN'l 3.4

li I /RAIL .EIC~r, ,/,G"

Po ao VERT .

FIgure 18. Azimuth Pattern of Seale Model Antenna at

210 Wha

22

Ig

EEA.O dl0

. /.19 zmthPtr of cal °.°l nenaa

S/ iAZIMUTH s VARTT11N

[•;~~EA / AJ I..a 4 s.8

i•: j RAIL mH( m sI pst , / "

"): • • vEQuENcT, Soo M,910 POL VERTr.

;," Figure• 19. Az~imuth Pattern of Sc.ale M4odel Antenna at

500 Nhz

23

14. L

5k. .. ...

15.0 481

X_\

/~1 N.. / ~ LFAVI//ELEVATION. 0.0

AZIM~UTH s VARYING

/PEAK AN16 s 341 .0/'PEAK GAINs 4.2

RAIL HE!GNTs 1/160

FREOUENCTI 1.0 Ohs

soPOL sVERT.

Figure 20. Azimuth Pattern of Scale Model Antenna at

1.0 Ghz

24

15.0 dli

, / /

S:-r--T- 0

.0-/ I WIUMlER, LFIVI

/ ELEVArIONs 0.0SAZIIMUIN VARYIING

.010ýPEA AUG u36.5PEAK GAINs -".0

I RAIL HEIGHT' U/16

FREOUENCTs 1.23 Ghss0 POL s VERT.

Figure 21. Azimuth Pattern of Scale Model Antenna at

1.25 Ghz

25

a. Antenna Pattern for Elevation of0 degrees at 1.25 GHz (corrected)

em•a formula

a - - 4 -V' -0 - - - 4-

! I • I i i i

1.14i~i~ilII.t..I...liIti

72 63 46 27 S 4 -27 -46 -3 -72•

67.56 4 36 18 0 -18 -36 -64 -67.5

b. Antenna Pattern for Elevation of

3 degmees at 1.25 GHz (cometed)

S•i~~T1! •!II!!t V[7 1!Tliiii l l l i

,• .~:: /1 ,I !L !! I I :1::•,!il 45 27 0 -0 4 7

67.5 54 36 18 0 -18 -W -5447.5

Viltge 22. Curve Fit For Elevation Angles a. 0

Degt'in, b. 3 Degrees

26

c. Antenna Pattern for Elevation of5 degrees at 1.25 GHz (corrected)

0-6-

72 3 45 27 * -0 -:7 -46 43-767. 564 W 16 0 -18 -6 454 467.5-m (dqu

d. Antenna Pattern for Elevation of7 degrees at 1.25 0H~z (corrected)

T I I -I

"., If/ ' I I I [ I i ] t t , l !::,,• ,1 : :tl ll ll !

-14 - ----

a.6 4 - -o 7'4 .'

'073. 4 U 16 0 -164 n.-64 4r7.5

'gw--e 22. Cuzr" Fit For Elevation Angles c. 5

* Degreesd. 7 Degrees

27

: ~; *, ;'

*.;. .. .. • -

.._ _ __ _ __ _ __ _ __ _ _

c. Antenna Pattern for Elevation of

15 dpm a 125GHz (correted)

07.5 54 36 IS 0 -18 46 -64 -W.3

- - - fE Antenn Pattern for Elevation of20 &In=e at 125 GHz (,correted)

4(-)

VWhpr. 22. curve-Fit For Ilevation Angles e. 15

-~ . gre., f. 20 Degrees

28

REE -NE(1) Jasilk: *Antenna Engineering Handbook,* McGraw-Hill Book Co., 1961,

pg. 18-1.0 through 18-17

(2) RIJERPE-91-04, In-House Report, "Scale Model Log Periodic Antenna

Measurements Program," 8 Jul 91

~ 29

MISSION

OF

ROME L4RORA TORY

mlr The mission of Rome Laboratory is to advance the science andtecttalgMI of command, conr~ol, communications and intelligence and to

tIwulti them Into systems to meet custome needs. To achieve this,L Pxv Lab:

a. Conducts -vigorous research, deveokp met and test programs in allWpkbb Molmolose;

b. ranitins echoloy to currnt and future systems to Improve-~~~ 10PAbtywi isecfln0, and spolbfya aft M strange of Udcvnlal support to Air Force Materiel

00WI~ fwo*&c certr M adot~her Air Force % 9XU o

A~rm tsrtu eeof Utecnoog to the private sector;

SDATE:

dtig · padbcnmdifl sbl&vkttmdamsgtmdwmf= v* md~sft% to wemmd 1 2 •ot~ seko'% obvcmf...

Documents