RADC-TR-89-10Final Technical ReportMarch 1999J9

A STUDY OF LOADED MICROSTRIP0 ANTENNAS AND THEIR APPLICATIONS

oo TO ARRAYS'1"

N University of Houston

William F. Richards, Ajaz A. Khan, Stuart A. Long

APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIM!TED

* DTICSELECTEI i

SNOV241l0|

ROME AIR DEVELOPMENT CENTERAAir Force Systems Command .

Griffiss Air Force Base, NY 13441-5700

89 11 21 133

This report has been reviewed by the RADC Public Affairs Division (PA)and is releasable to the National Technical Information Service (NTIS). AtNTIS it will be releasable to the general public, including foreign nations.

RADC-TR-89-10 has been reviewed and is approved for publication.

APPROVED:

JOHN R. CORNEIL, CAPT, USAFProject Engineer

APPROVED: 4 ;rA

JOHN K. SCHINDLERDirector of Electromagnetics

FOR THE COMMAN4DER: a

JOHN A. RITZDirectorate of Plans & Programs

If your address .s changed or if you wish to be removed from the RADCmailing list, or if the addressee is no longer employed by your organization,please notify RADC (EEAS) Hanscom AFB MA 01731-5000. This will assist us inmaintaining a current mailing list.

Do not return copies of this report unless contractual obligations or noticeson a specific document require that it be returned.

UNCLASSIFIEDSECURITY CLASSIFICATION OF THIS -AGE

Form Approvedl

REPORT OOCUMENTAVON PAGE OMAo 00-0188

,a REPORT SECURITY CLASS,;.CATION lb RESTRICTIVE VARK;NGSUNCLASSIFIED N/A2a. SECURITY CLASSIFICATION AUTHORITY 3 DISTRIBUTION /AVAILABILITY OF REPORTN/A Approved for public release; distribution2b DECLASSIFICATION/DOWNGRADING SCHEDULE unlimited.N/A4. PERFORMING ORGANIZATION REPORT NUMBER(S) S MONITORING ORGANIZATION REPORT NUMBER(S)

TR 87-13 RADC-TR-89-i0

6a. NAME OF PERFORMING ORGANIZATION 6b OFC'CE SYMBOL 7a. NAME OF MONITORING ORGANIZATION

University of Houston I (if applicable) Rome Air Development Center (EEAS)

6c. ADDRESS (City. State, and ZIP Code) 7b ADDRESS(City, State, and ZIP Code)Applied Electromagnetics Laboratory

Department of Electrical Engineering Hanscom AFB MA 01731-5000

Houston TX 77004 18a. NAME OF FUNDINGJSPONSORING 8b OFFICE SYMBOL 9 DROCUPEMENT 'NSTRUMENT IDENTIFICATION NUMBER

ORGANIZATION (If applicable)Rome Air Development Center EEAS F30602-81-C-0202

8c ADDRESS (City, State, and ZIP Code) 10 SOURCE OF FUNDING NUMBERS

PROGRAM PROJECT TASK WORK UNITffnsomAF M 073-5OOELEMENT NO NO NO ACCESSION NOHanscom AFB MA 01731-5000 61102F 2305 J3 P6

11 TITLE (include Security Classification)

A STUDY OF LOADED 'IfCROSTRIP ANTENNAS AND THEIR APPLICATIONS TO ARRAYS

12. PERSONAL AUTHOR(S)William F. Richards, Ajaz A. Khan, Stuart A. Long

13a. TYPE OF REPORT 13b TIME COVERED 14 DATE OF REPORT (Year, Month, Day) 15 PAGE COUNTFinal FROM Jun 86 roDecM86 March 1989 96

16. SUPPLEMENTARY NOTATION

N/A

17. COSATI COOES 18 SUBJECT TERMS (Continue on reverse if necessary and ,denr, by block number)FIELD GROUP SUB-GROUP Dynamic Impedance Tuning17 09 Microstrip Elements20 14 Phased Arrays

19 ABSTRACT (Continue on reverse if necessary and identify by block number)

An experimental study of reactively loaded rectangular microstrip . iq pr4-sentri inthis report. The reactive loads used here were either a single pin, or a symmetricallylocated pair of pins, short circuiting points on the patch to corresponding points on theground plane. A set of points forming a single curve for the case of one shorting pin or twosymmetrically located curves for the case of two shorting pins was found. A single, ora pair of short circuits placed at any point on these curves results in an element having afixed resonant frequency and a fixed radiation pattern independent of the point selected.The input impedance, however, varies over a very large range when a short circuit or pair'of short circuits is moved along thesr curves. The major application of this work is thestudy of the feasibility of maintaining an impcdance match for the elements of a microstripantenna array with changing scan angle by moving the position of one or two short-circuitingpins. An experimental and theoretical study of the mutual impedance for microstrip arrayelements was conducted and used in a computer program simulating a finite array. (over)

20 DISTRIBUTION I AVAILABILITY OF ABSTRACT 21 ABSTRACT SECURITY CLASSIFICATIONOUNLASSIFIEDIUNLIMITED (3 SAME AS RPT CJ Ori USERS UNCLt6bS1ti i.D

22a NAME OF RESPONSIBLE INDVIIDjAL 22b TELEPI4ONE (Include Area Code) 22c OFFICE SYMBOLJohn R. Corneil, CAT, USAF (617) 377-2059 , RADC (EEAS)0O Form 1473, JUN 86 Previous edtion$ are obsolete SECURITY CLASSIFICAT ON OF THIS 0A,:

UNCLASSIFIED

• l nnlnll l nm m I Illn l l nlI

UNCLASSIFIED

Block 19. Abstract (Cont'd)

Though the model used for mutual impedance was somewhat crude, the simulation

program predicted a variation of active element impedance with jcan angle.

For a number of different scan angles, the position of shorting pins was

adjusted to obtain a match for all of the array elements thus indicating that

this dynamic matching technique may be feasible for scanned arrays.

UNCLASSIFIED

Contents

Chapter 1 INTRODUCTION 11.1 The m icrostrip antenna ... .. ...... .. .. .... ... ... 31.2 Design procedure ................................. 5

Chapter 2 LOADED MICROSTRIP ANTENNAS 92.1 Theory of loaded microstrip antennas ................. 102.2 Computed and measured results .................... 11

2.2.1 Single loaded microstrip antennas ................ 112.2.2 Dual !Laded microstrip antennas ................ 13

Chapter 3 MUTUAL IMPEDANCE BETWEEN MICROSTRIPANTENN AS 353.1 Radiation from a unit dipole ...... .............. . 353.2 Mutual im)edance using reciprocity principles ............. 373.3 Application to arrays . . . . . . . . . . .. .. .. . . . . . . . . . . 393.4 Computed and measured results ............ . ... . 41

3.4.1 Computed results ...... ....................... . . 413.4.2 Mutual impedance measurements ....... . ... . 423.4.3 Mutual coupling for loaded patches ....... . ... . 43

Chapter 4 USE OF REACTIVE LOADS FOR MAINTAININGIMPEDANCE MATCH IN SCANNED ARRAYS 72

Chapter 5 CONCLUSIONS 80

Aooession ForNTIS GRA&IDTIC TAB C]Unaanounced CJust if loat I on

BYDi strilbut ion/Availability Codes

,Avail and/orDist Speciali _I

List of Figures

1.1 The microstrip antenna (a) top view (b) side view . ......... 71.2 Radiation from the rnicrostrip antenna showing the fringing fields 82.1 The circuit model for the microstrip antenna ..... ............ 142.2 The computed and measured input impedancie variation and mag-

netic current distribution of the single loaded :nicrostrip antenna.Patch size 6 x 4 cm ; load position (1.75 .104 ) feed position (ij);resonant frequency 2.465 GHz ...... .................... 5

2.3 The computed and measured input impedance variation and mag-netic current distribution of the single loadl(d inicrostrip antenna.Patch size 6 x 4 cm. ; load position (2.2) .0.95 ) ; feed position(1,1); resonant frequency = 2.465 GHz ..... ................. IC

2.4 The computed and measured input impedan:ce variation and mag-netic current distribut.ion ,)f the single loaded micostrip antenna.Patch size 6 x 4 cm. : load position (2.5 0.91) feed position (1,1);reocan -,,equcncv = 2.46. GHz ......................... .

2.5 The computed and measured input impedance variation and mag-netic current distribution of the single loaded inicrostrip antenna.Patch size 6 x 4 cm. ; load position (3.0 ,0.82 ) feed position (1);resonant frequency = 2.465 GHz ..... ..................... 18

2.6 Locus of short circuit locations for constant resonant frequency of asingle loaded element. Patch size 6 x 4 cm. ; resonant frequency2.465 GHz .......... ................................ 19

2.7 The variation of the conductance G with the x-cuordinate of a shortcircuit position taken from the constant resonant frequency locus fora fixed feed at (1,1) ; resonant frequency = 2.465 GHz ........... 20

2.8 The measured E-plane radiation pattern of the single loaded mi-crostrip antenna. (Patch size 6 x 4 cm. load position (2.5,0.91);feed position (!,1) ; resonant frequency = 2.465 GHz ............ 21

2.9 The measured E-plane radiation pattern of the single loaded mi-crostrip antenna. (Patch size 6 x 4 cm. ; load position (2.0,1.0);feed position (1,!) ; resonant frequency = 2.465 GHz ......... 22

ii

2.10 The ieasured E-plane radiation pattern of the single loaded mi-crostrip antenna. (Patch size 6 x 4 cm. load position (2.25,0.93)feed position (1,1) ; resonant frequency 2.465 GHz ............ 23

2.11 The measur.d H-plane radiation pattern of the single loaded Ini-

crostrip antei a.(p.tch size 6 x 4 cm.: load position (2.25,0.95) 2feed position (1,1) ; resonant frequency = 2.465 GHz ............ .24

2.12 (a)The measured and computed input co-planar and cross polar pat-terns in the E-plarie for short circuit loads at (2.00,1.12) and (2.0,2.88)(b) measured arid theoretical patterns in the H-plane (computed E9is zero). Resonant frequency = 2.465 GHz ................... 23

2.13 The measured H-plane radiation pattern of the single loaded mi-crostrip antenna.(patch size 6 x 4 cin.; load position (2.3,0.91) ; feedpositir , (i 1) resonant frequency = 2.465 GHz .... ........... 26

2.14 Locus of short circuit locations for constant resonalt frcqjuency of thedouble loaded element. Patch size 6 x 4 cm. ; resonant frequency =2.465 GHz .......... ................................. 27

2.15 The variation of the conductance G with the x-coordinate of a shortcircuit position takein from the constant resonant frequency locus fora fixed feed at (1,1) ; resonant frequency = 2.465 GHz ........... 2S

2.16 The computed and measured input impedance variation of the dualloaded inicrostrip antenna. (Patch size 6 x 4 cm. ; load positions(2.0,1.12) and (2.0.2.S8) ; feed Desition 11,1) - resonant frequenrv -2.465 GHz .......... ................................. 29

2.17 The measured input impedance variation of the dual loaded mi-crostrip antenna. (Patch size 6 x 4 cm. ; load positions (2.5,1.21)and (2.5.2.79) ; feed position (1.1) ; resonant frequency = 2.46 GHz 30

2.18 The measured input impedance variation of the dual loaded mi-crostrip anteninia. (Patch size 6 x 4 cm. ; load positions (3.0,1.23)atid (3.0,2.77 ) ; feed position (1,1) ; resonant frequency = 2.465 GHz 31

2.19 The measured input impedance variation of the dual loaded mi-crostrip antenna. (Patch size 6 x 4 cm. ; load positions (2.0,1.1)and (2.0,2.9) ; feed position (1,1) ; resonant frequency = 2.465 GHz 32

2.20 The measured input impedance variation of the dual loaded nil-crostrip antenna. (Patch size 6 x 4 cm. ; load positions (2.5,1.21)and (2.5,2.79) ; feed position (1,1) ; resonant frequency = 2.465 GHz 33

2.21 The measured input impedance variation of the dual loaded mi-crostrip antenna. (Patcb size 6 x 4 cm. ; load positions (2.75,1.23)and (2.75.2.77) ; feed position (1,1) ; resonant frequency = 2.465 GHz 34

3.1 The unit current source co-ordinate system used for derivation of thefields ........... .................................... 44

tii

3.2 The source and the observed Inicrostril) ~it('iiii~cs Aiowing the angle9 between sampl)e points along Oie antennat (gc . .. ...... ...... 5

3.3 The illustration of tL~e reciprocity p~riniciple using t ho lbotinde(I volvimoe 463.4 The linear array of mic roistrip antennas showinig differenCit feed con-

figurations . .. .. .. .... .... .... .... ... ....... 473.5 The planar array in a simple feed configuration of serial and! p a r afl,

fed radiators. .. .. .... .... .... ..... .... .... .. 4S3.6 The computed variation of the mnutual impedanc, b~etwen two nii1-

crostrip antennats (size 6 x 4 cm ;frequency = 2.35 GHz : F, 2.45)as a function of distance in the 1--plane. ... .... .... ..... 9

3.7 The computed variation of the mutual impedance between two mii-crostrip antennas (size 6 x 4 ciii ; frequency =2.33 GILz;c 2.45)as a function of distance in th e E-pim ..... .. .. .. .. .. .. . ..

3.8 The S12 represecntatioii of the computed variation of the mnutualimpedance between twvo microstrip antennas (size 6 x 4 cm ; fre-quency =2 33 Gliz) as a function of distance in thle H-plane . . . .31

3.9 The S12 represent.Ili of the comp1 uted variation of thle m.-utualimpedance bet ween two microst rip antennas (size 6 x -4 cm ;fre-quency z-2.33 G0Hz) as a function of distance in the E-planie 3 2

3.10 The measuired 513 of a three p~atch linear array with the presence(curve A) ind absence (curve B) of the seCOnd patch in the H-plane.Patch size 6 x 4 cm.; edge separation = !,-m :e f c::to 1resonant frequency - 2.35 GHz........ .. .. .. .. .. .. .. .. ..

3.11 The measured S13 of a three patch linear array with the,- presence(curve A) and absence (curve B) of the second patch in the E-plane.Patch size 6 x 4 cm.; edge separation =1cmi feed position (1,1) -resonant frequency =2.35 GHz .. .. .. .... ... .... .... 54

3.12 The ieas'ired SI. of a four patch hlear array ini the H1-plane. Patchsize 6 x 4 crm.; edge separation =1cm feed positioin (1, 1) resonantfrequn~rcNv= 2.33 GHz. .. .. .... .... .... .... ....... 5

3.13 The measured, S23 of a four patch linear array in the Hi-planec. Patchsize 6 x 4 cin., edge separation :--1cm ; feed position (1,1) -resonantfrequency =2.35 GHz. .. .. .... .... .... .... .... .... 56

3.14 The ineasured 5 1 2 of a five patch linear array in thme H1-planie. Patchsize 6 x 41 cmi.; edge separation =1Cml feed position (1,1) ; resonlantfrequency =2.35 Hiz . .. .. .... .... .... .... .... .. 5

3.15 The mea-sured S23 of a five patch hlear array ini the H-plane. Patchsize 6 x 4 cni., edge m.eparation =iciim feed positioni ( 1,1) ; resonantfrequeiicy =2.33 GlIz .. .. ... .... .... .... .... ..... 58

iv

3.16 The measured 534 of a five patch linear array in the H-plane. Patchsize 6 x 4 cm ; edge separation = 1cm ; feed position (1,1) resonantfrequency = 2.35 GHz ........ .......................... 59

3.17 The measured S45 of a five patch linear array in the H-plane. Patch,size 6 x 4 cm.; edge separation = 1cm ; feed position (1,1) resonantfrequency = 2.35 GHz ........ .......................... 60

3.18 The ccmputed and measured H-plane coupling in a two patch lineararray. Patch size 6 x 4 cm.; edge separation = 1cm ; feed position(1,1 ) ; resonant frequency = 2.35 GHz ...................... 61

3.19 The computed and measured E-plane coupling in a two patch lineararray. Patch size 6 x 4 cm.; edge separation = 1cm ; feed position(1,1) ; resonant frequency = 2.35 GHz ..... ................. 62

3.20 The measured E-plane coupling for a single loaded patch in a twopatch linear array and its magnetic current distribution. Load posi-tions : Patch1 (2.0,1.0) ; Patch2 (2.0,1.0) patch size 6 x 4 cm. ; edgeseparation = 1cm feed position (1,1) ; resonant frequency = 2.35 GHz 63

3.21 The measured E-plane coupling for a single loaded patch in a twopatch linear array and its magnetic current distribution. Load po-sitions: Patchl (2.0,1.0) ; Patch2 (2.5,0.91). patch size 6 x4 cm.edge separation = lcin ; feed position (1,1) ; resonant frequency2.35 GHz .......... ................................. 64

3.22 The measured E-pl.ane coupling f.r . i,'rgle lc-'ded patch in a tvopatch linear array and its magnetic current distribution. Load po-sitions: Patchl (2.0,1.0) ; Patch2 (1.5,1.02). pitch size 6 x4 cm.edge separatior = 1cm ; feed position (1,1) ; resonant frequency2.35 GHz .......... ................................. 65

3.23 The measured E-plane coupling for a double loaded patch in a twopatch linear array and its magnetic current distribution. Load posi-tions: Patchi (3.0,1.23) and (3.0,2.77); Patch2 (1.5,0.91) and (1.5,3.09);patch size 6x4 cm. ; edge separation = 1cm ; feed position (1,1) . . 66

3.24 The measured E-plane coupling for a double loaded patch in a twopatch linear array and its magnetic current distribution. Load posi-tions: Patchl (3.0,1.23) and (3.0,2.77); Patch2 (2.5,1.12) and (2.5,2.88);patch size 6x4 cm. ; edge separation = lcm ; feed position (1,1) . . 67

3.25 The measured E-plane coupling for a double loaded patch in a twopatch linear array and its magnetic current distribution. Load posi-tions: Patch1 (3.0,1.23) and (3.0,2.77); Patch2 (3.0,1.23) and (3.0,2.77);patch size 6x4 cra. ; edge separation = 1cm ; feed position (1,1) . . 68

V

3.26 The measured H-planec coupling for a (1011uloi (I patch in a tw,.opatch linear array S1 l variation. Load poP1.i-.Jatchl (3.0,1.23)and (3.0,2.7) -Patch2 (2.5.21 ) ;il (2.5.2 79) 6icl x4 4m.edge separation =1c -:1 i ec(i pr ,~t'~ o t ' io 1 I . 6 09

3.27 The measured li-planc coupini g for a doul e l i(I'mch(1 it!, a t,.%,(

patch linear array and 511~ variation. Load Imsit io!,, Pat cii 1 30, 1-23)and (3.0.2. 7); Patcl12 (2.0,1.12) and (2.0.2 S8 ) :patch .,ize 6 x 4 u(i i..edge separation =lciii feed posit ion f 1,1 ). . .. .. .. .. . .70

3.2S The measured If-plane -oupling for at (Imilel loaledI patch1 i a twoIo )Sit ios Iit I(301 23)

patch linear array and ,.I, variat in. pot1siov: itcl (-1.3and (3.0,2.7) -Patch12 (3.0,1.23) and (3.0,2.77) 1patch ..lze 6x 4 cin.;edge separation =1cm ; feed po~iitioli (1,1 ) .. .. ........ .... 71

4.1 Table showing computed and mecasured values ft t imutial couplingof two loaded patches in a linea r array ('1%iFiOniII(mit for the E-plane.Patch size 6 x 4 cm.; edge sep)aration =lciii ; fet,,d positionl (1,1) . 75

4.2 Table showing coxnputcd and ineasured values of the( mutual -ouplingof two loaded patches in a linear array environlment for the il1-plane.Patch size 6 x 4 cm.: edge separation =1cii ; ffed posit ioni (1,I1) . .76

4.3 Table showing measured values of active inipIedlance of two loadedlpatches in at inear array environmnirt for the H- ,,Iaiie Patch size 6x 4 cm.; edge sepa~ration =1cml :. feed positin ( . . .. .. . .77

4.4 Table showing measured values of active ,!idI cft -Avpatches in a linear array environment for the E-planie. Patch size 6x 4 cmi.; edlge separation =leni ; feed position (1.1).. .... ..... S

4.5 Table showing the computed position of short, circuit loads requiredto keep the microstrip antennia patchies in a 5 patchi clement linearphased array matched to tihe feed network. The ;irray is ass umed tobe matched at a progressive p~hase shift angle of 0"'. Patch size 6 x4cmn. ; edge separation 1 cmn. feed position (1,I 1.. .. ... ..... 79

Chapter 1

INTRODUCTION

In recent years exte.,>ix e rcsear 11 an(l developineiit in microstril) antennas has led to

a miore thorough uiiderstanding of r h, application pot eitiall of microst rip radiators.

This has also t-st alli-shed the inicrostrij) antenna as a special topic by itself in the

field of microwave radiators. Tlie idea of uiinIrg a printed conductor on a substrate a6

a radliating eleinciit camne from the ri irost-ip transmission lines that were JIeveloped

inuch earlie-r. The r rend to ininiaturi?,e and, integrate electronic systenis has created

a positive deinamd for this niew generation of antennias. Over the vears time miany

advantapes offe-red miciost rip ;oit~ias li;tv, been usedl withi good results. The,

main advantaiges are- light wei ght, . owv cost, lhaliar confi guration and comipatabilitv

ithinegat~lcircults [1. %"'otip"lt~lu "lwihitgrtd, ~ trpatenscnbe loaded reactively to change

the electric field( lis'tribution between thle pat ch and the ground plane. By an

appropriate choice ()f the type and location of the react ive loads anumber of antenna

paramieters can be alteredl. For exT .tihe resonant frequency can be shifted, dual

bands of variable 1ban(1 sY' ariat11 4'iTn he 1iia(le [2], thle polariz at ion can he) alt ered

3]anid modaes produ cing rompi let ely different radiation p~atterns can be mnade to

resonate at the saine fr-cquen-v .fl. The r ctvel;ri5 ii ;-Ollkai -

resollant striuctulre,, such as (,p~eni circu",d' k),, -JIjIt (iI11ilc(1I I tis:,rlI! o

varactor (liodes. or As simlple as a 11cu (;iil! ( Jweilhap- :i:eic~li:at N

diode). The purpose of this wvork Nvts to itiidv !1e fo-hli ! 11Si1n rriV;JlA

Short-circuit loads to keep) the elements imatched 1i a scali( !icros trip alltt::init

array.

In the first part )f t hi" report such reaict ively V 'Ic e i nit s arc stuj Ieci

both theoretically and experineiitally. M(r'asrcrineimts 'v1-re liit oil icroustfip

i' Autors with reactive loads posit ionied at vaious" p(o d 5ii thc patch. The(se-

exlperlnlelit s were dI IiCe to vcrjfv. thle thlo''rV develo pcd f,- c suh loded eleinen-s at

tte Appl.ed Elect roinagiet ics Laboratory. Depa rtm liit of E1 ''l'c icl1 Eigiiierillig at

the U niversitv of Hou ston [5H [li reactive 1lac ili ig~k wabrv e v short circiti

!,)ads PI1N dliode., cani also beC iS( cd all( thIe rn itir cii() pat crn ;1 111 iput resis:tanice

,.dc reactance wvere mneasuredl.

The second part of tihe report is the siiulattion of a finite iiiiciost rip antonna

array and the conmput at ion of the ii iii t.'ial e ,upl iii . Tlj s loS bcen dIone for in any

c iifferent edge to edge separations of a pair o~f rcct angular Imicro t rip radiators: anid

he variation of the inlt nal linledalloe with I dllaliwt fori1 Ihi t i E- ilamle aiv! the,

H-pllalle was comlput ed anid im jasi m '4.Xl the vara I' f ti jc iu lcl im1pedanice

in a linear array with nllore titan t'wo e1inleit, ha> he "),viir: aikI both Computed

and(I mea-sured resi ilt hlave bee o)t ai n((I.

The effect oii tHie atct ivfe lij ct ilijedillee o)f clectioiiic s1C ~llmmmig bv changing

the progressiVe phase shift angle et we( ii tile array elt'iiieiit s ha , aIso ) Ien ii

l)utCe. Tis clianget III eleinilt imipedanice is bie to til anw a otI-T)i'i1i beWIetaI-

the elemzents of thi arzav. flt' chane inatv 'inielc t Ilcotat:ltt

in a scanned array c'ause, jt wer to bct reflectedi b~ack into tile feed. iThe '.SCol

Imicrostril) anitennia ele'lleits with iiiovaidt' shiort Circuit loads in order to, mlatch the

inicrostril) anit uit en mu lt s III a scanned ai ray is stu ied. I ils could Imnprove the

array performance. A preliminiary study of the feasibility o doing this is the mainl

focus of this work. B~efore dliscuissinig the details of this studiy, a brief overv'cw of

rnncrostriJ) elemnenits Is given inI Section 1 .1 andi 1.2.

1.1 The inicrostrip antenna

A microstrip aliteiiiia n', Ilustrnated InI Figure 1.1 in1 it-; simplest form. It hias a

radliating patch which is i-rally miadle of coppter anti a ground( plane se--,t:ateod fron.

the radiator by a dielectric subostrate. The oielect ric constant c, most. poTu11.ar1v

uset-l is usually II tue raige ()f 2.0 to 10.0 . Thle actual choice of this dIependl

on the speIf appli cat ion . Foi- ex ample. low frequency app )icat ionls requi le if

dielectric conlstailts to keep tile size smnall. Substrate choice and evaluation aIre

an essential p~art ()f thle olesignl jroco(llire of the uxicrost rip antenna. The specific-

sub)strate properties of Iportance are the dielectric constant, the loss tanigent.

and the variation of both with temiperature. Other factors Include homiogeneity'.

isotrolpici tv, anld di ifeunsi(mal ,tability l1j. The radhiat ion from a inicrosti-rip antennia

c'an, as an approximnatittn, he 1 t ttribhuted to tilt' fi nigi rig field between thle edgeo or

the aunterliua auirl thet itroinid 1 lme. This is iilirtritmed In Fhniure 1.2. AdItionil

information canl be founi( I in referece II

M icrostrip aiitcunwxI > ht% c iniany 't alv 1it;t1 ", >aiLU f I 1 ;1i ;1 (,1 inl .::rl ! 1

other stahl(Iard aiiteilicts u)pcatiing In tlio:r ri1~ i1 i Y Ii

having a center frey 1ienCV whici (-;n range fromi 100 ,IH z to ;ibove 7'0 (AHz '1]. T'hey,

are typically low in cost itand l 11ht in weighlt. Iley Cll 1 lB sed to 'encrate bo0th

linear and circular polarization. Dual frw v iencv ii icrost ri p ati tc ilits call7 Ihe ',ily

m1ade so that thev can be usced at ififerent tirailsini t and~ p 'eive- f"61n'~sC

Also, the feed lines required for the antenna can be asil fabicated together with

the antenna using the well- developed etching t echniqueis use,,d for nakin' printed

circuit boards.

Oi- the other han d ii crost rip mntelilas have at iarr. w 1,aiidwidt h AlIso. lie

edfire performance (parallel to the p)1lne () t1 l1e pat Ii)isvrvp adheiav

aI low power handling capalci rv [1]. There are al-() 1 ossibilt ics of thec eXcitatioin of

surfaice inodes which lead to the loss o)f rztidiat ed power.

Microstrip antennas have a wide area of apphicat ioii 1 lwv (an b'- US(_,i in

D,)ppler radari systems, satellite coinn111iii~catiomi. in1ISSil' teleimtrvi. fe d elcii.j nts

in complex amteiinias andl biomedical ipp dicat i i. The areas of zippll cionls air

continuing to grow ats the aw'arciness of their is , iimc reases,

Nlicrostrip antennas atre nimdelle( in imanyv tliffleent %vaivs witl iti 28li ')dcl hay -

ing its own advantages and disaidvantages. 1 lie mn st pl'~i lar 1:nals are the t ramis-

mission line inodel [1), the cavi ty miodel ( Lo rt el. Lithle wire grid n(elAgit,-xzl

andt B~ailIey) [11, and~ H ie In ()dal expanisioni imodl (Cairver Wid Cofe' [1j. There are

several other models which have been developed, inform at iii on which cani bc found

iiin [1].

1.2 Design procedure

A general overview of the design of rectangular microstrip antennas is given and the

relevant design expressions are stated. The first step in the design is to choose an

appropriate dielectric constant. As mentioned before the frequency of operation has

an important role to play in this selection. Some of the substrates that have been

largely used are [1] rexolite (Er = 2.6), duroid (r = 2.32) and alumina (f = 9.S).

For a dielectric substrate with a thickness of h, it has been shown by Schroeder 6j

that the practical width for an efficient radiator is

v=2f, 2(ii

The expression for the element length is

L 2A1 (1.2)

where f, is the desired resonant frequency, 6, is the relative permittivity. c is the

speed of light and Al is given by

Al = 0.41 2 h( E + 0.3)(IV/h + 0.264) (1.3)06, - 0.258)(W/h + 0.8)

where Ee is the effective dielectric constant given by+ )1) ( 12h( r +1) + (~ 1 +i )I2(1.4)

in which IV is the width of the antenna (see Figure 1.1).

Several authors have suggested different expressions for the radiation pattern of

the rectangular microstrip radiator [1]. Richards et al. [7] have found good agree-

ment between theoretically predicted values of input impedance and experimentally

5

obtained values using the cavity model appro;wh. Exp r.,.siwis for uthcr factors suchas radiation resistance, quality factor, bandwidth, can be Iouti in r'f,'re'lc, .

6

radiating patchi

d~ tc n. r~1 Q pQjch c o nr,!j ctSUb S t a

ground Qflune co-axial feed

(b)

Figure 1.1: The microstrip antenna (a) top viewv (b) side viewv

7

Fantenn

Radiating sots,

Figure 1.2; Radiation from the mic/rostrip antenna showing tie fringing fields

3/

Chapter 2

LOADED MICROSTRIPANTENNAS

By loading the inicrostrip antenna with shorting pins from the radiating surface

to the ground plane the input impedance and rcsonant frequency can be changed

over a wide range without affecting its radiation pattern. It is possible to change

just the input impedance and keep tlue ieso:-mit frequency constant by piacing the

shorts along a properly chosen locus of points on the antenna patch. Both single

and dual loads can be used to vary the input impedance, bult as is shown in the

experimental results, using dual loads yields a wider range of input impedance

variation. Also, dual loads produce lower cross-polarization levels than single loads.

The main purpose of this work was the study of using movable loads in a microstrip

antenna array to control the input impedance variation of the individual elements

as phase scanning takes place. The theory of these loaded microstrip elements is

given here and is followed by experimental results.

9

2.1 Theory of loaded microstrip antennas

A brief overview is given in this section of th( ox)l (5,l ('-lVe( , by Richards

[5] which are based on the cavity model analysis. The resuilts that were obtained

theoretically were verified by the author experimentally with excellent agreement.

In a microstrip antenna the existence of a source-frete fi'ld at resonance implies

the existence of a non-zero feed current. Iirchoff's voltag( law applied to the

lumped circuit model of Figure 2.1, which is a source-fre, microstrip cavity loaded

at a single point with load reactance XL implies

i.(Xl, + Xin)IIL = 0, (2.1)

where IL is the load current and Xi,, is the input reactance of the loaded cavity.

Richards [5] showed that Xi,, may be represented as

Xi, = X[ 4- XL (2.2)

where XF is feed reactance, X, is the input reactance at the load point of an

unloaded inicrostrip cavity.

All these reactances are functions of frequency and the roots are the resonant

frequencies of the loaded element. The resonant mode usually used in microstrip

antennas is the (0,1) mode. The normalized voltage distribution at the observation

point (x, y) on the antenna for this miode is given by

V(.r. y) = cos 7Y) (2.3)

where the element dimensions are a x b. This is zero on the center line y = b/2

of the patch. This line of zero field is called the node line. The addition of loads

III I

on the microstrip antexina buys this nodal line closer to or further away from the

feed point. The closer the nodal line is to the feed point the lower is the inplt

impedance of the antenna and vice versa.

Multiple loads requir, the loots of

Z I.) ( Z 22 + Z I. ).. .• Z 2,,,"2.(let .L .. 0 (2.4)

ZI.-I. Z21N;... (Z, ,, z -l

where Zij is the z-paraineter between the I"' and the 3;h port of a multiple port

microstrip cavity, and Zf, is the load :mpedance connected to the Zth port. One can

solve these characteristic equations iiuiierically for the resonant frequency. The

input impedance, voltage distribution and the radiation pattern can then be found

as described in [7]. Additional information can be found in reference [5].

2.2 Computed and measured results

For short circuit reactive loads we have XL = 0. Extensive experiments were per-

formed to study the effect of short circuit loading on the characteristics of a mi-

crostrip antenna. All experiments were done using 3M Corporation's CuClad 250

glass reinforced double clad PTFE with a dielectric constant of , = 2.43. Obser-

vation was made of the change in input impedance, resonant frequency and the

radiation pattern of the microstrip antenna for (lifferent positions of reactive loads.

2.2.1 Single loaded microstrip antennas

Extensive experiments were done on microstrip antennas and the results were com-

pared with those evaluated numerically. The experimental results for the microstrip

11

antenna with single short-circuit loads will , ,iscssed first. The patch size was

6 cm by 4 cm and the antenua wa~s fed at a point (1 .0. I.0) with a co-axial ( S.MA

connecting probe from the grmnd plane as is shown in thl, F'igure 1A. The rcoriant

frequency that was predicted was 2.465 GHz. The variation of input impedance with

frequency is shown on Smith charts in Figures 2.2 to 2.5 where the short-circuit load

is at a different position on the patch for each case. The magnetic current distri-

bution, which is equivalent to the voltage distribution, was plotted along the patch

edges for each of the cases considered using the cavity model There are four sep-

arate plots corresponding to the four edges of the radiator. The positive dir.ction

of the axis of these magnetic current plots points away from the patch edge in each

case. These distributions are illustrated in Figures 2.2 to 2.5.

The theoretical prediction of the resonant frequency [51. was verified experimen-

tally. The experimental results for input impe'dance agree 'd exceptionally well with

the theoretically predicted ones in all the cases except for those shown in Figures

2.5. This could be due to experimental error and not due to error in the theoretical

model. It was also shown experimentally that if the short circuit load is placedl on

the locus of points as predicted theoretically by Richards [5) the resonant frequency

was almost unchanged. This locus of points is illustrated for the case of a single load

in Figure 2.6. The variation of the input impe(lance as the loads are moved along

this locus is illustrated in Figure 2.7. Measurements were also made of the radiation

pattern both in the E-plane and the H-plane and for the cross polaried component.

These are illustrated in Figures 2.8 to 2.13. An important application of loaded mi-

crostrip antennas, as mentioned before, is control of the active impedance of the

12

array elements as the array is scanned. Switching between these loads can keep

the active impedance within ieasonable levels of impedance mismatch while at the

same time keeping the resonant frequency constant.

2.2.2 Dual loaded inicrostrip antennas

Similar experiments were performed for the dual-loaded microstrip antennas and the

results obtained again agreed well with the theoretically predicted values. Here also

the loads were short circuits. The results obtained are illustrated in Figures 2.14 to

2.21. The input impedance of the microstrip antenna could be varied over a wide

range with the change in the position of the short circuit load. It was observed that

the cross polarized fields were reduced slightly for the dual-loaded case compared to

the single-loaded case. The locus of load placement to keep the resonant frequency

constant as illustrated in Figure 2.14 was predicted theoretically and was verified

experiment, Mly. The variation of the input impedance as the loads are moved along

this locus is illustrated in Figure 2.15. As can be seen in the figure, the locus of

points where the load can be placed is symmetric about the center line or the nodal

line of the unloaded patch.

13

i s the input rE-oct nnce nt tne iode poinlt

c)f Fin unlodded microstrip civl' tu

Figure 2.1: The circuit model for the m-icrostrip anitennia

~2500 MHECornpute

10 MH7, Increment

tino h igeladdmcoti nen. ac ie6x4c load poiin(I7 4 fe

position (1,1) ;resonant freque'ncy 2.46Gb Gliz

C2500 MHz

opute6 -

0 MHL Increment s u

'I I

load

feed 1

Figure 2.3: The computed and measured input impedance variatio. and magnetic current distribu-

tion of the single loaded microstrip antena Patch size 6 x 4 crn , load position (2.25 0.95) feed

position (1,1) resonant frequency = 2.465 GHz

I (

* '2500 MHZ CMputed t~-'-el0 -MHz IncrementMesue ------ &-

Figure 2.4 The computed and measured input impedlance variation and magnetic current distribu-

t ion of the single loaded microstrip antenna. Patch size 6 x 4 cm. -,load position (2.5 .0.91 ) i-eed

position (1,]) ,resonant ftcqiic.., -z 2,465 GHz

17

250 M~ Coputd -- ~---E

50 MHz C teen Mesrd -*-

10M~ IcemntP7 W sue

load

feed - - __ t_

Figure 2.5: The computed and measured input Impedance varlationi and magnetic current d Istr Ibu-

tion. of the single loaded microstrip antenna. Pa tch size 6 x 4 cm ;load position (3.0 ,0.82 )feedposition (1,1) resonant frequency = 2.465 Gliz

Patch

Feed Point

Figure 2.6: Locus of short circuit locations for constant resonant frequency of a single loaded element.Patch size 6 x 4 cm. ; resonant frequency = 2.465 GHz

19

200-

0

100

- I/11

0.00 2.00 4.00 6.00

z-component of short-circuit location

Figure 2.7: The variation of the conductance G with the x-coordinate of a short circuit positiontaken from the constant resonant frequency locus for a fixed feed at (1,1) , resonant frequency2.465 GHz

20

0

S30 P ~~' -ot~ f

\

S..

-20

',./ I~ i c y ; Z - riZ

Fpigure 2.8- The measured F,-plane radiation pattern of the single loaded microstrip antenna (Patch

Size 6 x 4 cm. load position (-2.bbO.g3 ; feed position (1,I) ; resonant frequency = 2.465 GHz

21

E7 Plore Pot ~err,

Figure 2.9: The measured E-plane radiation patterni of the single loaded microstrip antenna. (Patchsize 6 x 4 cm. ;load position (2.0,1.0) ;feed position (1,I) ;resonant frequency =2.465 GHz

22

0

-3030 ID Pon2 Pott2

Fr2qU~ncy: 2-

D29r-22 Ste'D: 5

Figure 2.10: The measured F-plane radiation pattern of the single loaded m Icrostr Ip antenna. (Patchsize 6 x 4 cm. ;load position (2.25,0.95) -1feed position (1,I) -,resonant frequency = 2.465 GHz

23

0 H

--30 -- 30

-60 0

-00

'N'

\ 3C C.-

IT

5ingie I oe d

igure 2.11: The measured lI-plane radiation pattern of the single loaded 11,1crOstrip antenna.(patc!,ize 6 x 4 cm.; load position (2.25,0.95) ; feed position (1,1) , resonait frequenco = 2 465 Gltz

2,1

40 dB scale -a) 30 -20 -10 0

40 dB scale -30 -20 -10 0(b)

Figurc 2.1I2: (a)'The inleaslired anid computed input co- and cross-polar ised field patterns I n theE-plaire for short circuit loads at (2.00,1.12) and (2.0,2.88) (b) measured a nd theoretical patterns itihe 11-plane (computed Ee is zero). Resonant frequcncy =2 465 Gllz

25

0

-30 ~ ~ 3

6

-66 50?C G

-rGuc?,-cy 2. 47 Hz

Figure 2.13: The measured ff-plane radiation pattern of the single loaded illicrostrip antenna (patchsize 6 x 4 cm., load position (2,5,0.91) -,feed position (1,I) , resonant frequency = 2.465 GHz

Feed yolint-

Patch X

Figure 2.14. Locus of short circuit locatious for constant resonant frequency of the double loadedelement. Patch size 6 x 4 cm. -.resonant frequency = 2.465 0Hz

27

200-

CD0

C.) 100-

0.00 2.00 41.00 6.00x-cornponent of short-circuit location

Figure 2.15: The Variation of the conductance G with the x-coordinate of a short circuit Positiontaken from the constant resonant frequency locus for a fixeee at (1 ,I) resonant frequtey2.465 GHz

28

Com1puted 2&~E 5 D MO HELM e as ure C~~'- Vi MH z ILc C nen-

Figure 2.16. The computed and nmeasured input impedance var;ation of the dual loaded microstripantenna. (Patch size 6 x' 1 cm load positions (2 0.1 12) and (2 0.2 88) 1feed position I 1resonant frequency = 2.465 G- fz

29

\ /

Conputed -0*-- / ', 50 H

Fiue21.eaud -m----'-aidnptmecne vaiaio of 0,1 1, 15.e micosti canter Pac

\ i /

feed !

Figure 2.17. The m~asurcd input ilnpedlnce variaOTio of tK, due.! ', ded microstri! antenna. (Patch

size 6 x 4 cm. load positions (2 5,1.21) and (2 :,.2 79) fed position (11) ,resonant fr'-qtiencNy2.465 Glz

C(

Computerl --F-, - / SD?-

311 - /

Miue 8oeasured inutipea-eaiai- fth-ua nae -- osrpatnn Pac

size 6 x 4 cm. ,load positions (3.0,1.23) and (3.0,2.77) ;feed position (1.1) ,resonant frequency=2.465 GHz

31

/4eosured Results

0

mfcl

0

o0

C)

7-CIH~

/ ncrem, nt.: 1L , Hz

Figure 2.19 The measured input iinpcdancc variation of the dual loadcd microstrip antenna (Patch

size 6 x 4 cm. load positions (2 0, 1.1) and (2 0,2,9) ,feed position (1,1) Iresonant frequency2.'165 Glhz

T2

Measured Results

~/

/ ,2500 ....~"j /

'K/Increm2nt: 1),M0

Figure 2.20. "i.e measured input impedance variation of the dual loaded microstrip antenna. (Patchsize 6 x 4 cm. ;load positions (2.5,1.21) and (2.5,2,79) ; feed position (1,1) ; resonant frequency :2.465 Gllz

33

Heosured Results

Figure 2.21: The measured input impedlance variation of the duald loaded microstrip antenna. (Patch

size 6 x 4 cm ,load positions (2 75,1.23) and (2,75,2.77) -,feed position (1,1) ;resonant frequency2 .465 Glz

34

Chapter 3

MUTUAL IMPEDANCEBETWEEN MICROSTRIPANTENNAS

An expression for the mutual impedance between rectangular microstrip patche

can be obtained by the use of the reciprocity principle, in particular the Rumse.

reaction formula, and the use of the formula for the radiation from a unit magneti

dipole.

3.1 Radiation from a unit dipole

We shall use an expression for the field from an infinitesimal current element o

dipole and then use the duality principle and the reaction formula to develop al

expression for the mutual impedance. The expression for the electric field in spher

ical coordinates due to a unit dipole in the direction as shown in Figure 3.1 cai

be written as [8]

E rk +- e -jk.cos 0 + 4- o r r 1 ) kosin00 (3.1)27rk0, ( r2 r 3 4irk, r r

35

where is a unit vector in the x-y plane, ko, = w It-, is the wave number or

propagation constant in free space and (, = (1Lo/Eo)i/2 is the intrinsic impedance of

free space. We make use of the duality principle to obtain the expression for the

H-field. For this we substitute 1/(o for (, and get the following expression for the

H-field

H = -rko + - e- kocsO + -0 eko r sinOO (3.2)

An expression for the H-field can be obtained in rectangular co-ordinates by

using the following conversion factors.

((X) - Xi) + (yj - Y)) (3.3)((Xj i* ,) + (Yj _ yi)2)1/2

((jX - ) (yj - y') f 2 (3.4)

= )-2x ) i /2 (3.5)((Xj - X,) 2 + (y, - )2)

r. (3.5)((Xj i~l (Yj yi)21/2

((X, -Xi)2 + (y, _ i)2)lf 3.6

In the above equations the subscript j represents the coordinates at the observation

point and the subscript i represents the coordinates at the source point. For the

case of the microstrip antennas shown in Figure 3.2 we have the unit tangents along

36

the x and the y co-ordinates to be

(x(,+j) - Xi)

((x, - xi)' + (y, - y()371/)

=Yi1 - i

((X , - x) 2 + (Y - )(3.8)

Using these unit tangents, the cosine of the angle 9 shown in Figure 3.2 can be

represented as

Cos 0 = f. -Fr + 'F (3.9)

3.2 Mutual impedance using reciprocity principles

The reciprocity principle is used in electromagnetics and especially in antenna the-

ory in deriving the relations between the receiving and transmitting properties of

an antenna. Consider Figure 3.3 where a volume V is bounded by a surface S

containing two sets of a-c sources, J,, M, and Jb, Mb. Using Maxwell's equations

and the divergence theorem we can easily derive an expression relating the fields

and the currents on the two surfaces. We denote the electric and magnetic fields

produced by source a to be E. and H, and that produced by the source b to be

Eb and Hb respectively. The reaction concept for the fields of source a on source b

and vice versa can be stated as

< a,b >=< b,a > (3.10)

37

where

<.a, b >= JJ(E,,- - H,, Mi)dv (3.11)

Equation 3.10 can be rewritten as

J IJJ(E. - Jb - H. Mb)dv = fJIf(Eb.J, - Hb - M,,)dv (3.12)

For the case of the two microstrip antennas of Figure 3.2, Mb and J, are zero.

Therefore the mutual impedance can be expressed as

Z b = Ij b n M,,dS (3.13)

where I. and Ib are the feeding currents of antennas a and b.

The magnetic current distribution around the microstrip patch is obtained by

using the cavity model [5]. The magnetic current values are obtained at discreet

points along the patch edges. The expression for the H held due to the radiation

from an infinitesimally small magnetic current source is used in the above expression

for computing the mutual impedance between any two microstrip antennas. The

currents I and 1b are assumed to be unity in the numerical evaluation of the above

integral. The samples of magnetic currents along the patch edges are considered

as infinitesimal current sources and the contribution of all such sources is used to

compute the mutual impedance.

In practice, at microwave frequencies the scattering parameters [S] are gener-

ally used rather than the impedance parameters. Applying the well-known matrix

relation [9]

[S] = (Z - [I])([Z] + [i)-1 (3.14)

38

an n x n [Z] matrix is converted to an [S] matrix of the same order. Here since

n = 2, Sil = S 22 and S12 = S21, the equation for S21 is

S21 = 2Z 2,1/((Z, + 2)- ,(3.15)

where Z 11 and Z 21 are the normalized self impedance, and mutual impedance re-

spectively.

3.3 Application to arrays

The computation of the mutual impedance in an array with more than two elements

can be done by using the same procedure as above. In this case mutual coupling

between each pair of elements must be calculated. A linear array of microstrip

antennas is illustrated in Figure 3.4. Here the mutual impedance can be calculated

by taking into ac.ount the mutual coupling duc to all individual elements. This

can be generalized for the case of a planar array which is illustrated in Figure 3.5.

The relation between the voltages and the currents flowing in the microstrip an-

tennas, which can be modelled as loads, is illustrated in matrix form as shown below

V Z21 Z2 ...... Z2N I2

W. ZNIZN2 ...... ZNN IN

Therefore, we can write the equations relating the voltage and currents as

VI = Zl1 I 1 + Z 12I + " + ZININ (3.16)

39

Assuming IN = 1 •ej N where ON is a specified phase shift angle, one obtains the

expression

Zactiv = Z11 + Z12.e±(02-' ) +" + ZIN-C (" - 6 ) (3.17)

In the above expression ZlN can be calculated using equation 3.13 and therefore,

we can calculate the active impedance of every element in either a linear array or a

planar array. As stated earlier the input impedance of loaded microstrip antennas

can be varied by reactive loading while keeping the resonant frequency and pattern

constant. As the array is scanned the varying active impedance of the array elements

produce an impedance mismatch which can possibly be minimized by loading the

microstrip antenna elements and switching the loads accordingly as the array is

scanned. The performance of the array may thus be improved. This is considered

in the next chapter.

40

3.4 Computed and measured results

3.4.1 Computed results

A Fortran program was written by the author to evaluate the mutual impedance

effects in a rectangular patch array. The program enables one to simulate the array

on the basis of the model developed previously in Section 3.2 and to compute the

active impedance of each element of a linear array with any patch size, edge to edge

separation, resonant frequency, number of sample points considered along the edges

where the magnetic currents have been computed, and any progressive phase shift

angle. The run time of the program depends on the number of sample points of

magnetic currents used in the computations. The program can be used for both

the E-plane and the H-plane scan. Here the plane perpendicular to the patch and

parallel to the electric field is called the E-plane, and that parallel to the magnetic

field is called the H-plane. The magnetic current samples were obtained by using

the cavity model. These magnetic current samples are evaluated at the resonant

frequency found from the program written by Dr. W. F. Richards of the Department

of Electrical Engineering of the University of Houston. The two programs have been

integrated together for the purpose of obtaining these computed results.

The various computed results are shown in the figures that follow. The variation

of the magnitude of mutual coupling was computed as a function of distance between

the patches. This is shown in Figure 3.6 for varying distance in the H-plane. In

Figure 3.7 we see the variation of mutual coupling with distance in the E-plane. The

change in mutual coupling with distance is observed to be less for the E-plane case

than for the H-plane. The mutual coupling is represented in Figures 3.8 and 3.9 in

41

decibels. The results obtained are approximate due to the approximate model used

for their computation. It may be stated here that for the purpose of examining the

feasibility of the use of loaded elements in an array for improved performance, the

app. .)ximate model is sufficient.

3.4.2 Mutual impedance measurements

Extensive experiments were performed to measure the mutual impedance between

antenna patches with varying separation. The patches were made on printed circuit

board manufactured by 3MI having a substrate of relative permittivity E, = 2.45

and substrate thickness of 0.152 cm. The patches were fed with a coaxial SMA

connector. All measurements were made on the HP 8510 network analyzer with an

absorbing chamber around the antenna. The frequency was varied from 1.65 to 3.15

GHz. Some of the results that were obtained are represented in the Figures 310

to 3.19. In the case where there are more than two patches present in an array the

effect of an open-circuited patch on the mutual coupling of any other two patches

was studied. It was found that this effect was more pronounced in the H-plane

than in the E-plane. A possible explanation for this could be the effect of surface

waves being more predominant in the H-plaiie than in the E-plane. Figures 3.10 and

3.11 illustrate this effect which contributes to the error in computation of mutual

impedance using the classical reaction formula. A correction to this is proposed in

the paper to be published [10] by D.R. Jackson, W.F. Richards and the author.

42

3.4.3 Mutual coupling for loaded patches

Measurements done for loaded patches are shown in Figures 3.20 to 3.28. In these

experiments two patches were studied in a linear array environment both in the

E-plane and in the H-plane. The HP 8510 network analyzer was used to make all

measurements. It was noted that significant error was introduced in all the mutual

coupling measurements from the reflections from the roof and other objects in the

laboratory. Therefore an absorbing chamber was made around the antenna and

measurements taken were then found to be very repeatable. The loads, as in the

previous cases of individual loaded antennas, were short circuits. The position of

the load was found to be critical in order to keep the resonant frequency unchanged.

Measurements were made for both single and double loads. The load position on

one patch was fixed and the load on the second patch was moved along the locus

(to keep the resonant frequency constant) and the effect on mutual coupling and

the active input impedance of both the patches was studied. It can be seen from

these results that the mutual coupling and also the active input impedance of the

patches can be varied with changes in load positioning.

43

ThE unit. cjrr-ent- ,p

Fi~gure 3.1. The unit current sourcc co-Ordinate systemn used for derivation of the fields

14

- IM E . _I

-jS' n- c

Figure 32 ' heoo ir n the observeddge mi r s rp antennas show ng the angle 0 betw een sam ple

poins alng te anennaedge

£E b

H b

Figure 3 3: The i !ustratIin (-f the rfcC)Ticjt) r Iijjipjc usingF Lhe bounded volumes

feed line microstrip entennd

Vl-e series fed )nr- drrdy of rnicrostr, antennds

rn ic ro str pD ~n te nn d

Food netwoi-K

The pdrol ci fed 1 inetnr errtoy or microstrip antennas

Figure 3.4 The lineatr Airr't, jnicr ostrip antennas showing different feed configurations

4 7

rnicrostrip ante2nnw1

Feed network

Figure 3 .5 The planar arraym il ',~j~pe fict1gutn sraadprallfd adators

Zo:1

o

zo

O

u01 ..~

GCO . 0 ? 60 .40 4. 20 5.00 5.80 6.60 7. 40 .ZO

H-PLANE EDGE SEPARATION IN CM

Figure 3.6: The computed variation of the mutual impedance between two microstrlp antennas (size

6 x 4 cm - frequency = 2.35 GHz ; c = 2.45) as a function of distance in the H-plane

49

)

.00 1'.20 1'.40 1.60 1.80 2'.00 2.20 2. 0 Ell . 0

E-PLANE SEPARATION INJ Cm.

Figure 3.7: The computed variation of the mutual impedance between two microstrip antennas (size6 x 4 cm ; frequency =2.35 GlIz = 2.45) as a fmction of distance in the E-plane

50

co \

100 1 PO 2 60 3 4 0 4.2 5.03 S. 80 5. 60 7 40 8. 20H-PLANE EDGE SEPARATION IN CM

Figure 3.8: The S12 representation of tile computed variation of the mutual impedaace between two

mlicrostrip antenna-s (size 6 x 4 cm frequency =2 3.5 CHz) a-s a function of distance in the H-plane

i5

Z-)

Do

on

------------- --

1 .00 I 20 1 40 1. 60 1 0 2.00 2 . 40 2 60 2.80E-PLANE SEPARAT:ON IN C)N

Figure 3.9: The S12 representation of the computed variation of the mutual impedance between twomicrostrip antennas (size 6 x 4 cm ; frequency = 2.35 Gtz) ws a function of distance in the E-plane

52

-. 30dD3~_______I

I ~~(curve A) T7 ___

-40dB (curve B

- 70d

ST TI IOC H

a 10 ~OO~ i-~FREQ(GHz-)

Figure 3. 10: The measured .513 of a three patch linear array with the presence (curve A) and absence(curve B3) of the second patch in the If-plane. Patch size 6 x 4 cm.; edge separattion =1cm :feedposition (1,I) -,resonant freque ncy =2.35 GlIz

53

hp -E PLAN 1I2iPrFr FU

1 (curve' A)

30dB ~ cre )- --

S 13(dB) -4SdBT - __

I _ __ __I

1-6 S K,31STC:I 3. _so- O G~ R QG

Figure~~~~~ ~ ~ ~ ~ ~ 3.11 Th esrdS3o he ac ierara ihtepeec creA n beccuv )o h eodpthi h E-lne Pac siz 6- x 4cm; egescaraon= nI ee

poiin(,);rsnatfeuny=23 GIz

* I _ _ I _54

hp 4 'TH LIN- . HRH-PLAN -T 2

__ lO dB L _ . . .. . ... . . .. .... . . .

20dB___(B _3_d ---- -

20dB L /..

_,__ _ o, . . __- - / - V

--40dB ---- /I __

-. ... ' _ . . ." ... .

1.65 START 1.500 0 0Hz 3.15

STOP . 15ozoooo Gz FREQ(GH)

Figure 3.12. The measured S12 of a fnur patch linear array in the H-plane. Patch size 6 x .4 cm.;edge separation = lcm ; feed position (1,1) ; resonant frequency = 2.35 GlHz

55

4'ThCH ....R HK[-) L ANV2 TL-.-- ...

1OdB- ___I .

- -20dB- Tt I

S23(dB) -30dB -- -- --

/I\I

- - \ :/ ___

GD START IGQCQ0 ~31STOP 5 H H

Fig-ure 3.13: The meastired S23 Of a four patch linear array In the 1- Plau atch size 6 x *lcm..edge separation =1cm feed position (1,1) ;resonant frequency =2 3,5 GlIz

6 6

h~p 5 5ATCH L IME F-R ARTT1 TO 2-10d.B L -_ _ .. .. .. __ ___ _

I ! j / ,I

Sod)-30d13@ - A, /f . I

S1 (B 3O B -_ _ __ _ _ -- _____ ---.-- - -- ,--- + -- _____ I,____/ I

I I C , E .- H

-40dB F "____._ - r.-

Figure 3.14: Tii- measured S 1 2 Of a fiv'e patch linear arrav III the 11-plane. Patch size 6 x 4 cmedge separation 1cm feed position (1,I) resonant frequency 72.35 Gliz

i ' "7

hp S ATCH LINEAR AR AY 2 TO3__

- lOdB i T/MARER1

25 GHz T _ _

- 20dB /A

A !

-40dB ! . . _ ,

S2(d)-3cB ~ I__/ E 2

-50dB _. ______ _ t "_. J_ _ _ _ _ _ _ _ I _ _

1.65 3.15

STOP -3.150o00OO Hz 'JL Q(GHz)

Figure 3.15: The measured 523 of a five patch linear array in the fl-plhne. Patch size 6 x 4 cm.edge separation = 1cm , feed position (1,1) ; resonant frequency = 235 HGtz

I C! 0CZ

edgec!B sep3OdB ,----- Ii-fedpsto I 1I I__ resnan ___ qeC = 2__ 35 l I

________ 59

h DATERMY417H _5 _

- 10dB.LfRFfY4T I-P N

- 20dB _'__-

S45(dB) - 30dB- A _

I II I

L3qdB___

S t . .. .. .. .. ...

STOP o- FJEQ(GHz-)

Figure 3.17- Thp measured S4 of a five patch linear array In tho 11-plane. Patch siz- C 4 cmedgc separation lcm ,feed position (1,1) , resonant frequency 2 35 lz

60

CD CDCD

-0.00 ?.00 00 6.cE 000 0.0I-- 4 -- L_ -fi~

0 G0 U PUT F0 PE AS U RE

CI)

CDc(fC)~

C'C)

",4-,

I I0

0.00 2. 00 00 0.00 8.00 10.00 .00H-- PLANE SEP-,Ai O , O N,

Figure 3.18: The computed and measured 11-plane coupllng in a twc patch linear arra. Patch size6 x 4 cm.; edge separation = lcm feed position (1,1) resonaant frequency = 2.35 CIfz

CD CC).C

- 0 . 20 1 .60 2.30 2.

c 'COPP U T ED

c22o >

C) ".E lRE C

C)ncr)

c) I

8 0\ 2\ 2 0

('--C)

L ~ ~ ~ ~ 1" SF

C) ... -

0.80 ;. 20 .60 2.00 .40 .81 .; o-

-? ,:'L SFA/,A, T ,0 H v, CwS

Figure 3.19. The computed and measured I-'plane coup1 ]g in a tto 1,,cih hiiear 6:rray. Patch iz.

6 x 4 cm.. edge separation lcm ; feed position (1 1) :rsorian! frc.queuiy -c 2 35 G liz

t.

"30113

S70o--' .15o00000e G.i FREQ(G, :)

I load

and its magnetic urrent distribution. Load positions : Patchl (2.0,1.0); Patc2 (2 0,1.0) patchsize

6 x 4 cm.; edge separation = lcm ; feed position (1,I) ; resonant frequency = 2 35 Gl.[z

6i3

- 1 O d B . ", E

- ?Odg

S, 2(dB) -3odB - - - - -

-40dB

-30dB _ _

OH, 31,G,3!-,TART -65CeZ01e2O GH, :

s7oP 1 ozeoeOO OH, FREQ(G112

feed 1 o

Figure 3.21: The measured E-plane coupling for a single loaded patch in a two patch hitear arra\

and its magnetic current distribution. Load positions: Patchi (2.0,1.0) , Patch2 (2..5,0.91). patchsize 6 x 4 cm.; edge separation = 1cm ; feed position (1,]) , resonant frequency = 2 35 Giz

6I

S,-,(dB) -30dB z 711 1 _

- -0dB -__ __

-50dB - _LC.3z FREQ(GP) '

fee xd load

Figure 3-22: The measured E-plane coupling for a single loaded patch in a two patch linear arm'%and its magnetic current distribution. Load positions: Patch 1 (2.0,1.0) -1Patch2 (1.5,1.02). patchisize 6 x 4 cm., edge separation =1cin feed position (1,I) ,resonant frequency =2.35 Cliz

65

-320dB

- 50dB __ __J i___ ___

2 ~~o~ ~FREQ(GHz)

feedc

Figure 3.23: The measured E.plane coupling for a double loaded patch in a two patch linear array

and its magnetic current distribution. Load positions: P~atch] (3.0,1 23) and (3.0,2.7) -Patcli2

(1.5,0.91) and (1.5,3.09) -patch size 6 x 4 cm.; edge separation =-n c feed position (1,1)

C 6

hp OAO) A 1~ 0.3 T II~ c2( .. 12/2.

- 10dB FMAFRER 1 1* .35 25 [6Hz _

- 90dB 1 _ _

Sj 2(dBI3-

- 30dBT,~___ / _____

L-/

-4('dB' 151' JjOO C_ - PR___ E 0(C

o j

F i u e 3 . .. T e e s r d - l n c u l n f .- d OI :C l a e a c n a tw a c n a r a

an ts m gtcL curen d ' -t b____ Lodpstos.P__ 301 3 n 30,47, P t

(2 ~~_______ 5,I2 n 2528 ).p thsz m, d esp rto c edp sto 1I

iGS <,n~ ~ r~. ~ . 3.1

p 0 / OADEO -P Ci ~- lOdB -- .....

__ _ I . . .. -

- 30dB - ------.. .--

- 40dB __ - - - - 7- 50dB . . . .... _.,. _... ,

STiR i n1.G5 s5T, r'.rn'v:,,,r'.,, 3.1.3

2 1TOk-: G9 FPt'E Q(G H z

Figure 3.25: T . "e.C.udL-... , e crunpJm/g foi a double loaded p,:Itch in Lav,o path lj arra'and iLts magnetic current distribution Load positions P'atch 1 (3 0,1 .23) aid (3.0,2,) . Patch2(3 0,1 23) and (3 0,2,77) , patch size 6 × 'x cm., edg,: separation - cm , feed positon (1 1)

C--,- C_

3 . 150"000

Figur ~ ~ ~ ~ ~ ~ ~ ~ ~ -i .6 h csrdS ai o o obela e a h:i at~)p ,c ar..L

positions.~ ~ ~ /ach (3 0 12 ) a d ( ., ) !P t h 2 5 1.1 rd ( , 9 a ic 6 x 4edg scaNIT CI edp~to 1I

69N

Ap

-- -- /

/ -. k/ " \'. ,

.. . 7",, 2.46 GHzZ .7533 - i.31 A-

START 1.650 O 0 >,0 GH-STOP 3. 15OeCo0 o G H;

rI gurc 3.27: The me&sured Sil variation for a double loaded patch in a two patch linear array Loadpositions: Patchl (3.0.1.23) and (3 0,2.7) ; P-itch2 (2 0,1.12) and (2 0,2.88) , patch size 6 x I, cm.,edge separa'ion = lcm , feed position (1,1)

/ '\ 7.. ' \

;'- - -//'\ ',

-.'- - - - ,,

,,- /' -- / / " ,

• ,, ' "

"/ /

• . / ~ / /

' t'

7- 7 Freq 2 .46 Ghz

-S T ART 1 GPO Hz,O,- 3 150CQ000 0 0Hz

Figure 3 28 The measured S., variation for a double loaded patch in a two patch linear array, Load

pOsitiOns PaLchl (30 I 3) and (3 0,27) Patch2 (3 0,1 23) and (3.0,2 77) patch size C x 4 rm.,

edge separatioi = cm ,feed position (1l1)

: 11

Chapter 4

USE OF REACTIVE LOADSFOR MAINTAININGIMPEDANCE MATCH INSCANNED ARRAYS

It has been found that in an electronicaly s(diillIcd array of inicrostrip antennas

the active driving point impedanc, of each clement varies si.gnificantly wben the

array is scjn~ed by ei t,'r varying the :pogre-ssive, si f or lb chani

the frequency [1!]. Vhen this happens, tlie transmission feed lines that are usu-

ally matched to the airay elements for maximun power transfer will experience a

mismatch and consequently a large amount of powcr will be reflected back instead

of radiating out. This reflected power is usually redrected by th, circulator at the

feed point and is coupled to an absorbing load to prevent it from going back to the

transmitter. But this loss of ,lower means los.s (f acciuracy and range in a radar

svstern.

The previous ,,tudy of loaded microstrip anitennas has revealed that the input

impedance of an antenna can be changed by placing reactive loads at specific points

72

on the antenna so that the resonant frequency remains unaltered. It was the purpose

of this study to see whether mutual coupling and the active impedance bet ,w.e in

microstrip antenna elements can be changed by positioning loads properly on the

microstrip antenna. A prograxn was wri'. en by the author to compute the mutual

coupling using the loaded elements for a finite array. An approximate theory is

used to compute mutual impedance between elcments which did not considcr t.e

effect of other patch elements when computing the imutual coupling between any

two elements. An exact expression should be used for accurate results as in [12 i and

[13]. For the purpose of the feasibility study the approximate' model tlha t was used

was sufficient. The computed results were close to experimentally obtained values

in most of the cases considered. The results, indicated that the mutual coupling and

the input active irnpedanc(e in an array cn be varied by using loaded elements. This)

was verified experimentally and the results ()taired are shown ir Figures 3.23 to

3.28 of the previous chapter. A comparison of the ,omputed -o- , ns;,red results

is shown in Figure 4.1 for the E-plane linear array and Figure 4.2 for the H-plane

linear array. A comparison of the computed and measured active impedance of two

loaded patches in a linear array environment is shown in Figures 4.3 anJ -1.4.

The position of the reactive load on the autenna was computed to match the

antenna elements at various scan angles as shown in Figure 4.5. The process is

iterative because if one element is matched by changing its load position. then

other elements of the arrav become mismatched. This requires changing the load

position iteratively until a reasoriable match is obtaind for all the array -leimemi s.

A test case of a fivc-patch linear array was considered. It was initially as:,mied

721

that the array was matched when all port currents were in phase. As the progressive

phase shift angle or the difference in phase angle between adjacent port currents is

changed to 22.50, 450 and 600, the angle that the main beam makes with the broad-

side direction also changes. The load positions are changed to reduce mi-nmatch.

These load positions and the change in the magnitude of ieflcctioii coefficient are

shown in Figure 4.5.

How much the performance of an array can be improved remains to be sen as

this study con~inues, but from these results it does seem that an impedance match

of array elements can be maintained within reasonable limits as beam scanning

takes place.

IT

MUTUAL COUPLING- OF A TWO PATCH4 1,NEAR ARRAY 'VTTH

LOADED PATCHESF-Plane Dual Loads:

Load Pos'Itons:Patch 1. Loadi (3.0 ,1.23) Loa62 (3.0 2

Patch 2: Loadi (1.5 )0.91) Load2 ( . .9

COMPUTED S21=-22.6,93 dBS -MEASUFRIL 311-lKO1

Load, PosiuionlsPatchL 1: LOad1 (3.0j J.23) Load3 " (0

Pa-zcl 2: -Lp' (2 1.12) :Lod<2, (2 233

Patc. 1 oad (3.0 1 23) , Lo n'2 (3 0 2.

PaIchL 2 Loadl (70 ).2) Loa2 (2 0 2 {

COMvPUjTED S2]1 579 lB-S - MEASURED S2 9 7 BS

Figure 4.1: Table showing computed and mneasured values of the mutual coupling or tw loade

patches in a linear array environmient for thei E-plane. Patclh s~ze 6 x 4 cm; edge separation = cm:l

fe )osl',Icjn (1,I)

MUTUAl. COUPLING OF A TWO PATCH LINEAR ARRAY WITH

LOADED PATCHESH-Plane Dual Loads:

Load -Positions:Patch 1: Loadl (3.0,1.2,) Load' (30 ,2 7

Tatch 2: Loadl (2.0 ,1.21) Lnad2 (2.0 ,2 79)COMPUTP' S2 3.i13 Kc"S MEASURED S21=-21 21, !Bc

* ,o, )i}ositions:

-t- . K J:, i () .1.23) Load2 , 0 277)c 2 L -!c' 1.12) L.oad- (2 S 2 88)

Datch 1: -<Da,,I (2.0 a (.0 2.77)tatch 2 >ai (,C) 1.23) Load2 (2.0 ,277)

CO,,,-, D S2i - 1i :,64 d-BS , , A CT P' S 21 ...- 407 7 ,

Figure 4 2 Table showing computed and m asured values of the mutual coupling of tv.- i:,:dpaTches in a linear arra) environrimeit for the 1l-plane Patch size 6 x 4 cm edge separation "feed position (1.1)

ACTIVE !MPED-ANCE OF A TWO PATCh T INEAPF A "'Y\,Vv'

LOADED PATCHES:!H-Plane Dual Loads~

Load Fr>2tMODSPa t ch -1 L o a (I (3 0 112 3.) ,L o ad2 (0 7' -azch 2 Loadi (2.0 P3) oad2 (2.('

MEASIYKED 75.33CE ~13

Load PositjIOns.?acrh jLoadl ('.0 13 IaD 2'~ 0

FPat Q T oadl (2.5 1 )Loa1 (' S

Fatcf 1 Load! (3).0 1.213) ,Lo,.,z (3 7?-r} 2: LoadI (3.0 Loa Q

MEASUREDZQ 7 5. 1 0 6

Figure 4 3: Tah~e showing meas5ured values of active impedlance of two loaded patches .n a lineararray enivironiment for the 11- plane Patch size 6 x 4 cmii edge separation =1cm = feed position (1,1'>

ACTIVE IMPEDANCE OF A TWO PA 'OH LINAR ARRAY V/LIP

LOADED PATCH ESF}--Plane single Lods.-

Load PositionsPatch 1: Loacl (2.0, 1 0Patch 2: Loadi (1.5 1 029

.\,EASURED ,,,, 78.93z! 44(

Loai ]Po. ]tions:

Patch 1 Lozdi (2.0 1 0)Patch 2 KJI (2.0 ,0)

>IIASTJR D Z,;, '10 r- ifl "9-, -

luad Fosmoirns:

Patch 1 Loidi (2.0 1.0

Patch 2: Loadi (2.5, 0.91jA4EASURED Z,,., = 51 66 -" jO 5527

Figure ;.4A: Talle showing neasured values of act ,c im'daice of two loaded patches mi a lineararray environment for the E-plane. Patch size 6 x 4 ci; edge separatiot -= 1cm feed positicn (1,!)

progresslYe

phase 5hift 0.0 22.0 43 0

00 0.179 0.30

-0.0 0 1 02 0.228

0 00 5.17E-2 0. 185

0.00 448E-2 0.41

0 Oi3 0.12 0 22

[ n t ref-lection .ot ffi i.aLb _I._._ t . .__

progressiYe

phase 'hift 0 0 22.5 e5 0

4- + + i-

1 O0 0147 9.73E-2

2C ', 5 09t-2 4+4'-

24 .57-2

0 CC S 66t-2 +.4[

S C 2.25E1-2 1

tr]AD POSITIOS

( 5,1. 17) (1,5,2.93)

(2 25,1.12) , (2 25,2.80)

+ - (2.75 ,1 i') (2.7 .5,2.8 3)

Figure 4 5: ]'able sho'i ng the Colpliuted position of short circuit loads required to keep -,e miCrostriL

anaenna patches in it 5 patch element Lnear phased array' matched to the feed netwc:k.. The array

is assumed to be matched at a progressive phase shift angle of 0'. Pat:h size 4' 4 cm edge

separation -- 1cm ; feed position (1,1)

79J

Chapter 5

CONCLUSIONS

The experimental study of loaded microstrip antennas showed that the properties of

microstrip antennas can be changed quite easily to suit a given set of applications.

By placing reactive loads on the microstrip antenna one can change its resonant

frequency and input impedance. If these loads are placed appropriately, then it ispossible to change only the input impedance of the antenna without changing the

resonant frequency.

The theoretical and experimental study of microstrip antenna arrays showed

that the active impedance of an array element is affected significantly by the mu-

tual coupling effects of surrounding elements. This active impedance was found

to vary with scan angle in phase scanned arrays. This causes an impedance mis-

match at the feed point of array elements resulting in a loss of radiated power cuo

to reflections. The feasibility study on the use of loaded microstrip antennas to

reduce the mismatch of these elements in a scammned array showed that dynamic

controlling of input impedance of array elements is possible. This can be achieved

by appropriately changing the reactive load position as phase scanning takes place.

80

The reactive loads could be shiort circuits for whichi PIN dliodes cani be used allowing

switching appropriately fromn a computer during phase scanning.,

Bibliography

[1] I. J. Bahl, P. Bliartia, "Microstrip Antennas ", Artech House, Inc. 1982, pp.

4-80.

[2] W. F. Richards, Shayla E. Davidson, S.A. Long, 'Dual band reactively loaded

microstrip antennas", IEEE Trans. on Antennas and Prop., Vol. AP-33, No. 5,

May 1985, p. 556.

[3] W. F Richards, Y.T. Lo, D.D. Harrison. "An improved theory for microstrip

antennas and applications", IEEE Trans. on Antennas and Prop. , Vol. AP-29.

No. 1, January 1981, pp. 38-46.

[4] NVA. F. Richards, Stuart A. Long, "Adaptive pattern control of a reactively

loaded, dual mode inicrostrip antenna", International Telemetry Conference

Proceedings, Las Vegas, Nevada, October 1986, pp. 291-296.

[5] W. F. Richards, Stuart A. Long, "Impedance control of microstrip antennas uti-

lizing reactive loading", International Telemetry Conference Proceedings., Las

Vegas, Nevada, October 1986, pp. 285-296.

[6] K. G. Schroeder, "Miniature slotted-cylinder antennas", Microwaves, Vl.3.

December 1964, pp. 28-37.

82

[7] WV. F. Richards, "Dynamic control of the input impedancue of a microstrij a:

tenna using loaded elements", Technical Report S-il1, Applied Elect ron iagi 1,

ics Laboratory, Department of Electrical Engineering, University of Hou ;to:,

1986.

[8] Robert E. Collin, Francis J1. Zucker, editors, "Antenna theory" part I and 2.

McGraw Hill, 1969.

[9] Sheng-fuhi R. Clhang, "Mutual coupling between inicrostrip anten7nas: 1'il ;i

bitrary spacing and orientation", IEEE Trans. 071 Antennas and Prop. y to

published)

[101 D. R. Jackson, NV. F. Richardls and Ajaz Ali-IN'hlami "An uxact co'' ,;-1nE

for microstrip patches". IEEE Trans. on Antcnnas and Prop.. ,to be pi-s

'[11] R. C. Hansen, "'Significant phiased array papers". Artech House. 1973.

[12] David M. Pozar, "Finite phased arrays of nflcrostri1 ) patches". IEEE Tra.;,

Antennas and Prop., Vol. AP-34, No. 5. Mlay 1936G. pp. GCjS-665.

[13] R. P. Jedlicka, 1K. R. Carvei . "Ilutual coupling between inicrost rip ucnt

Proceedings on the workshop on printed circuit antenna technology. New M,~x

ico State University, Las Criics, New 'Mexico 1979. pp. 4.1-4.8.

83

MISSION

Of*Rome Air Development Center

RADC plans and executes research, development, test andselected acquisition programs in support of Command, Control,

S Communications and Intelligence (C3I) activities. Technical andengineering support within areas of competence is protded toESD Program Offices (POs) and other ESD elements toperform effective acquisition of C3! systems. The areas oftechnical competence include communications, command and

N control, battle management information processing, surveillancesensors, intelligence data collection and handling, solid statesciences, elect romagnetics, and propagation, and elc, -)nic

Y' reliability/maintainability and compatibility.