Advances in Radio Science, 3, 143–146, 2005 SRef-ID: 1684-9973/ars/2005-3-143 Advances in © Copernicus GmbH 2005 Radio Science

Radiation Efficient Unidirectional Low-Profile Slot Antenna Elements for X-Band Application

C. Locker,¨ T. Vaupel, and T. F. Eibert FGAN-Research Institute for High Frequency Physics and Radar Techniques (FHR), Neuenahrer Str. 20, 53343 Wachtberg-Werthhoven, Germany

Abstract. Slots in metallic ground planes are very promis- be placed closely to the slot, resulting in a typical stripline ing candidates for conformal antenna applications. However, or triplate configuration. Unfortunately, this structure admits a low-profile unidirectional antenna requires a back reflector parallel-plate propagation and the discontinuity slot/stripline close to the slot and the resulting stripline feed causes strong excites the corresponding modes. Higher order modes can excitation of parallel-plate modes. In this contribution, we be avoided by choosing a thin substrate, because their cut-off consider unidirectional reflector-backed slot configurations frequencies increase by decreasing the distance between with parallel-plate mode suppression by shorting pins. Start- the ground planes. The suppression of the fundamental ing from a parametric study with respect to shorting pin lo- mode, however, is a challenging task, since it is a TEM type cation and back reflector distance, we present a stripline-fed mode starting at 0 Hz. One method to suppress the TEM rectangular slot element with radiation efficiency of more mode is the use of twin slot configurations for destructive than 80% and a bandwidth of about 5% at centre frequency interference of the excited waves. An alternative method is 10 GHz. A careful optimisation of shorting pin locations the cavity backed slot (Galejs, 1963; Biebl and Friedsam, guarantees reliable parallel-plate mode suppression without 1995; Omiya et al., 1998). Metallic walls force the electric deteriorating the slot radiation behaviour. Coupling coeffi- field of the TEM wave to zero. A similar procedure is cients between parallel and aligned rectangular slot elements the placement of shorting pins around the slot aperture, are presented. For increased bandwidth applications, a bow- connecting the slot’s ground plane with the back reflector tie slot element with about 8% bandwidth and radiation effi- (Sommers, 1955). Compared to cavity backed slots, only ciency of close to 80% is proposed. little information about guidelines on positioning of vias is available in literature. Some papers treat stipline-slot-fed patch antennas (Brachat and Baracco, 1995; Bhattacharyya 1 Introduction et al., 1998), where a theoretical analysis for via-slot interaction is given in the latter. Due to their low profile, light weight, robustness and low cost, antennas in printed circuit technology are widely In this contribution, a parametric study is concerned with used in electronic warfare, communication, automotive, the influence of via and reflector positions on the input etc. Traditionally, many design instructions can be found in impedance and radiation efficiency of a single rectangular literature for patch antennas (James et al., 1981). However, slot element. Then, an optimised unidirectional slot element due to its resonant behaviour the radiating patch element printed on a very thin substrate with parallel-plate mode sup- has only narrow bandwidth, unless thick substrates are pression by shorting pins is presented by simulated and mea- used, which is disadvantageous with respect to integration sured data, based on the approach proposed in (Sommers, into non-planar surfaces. The surrounding of the patch is 1955). Towards array applications, we present mutual cou- inherently dielectric, which benefits propagation of surface pling investigations for different slot arrangements. A second waves resulting in low radiation efficiency. radiating bow-tie like aperture for bandwidth enhancement is presented by numerical and experimental results. The radiating slot element is an attractive alternative. A slot in a metallic ground plane is the complementary structure of the corresponding electric dipole. Since the 2 Rectangular Slot Element radiation pattern of the slot is bidirectional, a back reflector The geometry of the investigated rectangular slot element is must be used to avoid back radiation. With respect to depicted in Fig. 1. The slot is located in the upper ground antenna integration in curved surfaces, the reflector should plane and is excited by a stripline located in the centre of Correspondence to: C. Locker¨ the substrate with r =2.2. In order to suppress the parasitic ([email protected]) parallel plate mode, the upper and lower ground planes are 1442 C. C.L L¨ockocker¨ er et etal.: al.:Lo Low-Profilew-Profile Slot SlotAntenna AntennaElements Elements 2 C. L¨ocker et al.: Low-Profile Slot Antenna Elements 2 C. L¨ocker et al.: Low-Profile Slot Antenna Elements x2 700 x x 2 700 x = 5.0 mm 1 0 = x x 600 x0 5.0 mm 0 x 1 600 x = 4.5 mm x 0 x 0 x 0 x 2 700 x = 4.5 mm 1 0 500 Real =0 x x x0x =4.50. 0m mmm 2 1 x1 500 Real 0 = x ) 600 x0 4.0 mm 2 x 400 =

) x =3.0 mm 0 x m 0x 4.5 mm 0 0 = h 400 x 3.0 mm m 0 x1 500 Real O h 300 x = 4.0 mm x ( 0 2 O ) 300 ( e 1 400 =

c x 3.0 mm m l 0 e 200 1 n h c l 1 1 a 200 n O y y 300 1 ( d 1 1 a

y 100 e y d e w 0 0 1 1 p

y y 100 e c l w 0 0 200 p n m y y 1

I 0 1 a 1 y m y d

I 0 1 w

1 100 e w 0 0

w -100 p y y -100 Imag m stripline I 0 Imag 1 stripline 1 r -200 l w r 1 -200

l -100 -300 Imag stripline via -300 slot r 1 -2050 10 15 20 slot via l 5 10 15 20 -300 Frequency (GHz) d x via 3 Frequency (GHz) slot d x3 5 10 15 20 x 4x ground plane 4 Frequency (GHz) d x3x ground plane 5x Fig. 2. Parametric study: Input impedance with respect to via- x 5 Fig. 2. Parametric study: Input impedance with respect to via- 4 positionFig. 2. xParametric0. study: Input impedance with respect to via- ground plane x 0 (a) (b) 5 positionFig. 2. xParametric. study: Input impedance with respect to via- (a) (b) position x0. position x0. (a) (b) 1000 Fig. 1. Geometries of investigated apertures: (a) stripline-fed rect- 1000 Fig.Fig 1.. 1. Geometries of ofin investigatedvestigated apertures: apertures:(a) stripline-fed(a) stripline-fedrect- = d =2.00 mm angular slot element, (b) bow-tie slot element 8010000 d 2.00 mm rectangularangularFig. 1.slotGeometries slotelement, element,of(b)(b)inbovestigatedw-tiebow-tieslot slotapertures:element element.(a) stripline-fed rect- 800 d = 1.50 mm d = 1.50 mm d= = 2.00 mm ) d 1.00 mm angular slot element, (b) bow-tie slot element 608000 = ) d 1.00 mm m 600 Real d= = 1.50 mm h m Real d =0.75 mm h O d = 0.75 mm ( ) d 1.00 mm

= 4O 06000 d 0.50 mm ( e m 2 Rectangular Slot Element 400 Real d = 0.50 mm

c = e connected2 Rectangular by shortingSlot pinsElement or vias. Antenna dimensions are h d 0.75 mm n c O a ( n

d = 0.50 mm d 204000 a

x x x = y y e (unit:2 millimetre):Rectangular Slot=4.5,Element=1.5, 0.25, =1.5, =2.2 e 0 1 2 0 1 d 200 c p e n p

The geometry of the investigated rectangular slot element is a m

(constant distance in y-direction), r=0.5, d=1, w =0.435 and I The geometry of the investigated rectangular slot1 element is d 0200 m I e 0 ldepicted=8.25.The geometryin Fig. 1.of theTheinslotvestigatedis locatedrectangularin the upperslot elementgroundis p 1depicted in Fig. 1. The slot is located in the upper ground m -2I 000 plane and is excited by a stripline located in the centre of the -200 Imag planedepictedand isineFig.xcited1. byThea striplineslot is locatedlocatedininthetheuppercentregroundof the Imag = 2 2 substrate with r = .2 .2 In order to suppress the parasitic -40-2000 Imag substrateToplane gainand awith broadis excitedr understandingby. a. striplineIn order oflocated theto influencesuppressin thethe ofcentre viaparasitic posi-of the -4050 10 15 20 parallel plate mode, =the2upper2 and lower ground planes are 5 10 15 20 tionsparallelsubstrate and reflectorplatewithmode,r distancesthe. upper. onIn order theand resonancelotowersuppressground behaviour,theplanesparasitic theare -400 Frequency (GHz) Frequency (GHz) slotconnectedconnectedparallel was investigatedbyplatebyshortingshortingmode, bypinsthepins a parametricupperororvias.vias.andAntennaAntennalo study.wer grounddimensions Todimensions avoidplanes trans-areareare 5 10 15 20 (unit: millimetre): x0 = 4.5, x1 = 1.5, x2 = 0.25, y0 = 1.5, Frequency (GHz) formation(unit:connectedmillimetre): andby compensationshortingx0 =pins4.5,orx effects1vias.= 1.Antenna5 by, x coupling2 = 0dimensions.25, fromy0 = 1 theare.5, Fig. 3. Parametric study: Input impedance with respect to re¯ector = 2 2 = 0 5 = 1 Fig. 3. Parametric study: Input impedance with respect to re¯ector y1 (unit:= .2millimetre):2(constant distancex0 = 4.in5,y-direction),x1 = 1.5, x2 r= 0=.25.0,,5yd0 ==1.,15, sloty1 to the. stripline(constant anddistance the openin y-direction), stripline stub,r the. stripline, d , distancedistanceFig. 3. dPdarametric study: Input impedance with respect to re¯ector w1 ==0.4352 2 and l1 = 8.25. = 0 5 = 1 Fig. 3. Parametric study: Input impedance with respect to reflector wy11= 0.435. (constantand l1 =distance8.25. in y-direction), r . , δd , was removed and the slot was excited by a magnetic -gap distancedistancedd. voltagew1 = source.0.435 and Thel1 terminal= 8.25. impedance was computed by a surface/volumeToTogaingaina abroadbroad integralunderstandingunderstanding equationofofthe methodtheinfluenceinfluence forof theofviavia analy-po-po- ofofthetheslot.slot.TheTheefefficiencficiencyycurvcurvesesshoshowwthatthatthetheparallelparallelplateplate sissitionssitions ofTo planarandgainandreflectorareflector circuitsbroad distancesunderstanding withdistances additionalonontheoftheresonancethe 3-D-componentsresonanceinfluencebehabehaofviourviaviour em-po-, , modemodeof thecancanslot.bebeTheefeffectifectiefficiencvelyvelysuppressedysuppressedcurves shobybywthethatthevias.thevias.parallelAtAtthetheplatefre-fre- the slot was investigated by a parametric study. To avoid quency band edges, Fig. 4 shows a decreasing efficiency for beddedthesitionsslot inandw aas multilayeredreflectorinvestigateddistances mediumby a parametricon ofthe infiniteresonancestudy extent. behaT (oVaupelaviourvoid, quencmodeTheycan fabricatedbandbeedges,effecti singlevFig.ely slot4suppressedsho elementws a decreasingby withthe dimensionsvias.efficiencAt the asyfre-for de- transformation and compensation effects by coupling from vias placed too far as well as too close to the slot. A dis- andtransformationthe Hansenslot w, as1999inandv;estigatedVaupelcompensation etby al.a,parametric2003effects). Thus,bystudycoupling the. T layeredo afromvoid scribedviasquencplacedy inband Fig.tooedges,1farisas excitedFig.well4 shoas bywstoo aa stripline,closedecreasingto whichtheefslot.ficienc is openAydis-for at the slot to the stripline and the open stripline stub, the tance x0 smaller than approx. 0.25λ should ensure efficien- mediumthetransformationslot environmentto the andstripline iscompensation analyticallyand the openef consideredfectsstriplineby coupling bystub, its spec-fromthe itstancevias end.placedx0 Thus,smallertoo thefthanar stubasapprox.well canas be0too used.25closeλ should to optimiseto theensureslot. impedanceefAficien-dis- stripline was removed and the slot was excited by a magnetic ciestanceover090smaller%. If thethanbackapprox.reflector0 25 isshouldtoo closeensureto theefficien-slot, tralstriplinethe domainslotwtoas Green’sremothe striplinev functioned and theand andslotthe onlywasopen theexcitedstripline magneticby a magneticstub, surfacethe matching.cies ovxer 90%. TheIf the slotback wasreflector printed. λ is ontoo aclose 90 mmto the×90slot, mm δ-gap voltage source. The terminal impedance was computed theciesparallelover 90plate%. modeIf the backis excitedreflectorstrongeris too, yieldingclose to athelowerslot, currentsδstripline-gap v inoltage thewas slotsource.remo aperturevedTheand asterminalthe wellslot asimpedancew theas e electricxcitedwbyas viaacomputed currentsmagnetic Rogersthe parallel 5880plate Duroidmode substrateis excited andstronger should, haveyielding a theoreticala lower byδa-gapsurfvace/voltageolumesource.inteThegralterminalequationimpedancemethod forwasthecomputedanal- eftheficiencparallely. plate mode is excited stronger, yielding a lower (modelledby a surf asace/v volumeolume currents)integral mustequation be discretised.method for the anal- directivityefficiency. of about 5.15 dB (gain of a lossless λ/2-dipole in- ysisbyofa planarsurface/vcircuitsolumewithintegraladditionalequation3D-componentsmethod for theem-anal- efficiency. Numericalysis of planar resultscircuits on thewith inputadditional impedance3D-components and radiation effi-em- creased by 3 dB due to missing back radiation). In order to beddedysis ofin planara multilayeredcircuits withmediumadditionalof infinite3D-componentsextent (Vaupelem- The fabricated single slot element with dimensions as de- ciencybedded showin a Figs.multilayered2, 3 andmedium Figs. 4, of5,infinite respectively,extent with(Vaupel re- decreaseThe fabricated the numbersingle of degreesslot element of freedom,with dimensions the widthas ofde- the andandbeddedHansen,Hansen,in 1999),a1999),multilayered(V(Vaupelaupelmediumetetal.,al.,2003).2003).of infiniteThus,Thus,extentthethelayeredlayered(Vaupel scribedscribedTheinfinabricatedFig.Fig.11isissingleexcitedexcitedslotbybyelementaastripline,stripline,with whichdimensionswhichisisopenopenas de-atat spect to via-slot distance x0 and back reflector-slot distance feeding stripline was chosen for 50. The slot’s dimensions mediummediumand Hansen,enenvironmentvironment1999),is(Visanalyticallyaupelanalyticallyet al., considered2003).consideredThus,bybytheitsitslayeredspec-spec- itsitsscribedend.end.Thus,inThus,Fig.thethe1 isstubstubexcitedcancanbebybeusedausedstripline,totooptimiseoptimisewhichimpedanceimpedanceis open at d. The curves for the real and imaginary part of the input were carefully optimised by Agilent‘s90 mm ADS2003A90 mm Momen- traltralmediumdomaindomainenGreen’vironmentGreen’s sfunctionfunctionis analyticallyandandonlyonlyconsideredthethemagneticmagneticbysurfitssurfspec-aceace matching.matching.its end. Thus,TheTheslotslotthewwstubasasprintedprintedcan beononusedaa 90tommoptimise90 mmimpedanceRogersRogers impedance show a typical resonator behaviour. The reflec- tum (ADS), including metallisation90 (copper)mm× 90 andmm dielectric currentscurrentstral domainininthetheslotGreen’slotapertureapertures functionasaswellwellandasasonlythetheelectrictheelectricmagneticviaviacurrentscurrentssurface 58805880matching.DuroidDuroidThesubstratesubstrateslot wasandandprintedshouldshouldonhahaavveeaatheoreticaltheoretical× directidirectiRogersv-v- tor and the vias cause a shift of the input impedance to the losses. The5 directivity15 dB was estimated by2 integrating× the ra- (modelled(modelledcurrentsasinasvtheolumevolumeslot aperturecurrents)currents)asmustwellmustbeasbediscretised.thediscretised.electric via currents ityity5880ofofaboutDuroidabout .5.substrate15 dB(gain(gainandofofshouldaalosslesslosslesshaveλ/aλ/theoretical2-dipole-dipoleincreasedincreaseddirectiv- inductive regime. The frequency of the first resonance (cor- diation fields5 over15 dB all radiation directions2 and the gain is de- NumericalNumerical(modelledresultsresultsas volumeononthethecurrents)inputinputimpedanceimpedancemust be discretised.andandradiationradiationefeffi-fi- bybyity33ofdBdBaboutdueduetoto.missingmissing(gainbackbackofradiation).radiation).a lossless λ/InInorder-dipoleordertotodecreaseincreaseddecrease responding to the peak input resistance) increases and the termined3 dB by the gain transfer method. A slightly higher res- ciencciencNumericalyyshoshowwFigs.resultsFigs.2,2,on33andtheandinputFigs.Figs.impedance4,4,5,5,respectirespectiandvelyvradiationely, with, withre-efre-fi- thethebynumbernumberdueofoftodedemissinggreesgreesofbackoffreedom,freedom,radiation).thethewidthInwidthorderofoftothethedecreasefeed-feed- peakcienc resistancey show Figs. decreases2, 3 and andFigs. broadens4, 5, withrespecti decreasingvely, with viare- onancethe number frequencyof degrees was chosenof freedom,50 inΩ ADSthe towidth obtainof athe resonancefeed- spectspecttotovia-slotvia-slotdistancedistancex0x0andandbackbackreflectorreflector-slot-slotdistancedistance ingingstriplinestriplinewwasaschosenchosenforfor 50 Ω. . TheTheslot’slot’ssdimensionsdimensions and reflector distance, resulting in a reduced effective length frequency of 10 GHz. Figure 650depictsΩ simulated (ADS) and d.dspect.TheThecurvtocurvvia-slotesesforfordistancethetherealrealxand0andandimaginaryimaginaryback reflectorpartpartof-slotofthethedistanceinputinput werewereing carefullystriplinecarefullywoptimisedoptimisedas chosenbybyforAgilent‘sAgilent‘s. TheADS2003AADS2003Aslot’s dimensionsMomen-Momen- of the slot. The efficiency curves show that the parallel plate measured return loss data. The measured bandwidth (return impedanceimpedanced. The curvshoshoweswaforatypicaltypicalthe realresonatorresonatorand imaginarybehabehaviourviourpart. .TheofThethereflec-reflec-input tumtumwere(ADS),(ADS),carefullyincludingincludingoptimisedmetallisationmetallisationby Agilent‘s(copper)(copper)ADS2003AandanddielectricdielectricMomen- mode can be effectively suppressed by the vias. At the fre- loss <−10 dB) is about 5%. Simulated and measured data tortorimpedanceandandthetheviasviasshocausewcausea typicala ashiftshiftresonatorofofthetheinputinputbehaimpedanceviourimpedance. Thetoreflec-tothethe losses.losses.tum (ADS),TheThedirectidirectiincludingvityvitywmetallisationwasasestimatedestimated(copper)bybyinteintegratinggratingand dielectricthethera-ra- quency band edges, Fig. 4 shows a decreasing efficiency for are in good agreement. The efficiency curve in Fig. 7 indi- inductiinductitor andveveretheregime.gime.viasThecauseThefrequencfrequenca shiftyofyofofthethetheinputfirstfirstresonanceimpedanceresonance(corto(corthe- - diationdiationlosses.fieldsfieldsTheodirectivoerverallallvityradiationradiationwas estimateddirectionsdirectionsbyandandintethegratingthegaingaintheisisde-de-ra- vias placed too far as well as too close to the slot. A distance cates, that the efficiency does not decrease below about 80% respondingrespondinginductivetoretogime.thethepeakpeakTheinputfrequencinputresistance)resistance)y of the firstincreasesincreasesresonanceandandthe(corthe- terminedtermineddiation fieldsbybythetheovgainergainalltransfertransferradiationmethod.method.directionsAAslightlyslightlyand thehigherhighergain isres-res-de- x smaller than approx. 0.25λ should ensure efficiencies over peak0peakrespondingresistanceresistancetodecreasesthedecreasespeakandinputandbroadensbroadensresistance)withwithincreasesdecreasingdecreasingandviaviathe onanceinonancetermined the bandfrequencfrequencby ofthe operation.yygainwwasastransferchosenchoseninmethod.inADSADStoAtoobtainslightlyobtainaaresonancehigherresonanceres- 90%. If the back reflector is too close to the slot, the parallel andandpeakreflectorreflectorresistancedistance,distance,decreasesresultingresultingandinbroadensinaareducedreducedwithefeffectidecreasingfectivevelengthlengthvia frequencfrequenconanceyfrequencyofof1010GHzyGHzwas. .Fig.chosenFig.66depictsindepictsADSsimulatedsimulatedto obtain (ADS)a(ADS)resonanceandand plateand modereflector is exciteddistance, stronger,resulting yieldingin a reduced a loweref efficiency.fective length frequency of 10 GHz. Fig. 6 depicts simulated (ADS) and C.C. LC.Locker¨¨ockL¨ocker eteter al.:al.:et al.: Low-ProfileLoLow-Profilew-Profile SlotSlotSlot AntennaAntennaAntenna ElementsElementsElements 14533 C. L¨ocker et al.: Low-Profile Slot Antenna Elements 3 C. L¨ocker et al.: Low-Profile Slot Antenna Elements 3 1.0 1.0 0 0 10..09 0.9 -05-5 1.0 0 00..98 0.8 -5-10

) -10 0.9 ) -5 B 00..87 0.7 B d d

( -10 ( ) 0.8 -15

-15 y s

y -10 B ) 0.6 s 00c ..76 = s c d =x 5.0 mm s

x 05.0 mm B n 0 ( o n 0.7 o -15

d -20 e -20 e L y i ( s L

i =

0.5 =x 4.5 mm -15 0.6 c c 0.5 x 04.5 mm

= s c

y 0 i n

x 5.0 mm s i

0 n f n

0.6 r o c f s f = r -20 f x = 5.0 mm e 0 -25 x 4.0 mm u n = L -25 i 0 o u E 0.4 x 4.0 mm

= t 0 E 00..54 x 4.5 mm -20 t e

c 0 L i e i = n

0.5 e f x = 4.5 mm r c =x0 3.0 mm f 0 R i x = 3.0 mm n -25 0 R -30 f 0.3 x 4.0 mm u 0 r -30 E 00..43 f = t -25

x = 4.0 mm u

x0 2.0 mm e E 0.4 = xx0 = 302..00 mm t e 0 R Measurement 00..32 0.2 x = 3.0 mm -3-0535 Measurement 0.3 = 0 R -30 Simulation (ADS) x0 2.0 mm Simulation (ADS) 00..21 0.1 x = 2.0 mm Measurement 0 -34-5040 Measurement 0.2 -358 8.5 Sim9ulati9o.5n (AD10S) 10.5 11 11.5 12 0.01 0 8 8.5 9Simu9l.a5tion1 (0ADS1)0.5 11 11.5 12 0.1 5 10 15 20 -40 5 10 15 20 -40 Frequency (GHz) 0 8 8.5 9 F9re.5quen1c0y (G1H0.z5) 11 11.5 12 Frequency (GHz) 8 8.5 9 9.5 10 10.5 11 11.5 12 50 10Frequency (GHz1)5 20 5 10 15 20 Frequency (GHz) Frequency (GHz) Frequency (GHz) Frequency (GHz) FigFig. 6.. 6.RectangularRectangularslotslotelement,element,simulatedsimulatedandandmeasuredmeasured returnreturn FigFig. 4..P4.arametricParametricstudy:study:Ef®ciencEf®ciency withy withrespectrespecttotovia-positionvia-positionx0x0 Fig. 4. Parametric study: Efficiency with respect to via-position x0. Figloss.Fig.loss.. 6. 6. RectangularRectangularslot slotelement, element,simulated simulatedand andmeasured measuredreturn return Fig. 4. Parametric study: Ef®ciency with respect to via-position x0 Fig. 6. Rectangular slot element, simulated and measured return Fig. 4. Parametric study: Ef®ciency with respect to via-position x0 loss. loss.loss. 1.0 1.0 1.0 1.0 0.9 10..09 0.9 1.0 10.09 0.8 1.0 00..98 0.8 0.9 0.98 0.7 0.9 00..87 0.7 0.8 0.87 y 0.8 0.6 y c y = 0.6 d 2.00 mm c y 0n .6

c 0.7

= 0n .76 e 0.7 d 2.00 mm c n i

e 0.7

0.5 = n i e

c d 1.50 mm y i

i 0.5 e c

= y y 00f ..65 i c i c d 1.50 mm

f = y

i 0.65 0.6 f c c d 2.00 mm c

n = f d = 1.00 mm f 0.6 i c E 0.4 d 2.00 mm f n n e = f E

i 0.4 n d 1.00 mm f e e E 00..54 = i i

c d 1.50 mm

= e

0.5 d = 0.75 mm E 0.4 i 0.5 i

c d 1.50 mm c f

i 0.3 0.5 Measurement = i c f

f d 0.75 mm f = i 0.3 f d 1.00 mm f 0.3 = f Measurement E 0.4 d = 0.50 mm

d 1.00 mm f Simulation (ADS) E 0.4 0.3 0.2 = E 0.4 d = 0.50 mm E 0.4 Simulation (ADS) 00..32 d d0.=750 .m75m mm 0.2 Measurement 0.3 0.32 Measurement 0.1 d = 0.=50 mm 0.3 Simulation (ADS) 00..21 d 0.50 mm 0.1 Simulation (ADS) 0.02 0.21 0.2 0.01 0.1 5 10 15 20 5 10 15 20 00.10.18 8.5 9 9.5 10 10.5 11 11.5 12 8 8.5 9 9.5 10 10.5 11 11.5 12 0 0 Frequency (GHz) Frequency (GHz) 0 0 Frequency (GHz) 5 5 10 10 15 15 20 20 Frequency (GHz) 8 8 8.58.5 9 9 9.59.5 1010 101.50.5 1111 111.51.5 1212 Frequency (GHz) Fig. 5. Parametric study: Ef®ciencFrequencyy with(GHz) respect to re¯ector dis- Frequency (GHz) Fig. 7. Rectangular slot element,Frequencsimulatedy (GHz) and measured ef®- Figtance. 5. Parametric study: Ef®ciency with respect to re¯ector dis- d Figcienc. 7.y. Rectangular slot element, simulated and measured ef®- FigtanceFig. 5.d. 5.ParametricParametricstudy:study:Ef®ciencEf®ciency withy withrespectrespectto tore¯ectorre¯ectordis-dis- FigciencFig. 7.y..7.RectangularRectangularslotslotelement,element,simulatedsimulatedandandmeasuredmeasuredefef®-®- Fig.tancetance 5.d Parametricd study: Efficiency with respect to reflector dis- Fig. 7. Rectangular slot element, simulated and measured effi- tance d. ciencciency. y. measured return loss data. The measured bandwidth (return ciency. measured return loss data. The measured bandwidth (return wave travelling along the slot ground plane creates the rel- loss < 10 dB) is about 5%. Simulated and measured data wave travelling along the slot ground plane creates the rel- measuredlossmeasured< 10−returndBreturn)lossislossaboutdata.data.5%.TheTheSimulatedmeasuredmeasuredandbandwidthbandwidthmeasured(return(returndata ative large coupling. By placing a metallic plate between are in−good agreement. The efficiency curve in Fig. 7 indi- watiawvvaeevlaretratragevellingvellingcoupling.alongalongtheBytheslotplacingslotgroundgrounda metallicplaneplanecreatesplatecreatesbetweenthetherel-rel- lossarelossin< good<1010dBagreement.dB) is) aboutis aboutThe5%.5%.efSimulatedficiencSimulatedy curvandande measuredinmeasuredFig. 7 indi-data80data 4the Bow-Tieslot aperture, Slotit Elementcan be demonstrated that the mutual cou- 3cates, Mutual−that couplingthe efficienc betweeny does rectangularnot decrease belo slotw elementsabout % atitheativeslotvelarlaraperture,gegecoupling.coupling.it canBybeBydemonstratedplacingplacinga ametallicmetallicthat theplateplatemutualbetweenbetweencou- arecates,arein inthatgoodgood−theagreement.efagreement.ficiencyThedoesTheefnotficiencefficiencdecreasey curvy curvbeloe inewinFig.aboutFig.7 indi-780indi-% pling due to parallel-plate mode propagation is effectively in the band of operation. theplingtheslotslotdueaperture,aperture,to parallel-plateit itcancanbebedemonstratedmodedemonstratedpropagationthatthatthetheismutualefmutualfecticou-velycou- cates,in cates,thethatbandthattheoftheefoperation.ficiencefficiency doesy doesnotnotdecreasedecreasebelobelow waboutabout8080%% Sincesuppressed the bandwidthby the vias. of the rectangular slot element is rela- Mutual coupling between antenna elements is an important plingsuppressedplingdueduetobytoparallel-platetheparallel-platevias. modemodepropagationpropagationisisefeffectifectivelyvely in thein thebandbandof ofoperation.operation. tively small, we also investigated bow-tie like slot elements suppressedsuppressedbybythethevias.vias. criterion in conformal array antenna applications. We in- (Kraus, 1988). The slot was widened at its ends and the vestigated3 Mutual collinearcoupling andbetween broadsiderectangular slot arrangements.slot elements Three 4 Bow-Tie Slot Element 3 Mutual coupling between rectangular slot elements 4lengthBow-T wasie increasedSlot Element to about 0.7λ0 for a centre frequency slots were considered in each configuration. The single slot 3 3MutualMutualMutualcouplingcouplingcouplingbetweenbetweenbetweenantennarectangularrectangularelementsslotslotiselementsanelementsimportant 4ofSince4 Bo 10Bow-T GHz,thew-TiebandwidthieSlot seeSlotElement Fig.Elementof1.the Inrectangular order toslot reduceelement edgeis diffrac-rela- dimensions are taken from Fig. 1a. In the collinear con- Mutualcriterioncouplingin conformalbetweenarrayantennaantennaelementsapplications.is an importantWe in- Sincetiontively effects,thesmall,bandwidth thewe also sideofin lengththevestigatedrectangular of thebow-tie substrateslotlikelemente slot waselements increasedis rela- figuration,MutualcriterionMutualvestigatedcouplingincoupling centre-centreconformalcollinearbetweenbetweenandarray distancesantennabroadsideantennaantennaelements withslotelementsapplications.arrangements. respectis isanan toimportantimportant theWeThree firstin- Sincetitov(Kraus,Sinceely 100thesmall, mm.the1988).bandwidthbandwidthwe In thisalsoThe structureofinslotofvtheestigatedthewrectangularasrectangular thewidenedbo viasw-tie positionsslotatslotlikitselementeelementslotends areelements notandisisrela- paral-rela-the slot are 22.5 mm and 52.5 mm. In the broadside slot con- criterionvestigatedcriterionslots wereincollinearinconformalconsideredconformalandarraybroadsideinarrayeachantennaantennaconfiguration.slotapplications.arrangements.applications.The singleWThreeWe ein-slotin- ti(Kraus,lelvlengthtielyvely tosmall, thesmall,w1988).as apertureincreasedwewealsoThealso axis.inslottoinvestigatedvaboutestigatedw Theas widened0 vias.7boλbo0w-tie positionsw-tieforatalikitslikcentreeeendsslot wereslotfrequencelementsandelements carefullythey figuration,vslotsestigatedvdimensionsestigatedwere collinear theconsideredcollinearare slotstakand couldeninandfrombroadsideeachbroadside beFig. arrangedconfiguration.1(a).slotslotarrangements.In closerarrangements.theThecollinear to eachsingleThreeconfig- other,Threeslot (Kraus,lengthoptimisedof(Kraus,10wGHz1988).as1988).increased to, see achieveTheTheFig.slottoslot1. theaboutwInwas designasorder0widened.widened7λ0 requirements.toforreduceataatcentreitsitsendsedgeendsfrequenc Theanddifandfrac-the bow-they resulting in separations of 15 mm and 45 mm. The mutual 10 GHz 0077 slotsdimensionsslotsuration,werewereconsideredcentre-centreareconsideredtaken fromin indistanceseacheachFig.configuration.1(a).configuration.withInrespectthe collinearThetoThesinglethesingleconfig-firstslotslotslot lengthoftietionlength dimensionsefwfects,aswas,increasedseeincreasedthe areFig.side (units:to1.lengthtoaboutaboutIn millimetre):orderof.the.λ0λtosubstrate0forreducefora acentrexcentre0=4.25,wedgeasfrequencincreasedfrequencdifx1frac-=4.45,yy coupling22 was5 mm determined52 5 bymm transmission measurements be- 1010010GHzGHzmm dimensionsuration,dimensionsare centre-centre. arearetakandtakenenfromdistances.fromFig.Fig.. 1(a).withIn1(a).theInrespectbroadsideInthethecollineartocollineartheslotfirstconfig-configu-config-slot oftionxtoof2=4.65,effects,x,3.=0.25,the,seeInseesidethisFig.Fig.x4structurelength=1.6,1.1. InxInof5order=2.5,theordertheviassubstrateyto0to=1.7,reducepositionsreducey1w=2.2,edgeasedgeareincreasedwnotdifdif1=0.425frac-parfrac-- tweenareuration,22 neighbouring5 mmcentre-centreand 52 slots5 mmdistances with. In athe HP8510withbroadsiderespect networktoslottheconfigu- analyser.first slot totion100efmmfects,. Inthethissidestructurelengththeof theviassubstratepositionswareas notincreasedpar- uration,ration,. centre-centrethe slots could.distancesbe arrangedwith respectcloser totoeachthe firstotherslot, re- tionandallelefl1tofects,=10.5.the aperturethe Asside for theaxis.length rectangularTheof viasthe substrate slotpositions element,wwereas theincreasedcarefully substrate Theareration,aresulting22 remaining.225the.mm5inslotsmmseparationsand thirdandcould5252. port5be.mm5ofmm wasarranged.15In. terminatedmmInthethecloserandbroadsidebroadside45to bymmeach aslot. 50slotTheotherconfigu-configu-match.mutual, re- toallelhasoptimisedto100100to anmmthe permittivitymm.apertureto.InachieInthisthisvstructureaxis.e ofstructurethe 2.2.Thedesign Figuretheviastherequirements.viasviaspositions10positionspositionsdepictswereTheare thearecarefullynot measuredbonotw-tieparpar- - 15 mm 45 mm Figuresration,sultingration,couplingthe8inandtheseparationsslotswslotsas9 showdeterminedcouldcould simulatedofbebearrangedbyarrangedtransmission andandcloser measuredclosertomeasurementstoeach. scatteringeachTheotherothermutual, re- co-,be-re- alleloptimisedanddimensionsallelto simulatedtothethetoapertureareachieaperture return(units:veaxis.theaxis. loss.millimetre):designTheThe Theviasviasrequirements. 10positionsxpositions dB0 = bandwidth4.25werewere,Thex1carefullybo=carefully amountsw-tie4.45, 15 mm 45 mm = 4 65 = 0 25 = 1 6 = =4 252 5 ==4 1457 efficientssultingcouplingsultingtweeninwneighbouringSinseparationsijas,separationsdetermined which describeofslotsof15bywithmmtransmission theaand couplingHP8510and45 mmmeasurementsnetw behaviour.. .TheorkTheanalysermutualmutualbe- For. optimiseddimensionstoxoptimised2 0.8 GHz. toareto,achie andxachie3(units: isvev higherethe. millimetre):the,designxdesign than4 requirements. forrequirements.. x, the0x5 rectangular. .,The,xThe1y0 bobo slotw-tie.w-tie. ele-,, 50 Ω == 42 652 ==0 4250 25 == 11065 = 42255 == 14 745 collinearcouplingtweencouplingTheneighbouringremainingw alignmentaswasdetermineddeterminedthird theslotsport radiationbywithbywtransmissionastransmissionaterminatedHP8510 null inmeasurements thenetwmeasurementsby correspondingorka analysermatch.be-be-. dimensionsxment.2ydimensions1 . The. , ,warex measured1are3 (units:(units:. . millimetre): antennaand,millimetre):x4l1 efficiency. .,x.x0 As50= for4 is.25.the. about,,,xrectangularyx101= 754−.4580%,.. ,, , 50 Ω =2==24246565 = 30=425=0 02525 4==1011656 5==22552 20==1177 directiontweenThetweenFigs.remainingneighbouring8 causesneighbouringand 9third asho smallerwslotsportslotssimulated couplingwwithaswithterminateda HP8510anda comparedHP8510measuredbynetwnetwa toorkscattering theorkanalyser couplinganalysermatch.co-. . xysee12slotx Fig.element,. .,11.w,1,x, in3thex the.substrate band. and. , ofx, l4hasx1 operation.an permitti...,.,xAs5x forvitythe.of.,rectangular,y.0y. Fig. .10., , forTheFigs.Theef broadsideremainingficients8remainingand 9S configuration.ijshothird, whichthirdw simulatedportportdescribewasw Inasterminatedand thisterminatedthemeasured case,couplingby thebya radiatedscatteringbeha50a 50ΩviourΩmatch. spacematch..co-For yslot1depictsy=1element,=2.2.,the2w, 1wmeasuredthe1==0substrate.0425.425andandhaslsimulated1l1=an=10permitti10.5..5returnAs. Asvityforforloss.theofthe2rectangularThe.2rectangular. Fig.10 dB10 0 8 GHz 2 2 waveFigs.efficientsFigs.collinear travelling8 and8 Sandijalignment9, sho9which alongshow wsimulateddescribe thethesimulated slotradiation groundtheandandcouplingmeasurednullmeasured planein thebeha createsscatteringcorrespondingscatteringviour the. Fco- rel-orco- slotdepictsbandwidthslotelement,element,the measuredamountsthethesubstratesubstratetoand .simulatedhashasananpermittiandpermittireturnis vityhighervityloss.ofof2thanThe.2.. .Fig.10forFig.dBthe1010 0 8 GHz 10 dB ativeefcollinearficientsefdirectionficients largealignmentS coupling.ijcausesS,ijwhich, whicha smallerthedescribe Bydescriberadiation placingcouplingthethenullcoupling acouplingcompared metallicin thebehacorresponding platebehatoviourtheviour betweencoupling. .ForFor depictsbandwidth5rectangulardepicts Conclusionthetheamountsmeasuredslotmeasuredelement.toandand. Thesimulatedsimulatedmeasuredandreturnisreturnantennahigherloss.loss.thanefTheTheficiencfor10 theydBis collineardirectioncollinearfor broadsidealignmentcausesalignmentaconfiguration.smallerthetheradiationcouplingradiationIn nullthiscomparednullcase,ininthethetocorrespondingradiatedthecorrespondingcouplingspace bandwidthrectangularaboutbandwidth75 amountsslot80amounts%,element.seetoFig.to0The08.11,8GHzGHzmeasuredin theandandbandisisantennahigherofhigheroperation.thanefthanficiencforforytheisthe the slot aperture, it can be demonstrated that the mutual cou- − . plingdirectionfordirectionbroadside duecauses tocauses parallel-plateconfiguration.a smallera smallercoupling modecouplingIn this propagationcomparedcase,comparedthetoradiatedto isthethe effectivelycouplingcouplingspace rectangularaboutRadiatingrectangular75 slots80slotslot%,element. inseeelement. aFig. metallicThe11,Theinmeasured groundmeasuredthe band planeantennaofantennaoperation. wereefef investigated.ficiencficiencyyisis 75− 80 suppressedforforbroadsidebroadside byconfiguration. theconfiguration. vias. InInthisthiscase,case,thetheradiatedradiatedspacespace aboutTheabout upper75 80 ground%,%,seesee planeFig.Fig.11, was11,in connectedinthethebandbandof withofoperation.operation. a back reflector −− 146444 C.C. C.C.LL L¨¨ockockocker¨ockerereretet etetal.:al.: al.:al.:LoLo Low-ProfileLow-Profilew-Profilew-ProfileSlotSlot SlotSlotAntennaAntenna AntennaElementsElements Elements 4 C. L¨ocker et al.: Low-Profile Slot Antenna Elements

--220-020 00 0 -20 -5 0 --3300 -5-5 -30 -5 ) ) ) -30 -10 B ) -10 B ) -10 B ) d )

-40 B d B ( -40 d

-40 B (

-10 d ( B )

d

( d g (

-15 d ( g -40 B

-15 g (

-15 n s

n d s i n s i l ( -50 s i g l s -50 -15 l -50 s p o n s p o -20 p o i L u -20 l s -20

-50 L u

L u

o

p o n o

o -20 -60 n r n C L -6u 0 r C

-60

r -25 C u l -25 o

u -25 l t

S Measurement n u l 21 a -60 S Measurement t t 2S1 Measurement r C e a 21 a

e u -25 e u l u t R

u -70 S Measurement t t S S21imulation (ADS) R -30 a

t -70

21 R u -70 S Simulation (ADS) -30 2S1 Simulation (ADS) e -30 u 21 u u Measurement t M -70 R MeMaesausruermeemnetnt M S S Me Saismuurelamtieonnt (ADS) -30

M 21

u 31 S3S1 M eMaesausruermeemnetnt -35 Simulation (ADS) --8800 31 -35-35 SiSmMiumelaaustluiaortnieo m(nA e(DnAStD)S) M -80 S31S S31i mMuelaastiuorne m(AeDnSt ) -80 S3S1 S iSmiumlautliaotnio (nA (DASD)S) -35 Simulation (ADS) 31 --4400 --9900 S31 Simulation (ADS) -40 -90 -4880 88..55 99 99.5.5 1100 1100.5.5 1111 1111.5.5 1122 -9880 88..55 99 99..55 110 10..5 11 11.5 12 8 8.5 9 9.5 10 10.5 11 11.5 12 8 8.5 9 9.5 10 10.5 11 11.5 12 8 8.5 9 9.5 10 10.5 11 11.5 12 8 8.5 9Freq9u.e5ncy 1(G0 Hz1)0.5 11 11.5 12 Frequency (GHz) FrFerqeuqeunecnyc (yG (HGzH)z) FrFerqeuqeunecnyc (yG (HGzH)z) Frequency (GHz) Frequency (GHz) Fig. 10. Bow-Tie element, measured and simulated return loss Fig.FigFigFig.. 8.8..8.8.MutualMutualMutualMutual coupling,coupling,coupling,collinear collinearcollinearslotslot slotarrangement. arrangement.arrangement. FigFig. 10.. 10.BoBow-Tw-Tieieelement,element,measuredmeasuredandandsimulatedsimulatedreturnreturnlossloss Fig. 8. Mutual coupling, collinear slot arrangement. Fig.Fig. 10.10.Bow-TieBow-Tie element,element, measuredand simulated return returnloss loss.

11..00 --1100 1.0 -10 0.91.0 -10 0.90.9 --1155 -15 0.80.9 ) -15 0.80.8 )

) -20 B -20 0.8 B

d -20 B 0.7 ) ( d

0.7 d 0.7 ( -20 B (

g

-25 y

d 0.7 g

n 0.6 ( y g -25 c y i -25

n l 0.6 c n n 0.6 i c g i l -25 y p e n l n i

n -30 0.6 c p u e

i 0.5 p e c i

-l 30 i n i

u -30 o 0.5 u

c 0.5 f p c e i f o i i C o

-30 f u f 0.5 f

-35 E 0.c 4 l C f C i

o

-35 E a 0f .4 l S Measurement -35 E 0.4 l 21 f C a u S Measurement a 2S1 Measurement Measurement t -35 E 0.4 l 21 0.3 u -40 u Measurement t u a S Simulation (ADS) Measurement t -40-40 21S21 Measurement 0.30.3 Simulation (ADS) u

u S Simulation (ADS) u M 2S1 Simulation (ADS) Measurement t -40 21 0.20.3 SimSiumlautliaotnio (nA (DASD)S) M

u S Measurement M -45 31S21 Simulation (ADS) 0.2 S3S1 M eMaesausruermeemnetnt 0.2 Simulation (ADS) -M 45-45 31 S S Si mMuelaastiuorne m(AeDnSt ) 0.10.2 -45 S 31 S31imulation (ADS) 0.1 -50 3S1 31 Simulation (ADS) 0.1 -50-508 8.5 9 9.5 S3110 Sim1u0l.a5tion1 1(AD1S1.)5 12 0 0.1 -5808 8.58.5 9 9 9.59.5 1010 101.50.5 1111111.51.5 1212 0 08 8.5 9 9.5 10 10.5 11 11.5 12 8 8.5 F9requ9e.5ncy (1G0Hz)10.5 11 11.5 12 08 8 8.58.5 9 9 9.59.5 1010 101.50.5 1111 111.51.5 1212 FrFerqeuqeunecnyc (yG (HGzH)z) 8 8.5 9 Freq9u.5ency1 (0GHz1)0.5 11 11.5 12 Frequency (GHz) FrFeqreuqeunecnyc (yG (HGzH)z) Frequency (GHz) Fig. 9. Mutual coupling, broadside slot arrangement. FigFig. 9.. 9.MutualMutualcoupling,coupling,broadsidebroadsideslotslotarrangement.arrangement. Fig. 11. Bow-Tie element, measured and simulated ef®ciency Fig. 9. Mutual coupling, broadside slot arrangement. FigFig. 11.. 11.BoBow-Tw-Tieieelement,element,measuredmeasuredandandsimulatedsimulatedefef®cienc®ciency y Fig. 9. Mutual coupling, broadside slot arrangement. Fig.Fig. 11.11.Bow-TieBow-Tie element,element, measuredand simulated ef efficiency.®ciency

5 Conclusion Omiya, M., Hikage, T., Ohno, N., Horiguchi, K., Itho, K.: Design of 55 ConclusionConclusion Omiya,Omiya,M.,M.,Hikage,Hikage,T.,T.,Ohno,Ohno,N.,N.,Horiguchi,Horiguchi,K.,K.,Itho,Itho,K.:K.:DesignDesignofof 5 Conclusion Omiya,Cavity-BackM., Hikage,ed Slot TAntennas., Ohno, N.,UsingHoriguchi,the Finite-DifK., Itho,ferenceK.: DesignTime- of by shorting pins. It is shown by simulation and measure- Brachat,CaCavity-Backvity-Back P. andededSlot Baracco,SlotAntennasAntennas J. M.:UsingUsing Dual-PolarizationthetheFinite-DifFinite-Differenceference Slot-CoupledTime-Time- Radiating slots in a metallic ground plane were investigated. DomainCavity-BackTechnique,ed SlotIEEEAntennasTransactionsUsing theon Finite-DifAntennas ferenceand Propa-Time- ment,RadiatingRadiating thatslots theslots parasiticinina ametallicmetallic TEMgroundground parallelplaneplane platewerewere modeininvestigated. canvestigated. be ef- DomainPrintedDomainT Antennasechnique,Technique, FedIEEEIEEE byTransactionsT Stripline,ransactions IEEEononAntennasAntennas TransactionsandandPropa-Propa- on An- TheRadiatingupper groundslots inplanea metallicwas connectedground planewithwerea backinvestigated.reflec- gation,46,Domain1853-1858,Technique, 1998IEEE Transactions on Antennas and Propa- fectivelyTheTheupperupper suppressedgroundgroundplaneplane by shortingwwasasconnectedconnected pins. However,withwitha aback theback slotreflec-reflec- de- gation,46,tennasgation,46, and1853-1858,1853-1858, Propagation,19981998 43, 738–742, 1995. torThebyuppershortinggroundpins.planeIt iswshoas wnconnectedby simulationwith a backand mea-reflec- Sommers,gation,46,D. J.:1853-1858,Slot Array1998Employing Photoetched Tri-Plate signtor by mustshorting pass apins. carefulIt optimisationis shown by process,simulation becauseand mea- of the Sommers,Galejs,Sommers, J.:D. AdmittanceD.J.:J.: SlotSlot ofArrayArray a RectangularEmploEmployingying SlotPhotoetchedPhotoetched Which is BackedTri-PlateTri-Plate by a tortor byby shortingshorting pins.pins. ItIt isis shoshownwn byby simulationsimulation andand mea-mea- Sommers,TransmissionD. Lines,J.: SlotIEEEArrayTransactionsEmployingon PhotoetchedMicrowave TheoryTri-Plate surement, that the parasitic TEM parallel plate mode can be TRectangularransmissionTransmission Cavity,Lines,Lines,IEEE IEEEIEEET TransactionsransactionsTransactionson ononMicro AntennasMicrowwavaev andeTheoryTheory Propa- strongsurement,surement, influencethatthatthethe ofparasitic theparasitic viasTEM onTEM theparallel resonanceparallelplateplate andmodemode broadbandcancanbebe andTransmissionTechniques, Lines,3, 157-162,IEEE1955Transactions on Microwave Theory efsurement,fectively suppressedthat the parasiticby shortingTEM parallelpins. Hoplatewevermode, thecanslotbe andgation,andTechniques,Techniques, 119–126,3,3, 1963.157-162,157-162,19551955 behaviour.efeffectifectivelyvelysuppressed Asuppressed stripline-fedbybyshortingshorting rectangularpins.pins. slotHoHowe elementweverver, ,thethe withslotslot a Brachat,andPT.,echniques,Baracco, J.3,M.:157-162,Dual-Polarization1955 Slot-Coupled Printed designeffectimustvely passsuppresseda carefulbyoptimisationshorting pins.process,Howebecausever, the ofslot Brachat,James,Brachat, J.P.,P R.,.,Baracco,Baracco, Hall, P.J. S.,J.M.:M.: andDual-PolarizationDual-Polarization Wood, C.: MicrostripSlot-CoupledSlot-Coupled AntennaPrintedPrinted Theory 10designdesign dB-bandwidthmustmustpasspassa ofacareful 5%careful atoptimisation centreoptimisation frequencyprocess,process, of 10becausebecause GHz andofof Brachat,AntennasP.,FedBaracco,by Stripline,J. M.: Dual-PolarizationIEEE TransactionsSlot-Coupledon AntennasPrintedand thedesignstrongmustinfluencepass aofcarefulthe viasoptimisationon the resonanceprocess,andbecausebroad-of AntennasandAntennas Design,FedFed PeterbybyStripline,Stripline, PeregrinusIEEEIEEE Ltd.,TransactionsT 1981.ransactionsononAntennasAntennasandand thethestrongstronginfluenceinfluenceofofthetheviasviasononthetheresonanceresonanceandandbroad-broad- Propagation,Antennas Fed43, by738-742,Stripline,1995IEEE Transactions on Antennas and anbandthe efficiencybehastrongviourinfluence of. moreA stripline-fedof thanthe 80%viasrectangular wason the fabricated.resonanceslot element Studiesand broad-with on Kraus,Propagation, J. D.: Antennas,43, 738-742, McGraw1995 Hill, 1988. band behaviour. A stripline-fed rectangular slot element with BhattacharyyPropagation,Propagation,, A., 43,F43,ordham,738-742,738-742,O.,19951995Liu, Y.: Analysis of Stripline-Fed couplingband10 dBbeha effectsviour. betweenA stripline-fed5 alignedrectangular slots showslot thatelement10 collinearGHzwith BhattacharyyOmiya, M.,, Hikage,A., Fordham, T., Ohno,O., N.,Liu, Horiguchi,Y.: Analysis K.,of andStripline-Fed Itho, K.: De- a 10banddBbeha-bandwidthviour. A stripline-fedof 5% at centrerectangularfrequencsloty ofelement10 GHzwith BhattacharyyBhattacharyySlot-Coupled,, A.,PA.,atchFFordham,ordham,AntennasO.,O.,withLiu,VYias.:.: forAnalysisParallel-Plateof Stripline-FedMode slotsa a 1010 aredBdB non-critical.-bandwidth-bandwidthof Forof %5 bandwidth5%atatcentre80centre enhancement,frequencfrequencyyofof a10 bow-tie10 GHzGHz Slot-Coupled Patch Antennas with Vias for Parallel-Plate Mode anda an efficienc-bandwidthy of moreof than% at centre% wasfrequencfabricated.y ofStudies Suppression,signSlot-CoupledSlot-Coupled of Cavity-BackedIEEEPPatchatchTransactionsAntennasAntennas Slot Antennaswithon AntennasVias Usingias forfor thePandarallel-Plate Finite-DifferencePropagation,Mode likeandand slotananef elementefficiencficiency withyofofmoremore 10 dBthanthan bandwidth8080%%wwasas offabricated.fabricated. 8% and radiationStudiesStudies Suppression, IEEE Transactions on Antennas and Propagation, onandcouplingan efficienceffectsy betweenof more thanaligned80%slotswasshofabricated.w that collinearStudies 46,Time-DomainSuppression,Suppression,538-545, 1998IEEEIEEE Technique,TTransactionsransactions IEEEon Transactionson Antennas and on AntennasPropagation, and ononcouplingcouplingefeffectsfectsbetweenbetweenalignedalignedslotsslotsshoshowwthatthatcollinearcollinear 46, 538-545, 1998 efficiencyslotson couplingare non-critical. of closeeffects to 80%betweenFor bandwidth wasaligned presented.enhancement,slots show thata bocollinearw-tie Vaupel,Propagation,46,46,T538-545,538-545,., Hansen, 46,19981998V.: 1853–1858,Electrodynamic 1998.Analysis of Combined Mi- slotsslotsarearenon-critical.non-critical.ForForbandwidthbandwidthenhancement,enhancement,a abobow-tiew-tie Vaupel, T., Hansen, V.: Electrodynamic Analysis of Combined Mi- likslotse slotareelementnon-critical.with 10FordBbandwidthbandwidthenhancement,of 8% and radiationa bow-tie Sommers,VVcrostripaupel,aupel,TTand.,., D.Hansen,Hansen,Coplanar/Slotline J.: SlotVV.:.: ElectrodynamicElectrodynamic ArrayStructures EmployingAnalysiswith Photoetched3-Dof ComponentsCombined Tri-PlateMi- like slot element with 1010dBdBbandwidth of 8%8 and radiation crostrip and Coplanar/Slotline Structures with 3-D Components efliklikficienceeslotslotyelementelementof closewithtowith80%10wdBasbandwidthpresented.bandwidthofof 8%%andandradiationradiation BasedTransmissioncrostripcrostripon aandandSurfCoplanar/SlotlineCoplanar/Slotline Lines,ace/Volume IEEEInte TransactionsgralStructuresEquationwith onApproach, Microwave3-D ComponentsIEEE Theory 8080 Based on a Surface/Volume Integral Equation Approach, IEEE Referencesefefficiencficiencyyofofcloseclosetoto 80%%wwasaspresented.presented. TransactionsandBasedBased Techniques,onon aaonSurfSurfMicroace/V 3,ace/V 157–162,wolumeolumeave TheoryInteInte 1955.gralgralandEquationTechniques,Approach,47, 1788-IEEE efficiency of close to % was presented. Transactions on Microwave Theory and Techniques, 47, 1788- Vaupel,1800,TTransactionsransactions1999 T. and Hansen,onon MicroMicro V.:wwaa Electrodynamicvvee TheoryTheory andand T Analysisechniques,echniques, of47, Combined1788- Bhattacharyya, A., Fordham, O., and Liu, Y.: Analysis of Stripline- 1800, 1999 Vaupel,Microstrip1800,1800,T.,19991999Hansen, andV Coplanar/Slotline., Schfer, F.: Radiation StructuresEf®cienc withy Analysis 3-D Compo-of Fed Slot-Coupled Patch Antennas with Vias for Parallel-Plate Vaupel, T., Hansen, V., Schfer, F.: Radiation Ef®ciency Analysis of References VVSubmm-Waupel,aupel,nentsTT Based.,.,Hansen,aHansen,ve Recei on aVVv Surface/Volume.,.,ersSchferSchferBased,, FF.:.:onRadiationa Modi®ed IntegralEf®cienc EquationSpectraly AnalysisDomain Approach,of ReferReferModeencesences Suppression, IEEE Transactions on Antennas and Prop- Submm-Wave Receivers Based on a Modi®ed Spectral Domain References Technique,IEEESubmm-WSubmm-W TransactionsRadioaavvee ReceiReceiSience, onvversers Microwave4,BasedBased13-1-13-2,onon aa TheoryModi®ed2003 andSpectral Techniques,Domain 47, Technique, Radio Sience, 4, 13-1-13-2, 2003 James,agation,J. R., 46,Hall, 538–545,P. S., W 1998.ood, C.: Microstrip Antenna Theory and Kraus,1788–1800,TTechnique,echnique,J. D.: Antennas,RadioRadio 1999.Sience,Sience,McGraw4,4,Hill,13-1-13-2,13-1-13-2,1988 2003 James,James,J.J.R.,R.,Hall,Hall,P.PS.,. S.,WWood,ood,C.:C.:MicrostripMicrostripAntennaAntennaTheoryTheoryandand Biebl,James,Design, E. M.J.PeterR., andHall,Pere Friedsam,Pgrinus. S., W G.Ltd.,ood, L.:1981C.: Cavity-BackedMicrostrip Antenna ApertureTheory Anten-and Kraus,Vaupel,Kraus,J.J.D.: T.,D.: Hansen,Antennas,Antennas, V.,McGra andMcGra SchwwHill,afer,¨ Hill,1988 F.:1988 Radiation Efficiency Anal- Design, Peter Peregrinus Ltd., 1981 Kraus, J. D.: Antennas, McGraw Hill, 1988 Galejs,nasDesign,Design, withJ.: Admittance DielectricPeterPeterPerePere andgrinusofgrinusa MagneticRectangularLtd.,Ltd.,19811981 Overlay,Slot Which IEEE Transactionsis Backed by ona Galejs, J.: Admittance of a Rectangular Slot Which is Backed by a ysis of Submm-Wave Receivers Based on a Modified Spectral Galejs,Galejs,AntennasRectangularJ.:J.:Admittance andAdmittanceCa Propagation,vity, IEEEofofaaRectangularTRectangular 43,ransactions 1226–1232,SlotSloton AntennasWhich 1995.WhichisisandBackBackPropa-ededbybyaa Rectangular Cavity, IEEE Transactions on Antennas and Propa- Domain Technique, Radio Sience, 4, 13-1-13-2, 2003. gation,RectangularRectangular119-126,CaCavity1962vity,,IEEEIEEETTransactionsransactionsononAntennasAntennasandandPropa-Propa- gation, 119-126, 1962 Biebl,gation,gation,E. M.,119-126,119-126,Friedsam,19621962G. L.: Cavity-Backed Aperture Antennas Biebl, E. M., Friedsam, G. L.: Cavity-Backed Aperture Antennas Biebl,Biebl,with DielectricE.E.M.,M.,Friedsam,Friedsam,and MagneticG.G.L.:L.:OvCaCaerlayvity-Backvity-Back, IEEEededTransactionsApertureAperture AntennasAntennason An- with Dielectric and Magnetic Overlay, IEEE Transactions on An- tennaswithwithDielectricDielectricand Propagation,andandMagneticMagnetic43, 1226-1232,OvOverlayerlay,,IEEE1995IEEETTransactionsransactionsononAn-An- tennas and Propagation, 43, 1226-1232, 1995 tennastennasandandPropagation,Propagation,43,43,1226-1232,1226-1232,19951995