Açık kuplajlı mikroşerit yama antenler için yapay sinir ağ modeli

'LFOHhQLYHUVLWHVL0KHQGLVOLN)DNOWHVL PKHQGLVOLNGHUJLVL

*_{<D]ÕúPDODUÕQ\DSÕODFD÷Õ\D]DUøVD$7$ùLVDBDWDV@dicle.edu.tr; Tel: (412) 248 80 30 (3739)}

Özet

7HNQRORMLQLQ JHOLúPHVLQH SDUDOHO RODUDN NDEORVX] LOHWLúLPLQ LOJL J|UPHVL VRQ \ÕOODUGD DQWHQ WHNQRORMLVLQLQ KÕ]OÕLOHUOHPHVLQHRODQDNVD÷ODPÕúWÕU.DEORVX]LOHWLúLPDUDoODUÕQGDQELULGHPRELOX\JXODPDODUGDYHX]D\ DUDoODUÕQGD NXOODQÕODQ 0LNURúHULW <DPD $QWHQOHUGLU 0<$ .LúLVHO WDúÕQDELOLU FLKD]ODUÕQ \D\JÕQODúPDVÕ 0<$¶QÕQ|QHPLQLGDKDGDDUWWÕUPÕúWÕU

%X oDOÕúPDGD *+] LOH *+] DUDVÕQGDNL IUHNDQV GH÷HUOHUL LoLQ <DSD\ 6LQLU $÷ <6$ PRGHOLQH GD\DOÕ $oÕNOÕN .XSODMOÕ 0LNURúHULW <DPD $QWHQ $.0<$ WDVDUÕPÕ \DSÕOPÕúWÕU $.0<$ODU PLNURúHULW KDW LOH EHVOHQLUOHUYHNHQGLVÕQÕIÕQGDNL0<$WLSOHULLoHULVLQGHHQ\NVHNEDQWJHQLúOL÷LQHVDKLSWLUOHU

*HRPHWULN \DSÕODUÕ IDUNOÕ  DGHW $.0<$¶QÕQ VLPODV\RQX )LQLWH (OHPHQW 0HWKRG  )(0  \|QWHPLQL NXOODQDQER\XWOXWDPGDOJD(OHNWURPDQ\HWLN$ODQ6LPODW|U($6\D]ÕOÕPÕLOH\DSÕOPÕúYHKHUELUDQWHQ LoLQUH]RQDQVIUHNDQVGH÷HULKHVDSODQPÕúWÕU

Levenberg-0DUTXDUGW /0 |÷UHQPH DOJRULWPDVÕ WHPHOLQGH JHOLúWLULOHQ <6$ PRGHOL ($6 LOH UHWLOHQ |UQHNOHU LOH H÷LWLOPLú H÷LWLP VUHVLQFH J|UPHGL÷L WHVW YHUL VHWL NXOODQÕODUDN GR÷UXOX÷X |OoOPúWU *HOLúWLULOHQ<6$PRGHOLQLQEDúDUÕPÕQÕQ|OoOPHVLQGHNDWoDSUD]ODPDGR÷UXOXN\|QWHPLNXOODQÕOPÕúYH  WHVW KDWD RUDQÕ WHVSLW HGLOPLúWLU =DPDQ YHULPOLOL÷L DoÕVÕQGDQ EDNÕOGÕ÷ÕQGD |QHULOHQ \|QWHPLQ ($6 \D]ÕOÕPÕQD J|UH HQ D]  NDW GDKD KÕ]OÕ oDOÕúWÕ÷Õ WHVSLW HGLOPLúWLU gQHULOHQ <6$ PRGHOLQLQ $.0<$ODUÕQ rezRQDQVIUHNDQVÕQÕQEHOLUOHQPHVLQGHHWNLQYHYHULPOLELU\|QWHPRODFD÷ÕGúQOPHNWHGLU

Anahtar Kelimeler: Yapay sinir a÷ODUÕ $oÕNOÕN NXSODMOÕ PLNURúHULW \DPD DQWHQOHU 5H]RQDQV

IUHNDQVÕ$QWHQWDVDUÕPÕ.

$oÕNOÕNNXSODMOÕPLNURúHULW\DPDDQWHQOHULoLQ

\DSD\VLQLUD÷PRGHOL

øVD$7$ù*1_{, M. Bahattin KURT}1_0XVD$7$ù2

1_{DLFOHhQLYHUVLWHVL(OHNWULN(OHNWURQLN0KHQGLVOL÷L%|OP'L\DUEDNÕU} 2_{6LLUW hQLYHUVLWHVL %LOJLVD\DU 0KHQGLVOL÷L %|OP 6LLUW}

mühendislikdergisi

Cilt: 3-9

Dicle Üniversitesi Mühendislik Fakültesi

Neural network model for aperture

coupled microstrip antennas

Extended abstract

Recently, in parallel with the development of technology, attention on wireless comunication provides rapid progress in antenna technology. One of the wireless communication tools is microstrip patch antennas (MPA) that is used in mobile applications and space vehicles in general. Heavy demand on personal portable device further increases the significance of MPA.

MPAs, due to its small size and sharing the same dielectric layer with circuit members are harmonized easily with integrated circuits. Neverthless, narrow bandwidth, high loss in feeder circuit, low cross polarization and low power control capacity are the main weaknesses of the basic MPAs. Previous researches and studies show that, most of the aforementioned disadvantages might be removed or reduced by means of making various extensions on the basic MPA units. MPAs are utilized efficently on various system applications including wireless and satellite communication, biomedical irradiator, environmental instruments and remote sensing systems. Number of these applications will be raised with the parallel in development of technology. Defining resonant frequency is crucial issue since MPAs operate at narrower bandwidth than other antennas. Parameters that influence the resonant frequency of MPA as follows; thickness of used dielectric material, dielectric constant, size of ground surface thickness and width of conductive patch. In this study, Aperture Coupled Microstrip Antennas (ACMA) extended from MPA class are investigated. Note that, ACMA is fed microstrip line which has relatively higher bandwidth compared to other microstrip patch antennas.

ACMA prototype is used throughout the study. It is prepared via High Frequency Structure Simulator (HFSS) software. HFSS software is a high performance full-wave electromagnetic simulator and has an effective graphical user interface. HFSS software is basically providing reference data. Simulation of the HFSS software package can remove the excessive cost during the fabrication and

provides positive contribution on the results during the produciton stage. On the other hand, this model has high learning curve and fairly low physical anlaysis capability.

In order to obtain desired parameters of antenna, simulation programs generates results in a long period of time due to its heavy computation load and complex analytical algorithm behind. Therefore as an alternative to the HFSS, new computer aided methods should be investigated. One of these methods is the Artificial Neural Network (ANN). Learning ability, rapid applicability on various problems, generalization capability, requiring less information, fast and easily processing power make ANN popular in recent years for this particular problem.

According to many studies, ANN can address the chalenging problems particularly resonant frequency which are actually quite complex and time consuming processes. In this study, producing desired parameters in the range of 1 – 3.5 GHz for ACMA, eligible ANN model was developed. Outputs of the developed ANN model were evaluated and then compared to HFSS simulation software results. It was observed that our proposed method is more efficient (100 times faster than HFSS software) and has acceptable accuracy rate (96.5 %) with respect to the reference HFSS model.

Keywords: Artificial neural networks, Aparture

coupled microstrip patch antennas, Resonance frequency, Antenna design.

0LNURúHULWDQWHQOHULQ\DSD\VLQLUD÷PRGHOL

GLULú

7HNQRORMLQLQ JHOLúPHVLQH SDUDOHO RODUDN

NDEORVX] LOHWLúLPLQ LOJL J|UPHVL VRQ \ÕOODUGD anten teknolojisinin JHOLúPHVLQH olanak VD÷ODPÕúWÕU 7UNHU YG  =RU NDUPDúÕN YH ]DPDQ DOÕFÕ DQWHQ WDVDUÕPODUÕQGDki anten parametreleri, uzun analizler ve WDVDUÕP G|QJOHUL NXOODQÕODUDN KHVDSODQÕUODU ;LDR YG 2003).

/LWHUDWUGHNL PHYFXW NODVLN YH QPHULN \|QWHPOHU NXOODQÕODUDN PLNURúHULW DQWHQOHULQ UH]RQDQV IUHNDQVÕ LoLQ HOGH HGLOHQ WHRULN VRQXoODUGHQH\VHOVRQXoODUODL\LELUX\XPOXOXN

sergilememektedir. gQFHNL oDOÕúPDODUGD

|÷UHQPH EHFHULVL NROD\FD IDUNOÕ SUREOHPOHUH X\DUODQDELOLUOL÷L JHQHOOHPH NDELOL\HWL GDKD D] ELOJL JHUHNWLUPHVL SDUDOHO \DSÕODUÕQGDQ GROD\Õ KÕ]OÕ oDOÕúDELOPH \HWHQH÷L YH NROD\ ELU úHNLOGH X\JXODQDELOPHVL JLEL SHN oRN DYDQWDMODUÕQGDQ GROD\Õ <6$ HWNLQ YH YHULPOL ELU úHNLOGH NXOODQÕOPÕúWÕU*OWHNLQYG

YSA PRGHOOHULQLQ DQWHQ WDVDUÕPODUÕQGDNL NDUPDúÕNYH]DPDQDOÕFÕPDWHPDWLNVHOLúOHPOHUL 0RPHQW \|QWHPL JLEL   RUWDGDQ NDOGÕUDUDN NDEORVX] KDEHUOHúPH PKHQGLVOL÷LQGH de etkin RODUDN NXOODQÕODELOHFH÷L J|VWHULOPLúWLU âPtG ve Raida, 2006). )DUNOÕ úHNLOOHUGHNL GLNG|UWJHQ oJHQYEPLNURúHULW\DPDDQWHQOHULQUH]RQDQV IUHNDQVÕQÕ EHOLUOHPHN LoLQ GH÷LúLN <6$ PRGHOOHUL|QHULOPLúWLU.D\DEDúÕYG dHúLWOLWLSWHNLPLNURúHULW antenlerin karakteristik SDUDPHWUHOHULQLQ EDúDUÕOÕ ELU úHNLOGH <6$ LOH KHVDSODQGÕ÷Õ oDOÕúPDODU OLWHUDWUGH PHYFXWWXU *OWHNLQYG

AKMYA  \ÕOÕQGD ilk olarak Pozar WDUDIÕQGDQ |QHULOPLú YH DOWDú PDO]HPH \NVHNOL÷L LOH E\N EDQG JHQLúOL÷L DUDOÕ÷ÕQÕ JHOLúWLUPHGHNXOODQÕOPÕúWÕU

%XoDOÕúPDGDJHOLúWLULOHQ|QHULOHQYSA modeli,  *+] LOH  *+] IUHNDQV DUDOÕ÷ÕQGD oDOÕúDQ

$.0<$¶QÕQ UH]RQDQV IUHNDQVÕQÕQ

KHVDSODQPDVÕQGDNXOODQÕOPÕúWÕU

Anten tDVDUÕPÕ

Bir AKMYA ùHNLO ¶GH J|VWHULOPHNWHGLU AKMYA, tRSUDN \]H\L WDUDIÕQGDQ D\UÕOPÕú LNL tabakadan meydana gelmektedir. hVW \DOÕWNDQ

İ

İ

PLNURúHULW EHVOHPH KDWWÕQÕ LoHULU Toprak \]H\LQGH NHVLOPLú NoN DUDOÕN DoÕN GHYUH

GXUXPXQGD EXOXQDQ PLNURúHULW EHVOHPH

KDWWÕQGDQ ÕúÕPD \DSDQ \DPD\D ED÷ODQWÕ \DSÕOPDVÕQD L]LQ YHULU $QWHQLQ rezonans IUHNDQVÕ, \D\ÕOÕP \DSDQ \DPDQÕQ úHNOL YH ER\XWODUÕ WRSUDN G]OHPLQLQ ER\XWODUÕ DoÕNOÕN (slot) NÕVPÕQÕQ ER\XWODUÕ EHVOHPH KDWWÕQÕQ pozisyonu ve ER\XWODUÕ YH LNL \DOÕWNDQ malzemenin dielektrik sabitiQHED÷OÕGÕU(Bose ve Gupta, 2008).

ùHNLO$.0<$VWWHQYH\DQGDQJ|UQú (Kuchar, 1996)

%LU $.0<$ X\JXQ ELU GLHOHNWULN DOWDú malzeme ]HULQGH IDUNOÕ IUHNDQVODU LoLQ WDVDUODQPÕú RODELOLU <D\ÕOÕP \DSDQ \DPD YH DoÕNOÕN NÕVPÕ GLNG|UWJHQ olarak WDVDUODQÕU. Bir

$.0<$ WDVDUÕPÕQGD NXOODQÕODQ anten

SDUDPHWUHOHULúXQODUGÕU

x <DPD ER\XWODUÕ <DPDQÕQ X]XQOX÷X Lp

YHJHQLúOL÷LWp

x $oÕNOÕN ER\XWODUÕ $oÕNOÕ÷ÕQ X]XQOX÷X Lap YHJHQLúOL÷LWap

x %HVOHPH ER\XWODUÕ %HVOHPH KDWWÕQÕQ X]XQOX÷XLf YHJHQLúOL÷L Wf

Veri kümesinin elde edilmesi

%X oDOÕúPDGD NXOODQÕODQ YHUL NPHVL |UQHNOHUL PRPHQWOHU \|QWHPLQH GD\DOÕ +)66 VLPODW|U \D]ÕOÕPÕNXOODQÕODUDNUHWLOPLúWLUToplamda 500 DGHW |UQHN UHWLOPLúWLU Veri setindeki her bir |UQHN EHú DGHW $.0<$ JLULú SDUDPHWresi (Lp, Wp, Lap, Wap, Lf) YHELUDGHWoÕNÕúUH]RQDQV

frekans Fr GH÷HULQGHQ PH\GDQD JHOPLúWLU +HU

LNL DOWDú \DOÕWNDQ PDO]HPHQLQ GLHOHNWrik sabitleri, 2. \H VDELWOHQPLúWLU 'úN PDOL\HWOL ROXúXNROD\NXOODQÕODELOLUOL÷LYH GLHOHNWULNND\ÕS WDQMDQWÕQÕn 0.¶G% VHYL\HVLQGH ROPDVÕndan GROD\Õ 5RJHUV 57GXURLG Œ dielektrik malzemesi VHoLOPLúWLU +)66 3 boyutlu tam dalga elektromanyetik alan sLPODW|U arabirimi ùHNLO¶GHJ|VWHULOPHNWHGLU

ùHNLO2. +)66\D]ÕOÕPDUD\]

+)66¶GH NXOODQÕODQ $.0<$¶QÕQ JLULú

SDUDPHWUH GH÷HUOHULQH NDUúÕOÕN elde edileQ oÕNÕú UH]RQDQV IUHNDQV GH÷HUOHULQGHQ ED]ÕODUÕ 7DEOR ¶GHOLVWHOHQPLúWLU Tablo 1. +)66¶GHNXOODQÕODQED]Õ$.0<$¶QÕQ JLULúSDUDPHWUHGH÷HUOHULYHHOGHHGLOHQoÕNÕú UH]RQDQVIUHNDQVGH÷HUOHUL 6ÕUD No *LULú3DUDPHWUHOHUL dÕNÕú Parametresi Lp (cm) (cm) Wp (cm) Lap (cm) Wap (cm) Lf F(HFSS) r (GHz) 1 4.446 3.602 0.203 1.371 6.948 1.950 2 3.977 3.123 0.197 1.656 6.164 2.090 3 3.838 3.415 0.164 1.895 5.786 2.190 4 3.010 3.639 0.160 1.836 5.396 2.500 5 2.513 2.648 0.201 1.162 6.225 3.110

*HOLúWLULOHQyapay sinir a÷modeli

%X oDOÕúPDGD |QHULOHQ <6$ PRGHOL YH VÕQÕIODQGÕUÕFÕ SHUIRUPDQV DOJRULWPDODUÕ *18 2FWDYH \D]ÕOÕPÕ NXOODQÕODUDN JHOLúWLULOPLúWLU (Eaton, 2001 ).

<DSD\VLQLUD÷WDVDUÕPÕ

dDOÕúPDGD o NDWPDQOÕ <6$ PLPDULVL EHQLPVHQPLúWLU ho NDWPDQGDQ GDKD E\N

PLPDULOHUGH NDUPDúÕNOÕN DUWÕúÕ VWHO

H[SRQHQWLDO RODUDN |÷UHQPH YHULVLQLQ DUWÕúÕQÕ JHUHNWLUGL÷LQGHQ HQEDVLWPLPDUL RODQWHN JL]OL NDWPDQ PLPDULVL VHoLOPLúWLU AKMYA

WDVDUÕPÕQGD NXOODQDFD÷ÕPÕ] JHRPHWULN

SDUDPHWUH VD\ÕVÕ JLULú NDWPDQÕQGDNL Q|URQ VD\ÕVÕQÕ UH]RQDQV IUHNDQV VD\ÕVÕ LVH oÕNÕú NDWPDQÕQGDNLQ|URQVD\ÕVÕQÕEHOLUtir.

YSA modelinde gLULú SDUDPHWUHOHUL olarak Lp,

Wp, Lap, Wap ve Lf NXOODQÕOPÕúWÕU dÕNÕú

parametresi ise rezonans freNDQVÕ Fr

dir

<DSÕODQ GHQHPHOHU VRQXFXQGD HQ X\JXQ <6$ topolojisinin; 5 Q|URQOX JLULú katman, 8 Q|URQOX gizli katman ve 1 Q|URQOX oÕNÕú NDWPDQ modeli ROGX÷X WHVSLW HGLOPLúWLU gQHULOHQ <6$ PRGHOL ùHNLO 3’ te J|VWHULOPLúWLU

ùHNLO3*HOLúWLULOHQ\DSD\VLQLUD÷PRGHOLQLQ \DSÕVÕ

ùHNLO WHNL ;;5 JLULú NDWPDQÕQGDNL JLULú Q|URQODUÕ 7DEOR GHNL JLULú SDUDPHWUHOHULQL

0LNURúHULWDQWHQOHULQ\DSD\VLQLUD÷PRGHOL

temsil etmektedir. H1..H8 ise gizli katmandaki Q|URQODUÕ<LVHoÕNÕúNDWPDQÕQGDNL7DEORGH J|VWHULOHn oÕNÕú SDUDPHWUHVLQL J|VWHUPHNtedir. øONED÷ODQWÕ D÷ÕUOÕNODUÕ YHELDVGH÷HUOHULrandom IRQNVL\RQXLOHUHWLOPHNWHGLU. dDOÕúPDVUHVLQFH \DSÕODQ ELUoRN GHQHPHOHU VRQXFXQGD JHUHN <6$GDNL VHUEHVW SDUDPHWUHOHULQ GH÷HUOHUL JHUHNVH GH JL]OL NDWPDQGDNL HQ X\JXQ Q|URQ VD\ÕVÕ HQ NoN WHVW KDWDVÕ UHWHFHN úHNLOGH JQFHOOHQHUHN HQ X\JXQ |÷UHQPH NDWVD\ÕVÕ  H÷LWLP HSRN VD\ÕVÕ 150 ve gizli katmandaki Q|URQVD\ÕVÕRODUDNVDELWOHQPLúWLU

6LQLUD÷PRGHOLQLQH÷LWLPL

*HOLúWirilen YSA modeli, Levenberg-Marquardt |÷UHQPH DOJRULWPDVÕ NXOODQÕODUDN momentler

\|QWHPLQH GD\DOÕWLFDULELUVLPODV\RQ\D]ÕOÕPÕ

RODQ+)66¶GHQUHWLOHQ |UQHNOHU LOHH÷LWLOPLúWLU. <6$QÕQ JL]OL NDWPDQÕQGD KLSHUEROLN WDQMDQW VLJPRLG IRQNVL\RQX YH oÕNÕú NDWPDQÕQGD VDI GR÷UXVDOIRQNVL\RQXWUDQVIHUIRQNVL\RQXRODUDN NXOODQÕOPÕúWÕU (÷LWLP VÕUDVÕQGD JLULú |UQHNOHUL VLQLU D÷ÕQGDNL JLULú NDWPDQÕQGD EDúDUÕOÕ ELU úHNLOGH |÷UHWLOHUHN VLQDSWLN D÷ÕUOÕNODU LVWHQHQ oÕNÕú \DQÕWODUÕQD XODúPDN LoLQ GH÷LúWLULOLU - HSRN FLYDUÕQGD H÷LWLP KDWDVÕ PLQLPL]H HGLOGL÷LQGHQ GROD\Õ H÷LWLP LúOHPL VRQODQGÕUÕOÕU %XQXQ VRQXFXQGD H÷LWLOPLú VLQLU D÷Õ H÷LWLP VUHVLQFH J|UPHGL÷L WHVW YHUL VHWL NXOODQÕODUDN GR÷UXOX÷X|OoOU

1|UDOPRGHOLQLQGR÷UXOX÷X

9HUL PDGHQFLOL÷L YH VDQDO |÷UHQPH

X\JXODPDODUÕQGD . NDW oDSUD] GR÷UXODPD \|QWHPL oRN VÕN NXOODQÕOPDNWDGÕU . ED]Õ oDOÕúPDODUGD  ED]ÕODUÕQGD LVH  RODUDk VHoLOPHNWHGLU $WDV YG.  . VHoLPLQGHNL JHQHO \DNODúÕP YHUL NPHVLQLQ E\NO÷ LOH LOJLOLGLU%\NYHULNPHOHULQGH.VD\ÕVÕNoN ELU GH÷HUNHQ D] HOHPDQOÕ YHUL NPHOHULQGH VÕQÕIODQGÕUÕFÕODUÕQ |÷UHQPH VUHFLQGH YHUL NPHVLQGHQPDNVLPXP\DUDUODQDELOPHVLLoLQ. VD\ÕVÕ E\N WXWXOXU $WDV YG., 2012). *HOLúWLULOHQ <6$ PRGHOLQLQ GR÷UXOX÷X  NDW oDSUD]GR÷UXODPD-Fold Cross Validation) ile GH÷HUOHQGLULOPLúWLU  NDW oDSUD] GR÷UXODPD úX úHNLOGH oDOÕúPDNWDGÕU 9HUL NPHVL UDQGRP IRQNVL\RQX NXOODQÕODUDN \HQLGHQ VÕUDODQGÕNWDQ

VRQUD EHú SDUoD\D E|OQU +HU ELU NDWWD EHú E|OPGHQ ELUL WHVW NPHVL RODUDN D\UÕOGÕNWDQ VRQUD JHUL\H NDODQ  SDUoD ]HULQGHQ H÷LWLP LúOHPL JHUoHNOHúWLULOLU (÷LWLP LúOHPL ELWWL÷LQGH GR÷UXOXN SHUIRUPDQVÕ D\UÕOPÕú WHVW SDUoDVÕ ]HULQGHQKHVDSODQÕU %XLúOHP GL÷HUSDUoDODUÕQ GDHQD]ELUNH]WHVWYHULNPHVLRODFD÷ÕúHNLOGH EHú GHID WHNUDUODQÕU (Q VRQGD EHú IDUNOÕ GR÷UXOXN GH÷HULQLQ RUWDODPDVÕ DOÕQÕU %X \|QWHPYHULNPHVLQGHNL YHUL\HHQD]H÷LOLPOL

ROPDVÕQGDQ GROD\Õ VDQDO |÷Uenme

X\JXODPDODUÕQGD oRN VÕN NXOODQÕOPDNWDGÕU (Breiman ve Spector, 1992). YSA modelindeki JLULú ED÷ODQWÕ D÷ÕUOÕNODUÕ YH ELDV GH÷HUOHUL H÷LWLPH EDúODPDGDQ |QFH UDQGRP GH÷HUOHU DODFD÷ÕQGDQ KHU oDSUD]ODPD NDWÕ LoLQ <6$ PRGHOL  GHID H÷LWLP YH WHVW \DSÕODUDk RUWDODPD KDWD RUDQÕ YH VWDQGDUW VDSPD PLNWDUÕ EHOLUOHQPLúWLU  'ROD\ÕVÕ\OD QLKDL KDWD RUDQÕ  NDW oDSUD]ODPD LoLQ  GHQHPH VRQXFXQGD KHVDSODQPÕúWÕU +HUELUWHVW|UQH÷LLoLQ+DWDDúD÷ÕGDNLIRUPO ile KHVDSODQPÕúWÕU

Uygulama

$QWHQOHUGH SHUIRUPDQVÕ HWNLOH\HQ \D GD SHUIRUPDQVÕQÕQQHGHUHFHL\LROGX÷XQXJ|VWHUHQ parametrelerden biri Return Loss (RL) olarak ta bilinen geri G|Qú ND\EÕGÕU. dHúLWOL $.0<$ WDVDUÕPODUÕ +)66 \D]ÕOÕPÕQGD WHVW HGLOHUHN JHUL G|QúNDUDNWHULVWL÷LJUDIL÷LLQFHOHQPLúWLU ø\L ELU DQWHQ SHUIRUPDQVÕ  5/ Ӌ 9.5dB ) ROGX÷X GXUXPODUGD VD÷ODQÕU *KRVK YH 3DUXL 2010). ùHNLO ¶WH LGHDO ELU DQWHQ WDVDUÕPÕQD örQHNYHULOPLúWLU*HULG|QúND\EÕQÕQ-9.5 dB¶ den NoN ROPDVÕGXUXPXQXJ|VWHUHQ ùHNLO5¶WH ise LGHDO ROPD\DQ ELU DQWHQ WDVDUÕPÕ |UQHN YHULOPLúWLU.

ùHNLO. $.0<$¶QÕQJLULúSDUDPHWUH GH÷HUOHULQH J|UHLVWHQHQUH]RQDQVIUHNDQV

FHYDEÕ

ùHNLO5. AKMYA¶QÕQJLULúSDUDPHWUH GH÷HUOHULQHJ|UHistenilmeyen rezonans frekans

cevaEÕ

YSA modeli ve HFSS’den elde edilen sonuçlardan ELUNÕVPÕ Tablo 2’GH J|VWHULOPLúROXS frekans GH÷HUOHULQLQuyXPLoLQGHROGX÷XJ|UOPHNWHGLU 7DEORgQHULOHQ<6$PRGHOLYH+)66¶QLQIUHNDQV VRQXoODUÕQGDNLKDWD|OoPOHUL 6ÕUD No *LULú3DUDPHWUHOHUL dÕNÕú3DUDPHWUHOHUL Lp (cm) Wp (cm) Lap (cm) Wap (cm) Lf (cm) Fr(GHz) (HFSS) Fr(GHz) (YSA) % Hata 1 4.390 3.651 0.201 2.024 5.349 1.830 1.8 1.64 2 3.977 3.123 0.197 1.656 6.164 2.090 2.1 0.47 3 3.838 3.415 0.164 1.895 5.786 2.190 2.2 0.45 4 3.562 2.732 0.207 1.892 6.581 2.290 2.3 0.43 5 3.380 3.337 0.177 1.863 5.915 2.400 2.4 0.00 6 3.010 3.639 0.160 1.836 5.396 2.500 2.5 0.00 7 2.847 3.684 0.205 1.595 5.718 2.690 2.7 0.37 8 2.682 2.745 0.185 1.828 6.752 2.890 2.9 0.34 9 2.572 2.555 0.176 1.669 6.479 2.980 3.0 0.67 10 2.513 2.648 0.201 1.162 6.225 3.110 3.1 0.32

$\UÕFD 2FWDYH¶GH JHUoHNOHúWLULOHQ <6$

modelinin test sonucu ile HFSS test sonucunun NDUúÕODúWÕUÕOPDVÕ ùHNLO 6’da J|VWHULOPLúWLU Elde HGLOHQ <6$ VRQXoODUÕQÕQ +)66 VRQXoODUÕQD L\L GHUHFHGHX\XPVHUJLOHGL÷LL]OHQPHNWHGLU

ùHNLO. YSA ile HFSS VRQXoODUÕQÕQ NDUúÕODúWÕUÕOPDVÕ

0LNURúHULWDQWHQOHULQ\DSD\VLQLUD÷PRGHOL

Sonuçlar

7HVWHGLOHQIUHNDQVGH÷HUOHUL+)66VLPODV\RQ \D]ÕOÕP VRQXoODUÕ LOH NDUúÕODúWÕUÕOPÕúWÕU <6$ PRGHOL LOH KD]ÕUODQPÕú $.0<$¶QÕQ \XNDUÕGD EHOLUWLOHQ |]HOOLNOHUL NDEXO HGLOHELOLU KDWD VÕQÕUÕ LoLQGH  EXOXQPXúWXU %XQXQOD EHUDEHU ¶OLN JHQHOOHPH KDWDVÕ LOHULGH \DSÕODFDN GHWD\OÕ oDOÕúPDODUGD WHNUDU HOH DOÕQDUDN GDKD DúD÷ÕODUD oHNLOPHVLJHUHNPHNWHGLU

(÷LWLOPLú D÷ KHUKDQJL ELU JHRPHWULN $.0<$ WDVDUÕPÕ LoLQ JHUHNOL IUHNDQV GH÷HULQLQ KHVDSODQPDVÕQGD VLPXODW|U \D]ÕOÕPÕ +)66 GHQ GDKDKÕ]OÕFHYDSYHUPHVUHVLQHKHUELULúOHPGH ortalama olarak YS$+)66¶GHQNDWGDKD KÕ]OÕ) sahLSWLU gQHULOHQ <6$ PRGHOLQLQ FHYDSODPDVUHVLVD\ÕVDOPRGHOHJ|UHoRNGDKD NÕVD YH KÕ]OÕ ROGX÷XQGDQ YH H÷LWLOHQ VLQLU D÷Õ JLULú|UQWOHULLoLQ\HWHUOLELUKDVVDVL\HWOHoÕNÕú FHYDEÕQÕ YHUGL÷LQGHQ VLPODW|U +)66 \D]ÕOÕPÕ \HULQH DúÕUÕ KDVVDV ROPD\DFDN $.0<$ WDVDUÕPODUÕQGDNXOODQÕODELOLU

$÷ÕQ JL]OL NDWPDQODUÕ YH LoHUGL÷L LGHDO Q|URQ VD\ÕVÕ GDKD L\L D\DUODQDUDN YH H÷LWLP |UQHNOHUL

oR÷DOWÕODUDN <6$ PRGHOLQLQ JHQHOOHPH

GR÷UXOXN SHUIRUPDQVÕ YH JUE]O÷

DUWWÕUÕODELOLU

%X oDOÕúPDGD JHOLúWLULOHQ <6$ PRGHOL anten SDUDPHWUHOHULQGHQ VDGHFH UH]RQDQV IUHNDQVÕQÕQ EXOXQPDVÕQGD NXOODQÕOPDNWDGÕU 'L÷HU WDUDIWDQ EDQW JHQLúOL÷L JHUL G|Qú ND\EÕ YE DQWHQ SDUDPHWUHOHULQL GH LoLQH DODFDN NDSVDPOÕ ELU

oDOÕúPDQÕQ LOHULNL ]DPDQODUGD \DSÕOPDVÕ

GúQOPHNWHGLU

$\UÕFD $.0<$¶GD oÕNÕúÕ ELOLQHQ UH]RQDQV IUHNDQV YH JHUL G|Qú ND\EÕ GH÷HUOHUL <6$ PRGHOLQGH JLULú SDUDPHWUHVL RODUDN NXOODQÕOÕS oÕNÕúWDDQWHQLQJHRPHWULNER\XWODUÕQÕEXODFDNELU DUDúWÕUPDGD\DSÕODELOLU

Kaynaklar

$WDú 0 <DUGLPFL < ve Temizel, A., (2012). A new approach to aflatoxin detection in chili pepper by machine vision. Computers and Electronics in Agriculture, 87, 129-141.

$WDú08\DU0 ve Kaya Y., (2103). An efficient rotation invariant feature extraction method based on ring projection technique6,8.ÕEUÕV Bose, T. ve Gupta, N., (2008). Neural Network

Model for Aperture Coupled Microstrip Antennas, Microwave Review, pp. 21-24.

Breiman, L. ve Spector, P., (1992). Submodel selection and evaluation in regression: The X-random case, 524 International Statistical Review, 60, 291-319.

Ghosh, C. K. ve Parui, S. K., (2010). Design, Analysis and Optimization of A Slotted Microstrip Patch Antenna Array at Frequency 5.25 GHz for WLAN-SDMA System. International Journal on Electrical Engineering and Informatics - Volume 2, Number 2

*OWHNLQ 6 *QH\ . YH 6D÷ÕUR÷OX ù  )DUNOÕ g÷UHQPH $OJRULWPDODUÕ .XOODQÕODUDN (÷LWLOHQ<DSD\6LQLU$÷ODUÕøOH(OHNWULNVHO2ODUDN øQFH YH .DOÕQ 'LNG|UWJHQ 0LNURúHULW $QWHQOHULQ 5H]RQDQV 'LUHQFLQLQ +HVDSODQPDVÕ URSI-7h5.ø<(¶ 18-20ø7h

Eaton, J. W., (2001). Octave: Past, present and future. In Proceedings of the 2nd International Workshop on Distributed Statistical Computing. .D\DEDúÕ $%LoHU0%$NGD÷OÕ$ YH7RNWDú

$  8KI %DQGÕQGD dDOÕúDQ + ùHNLOOL .RPSDNW 0LNURúHULW $QWHQOHULQ 5H]RQDQV )UHNDQVÕQÕQ <DSD\ 6LQLU $÷ODUÕ .XOODQDUDN +HVDSODQPDVÕ *D]L hQLYHUVLWHVL 0KHQGLVOLN 0LPDUOÕN )DNOWHVL 'HUJLVL, Cilt 26, No 4, 833-840.

Kuchar, A., (1996). Aperture-Coupled Microstrip Patch Antenna Array, Friedhofallee 4a/11 A-2232 Deutsch-Wagram.

Pozar, D. M., (1985). A Microstrip Antenna Aperture Coupled to a Microstrip Line, Electronics Letters, Vol. 21, pp.49-5O.

âPtG3YH Raida, Z., (2006). Application of Neural Networks:Enhancing Efficiency of Microwave Design, Microwave Review, pp. 2-9.

7UNHU 1 *QHú ) ve Yildirim, T., (2006). Artificial Neural Networks Applied to the Design of Microstrip Antennas, Microwave Review,. pp. 10 – 14.

Xiao, S., Wang, B. Z., Zhong, X. ve Wang, G., (2003). Wideband Mobile Antenna Design Based on Artificial Neural Network Models, Wiley Periodicals, pp. 316 - 320.