Advances in Antenna Technology for Wireless Handheld Devices

Advances in Antenna Technology for Wireless Handheld Devices

Guest Editors Jaume Anguera Aurora Anduacutejar Minh-Chau Huynh and Charlie Orlenius

International Journal of Antennas and Propagation

Advances in Antenna Technology for WirelessHandheld Devices



Guest Editors Jaume Anguera Aurora AndujarMinh-Chau Huynh and Charlie Orlenius

Copyright copy 2013 Hindawi Publishing Corporation All rights reserved

This is a special issue published in ldquoInternational Journal of Antennas and Propagationrdquo All articles are open access articles distributedunder the Creative Commons Attribution License which permits unrestricted use distribution and reproduction in any medium pro-vided the original work is properly cited

Editorial Board

M Ali USACharles Bunting USAFelipe Catedra SpainDau-Chyrh Chang TaiwanDeb Chatterjee USAZ N Chen SingaporeMichael Yan Wah Chia SingaporeChristos Christodoulou USAShyh-Jong Chung TaiwanLorenzo Crocco ItalyTayeb A Denidni CanadaAntonije R Djordjevic SerbiaKaru P Esselle AustraliaFrancisco Falcone SpainMiguel Ferrando SpainVincenzo Galdi ItalyWei Hong ChinaHon Tat Hui SingaporeTamer S Ibrahim USAShyh-Kang Jeng Taiwan

Mandeep Jit Singh MalaysiaNemai Karmakar AustraliaSe-Yun Kim Republic of KoreaAhmed A Kishk CanadaTribikram Kundu USAByungje Lee Republic of KoreaJu-Hong Lee TaiwanL Li SingaporeYilong Lu SingaporeAtsushi Mase JapanAndrea Massa ItalyGiuseppe Mazzarella ItalyDerek McNamara CanadaC F Mecklenbrauker AustriaMichele Midrio ItalyMark Mirotznik USAAnanda S Mohan AustraliaP Mohanan IndiaPavel Nikitin USAA D Panagopoulos Greece

Matteo Pastorino ItalyMassimiliano Pieraccini ItalySadasiva M Rao USASembiam R Rengarajan USAAhmad Safaai-Jazi USASafieddin Safavi Naeini CanadaMagdalena Salazar-Palma SpainStefano Selleri ItalyKrishnasamy T Selvan IndiaZhongxiang Q Shen SingaporeJohn J Shynk USASeong-Youp Suh USAParveen Wahid USAYuanxun Ethan Wang USADaniel S Weile USAQuan Xue Hong KongTat Soon Yeo SingaporeJong Won Yu Republic of KoreaWenhua Yu USAAnping Zhao China

Contents

Advances in Antenna Technology for Wireless Handheld Devices Jaume Anguera Aurora AndujarMinh-Chau Huynh and Charlie OrleniusVolume 2013 Article ID 376531 2 pages

Advances in Antenna Technology for Wireless Handheld Devices Jaume Anguera Aurora AndujarMinh-Chau Huynh Charlie Orlenius Cristina Picher and Carles PuenteVolume 2013 Article ID 838364 25 pages

Evaluation of SAR Distribution in Six-Layer Human Head Model Asma Lak and Homayoon OraiziVolume 2013 Article ID 580872 8 pages

Printed Internal Pentaband WWAN Antenna Using Chip-Inductor-Loaded Shorting Strip for MobilePhone Application Yong-Ling Ban Shun Yang Joshua Le-Wei Li and Rui LiVolume 2012 Article ID 516487 7 pages

Compact Dual-Band Dual-Polarized Antenna for MIMO LTE Applications Lila MouffokAnne Claire Lepage Julien Sarrazin and Xavier BegaudVolume 2012 Article ID 398423 10 pages

Band-Notched Ultrawide Band Planar Inverted-F Antenna H T Chattha M K Ishfaq Y SaleemY Huang and S J BoyesVolume 2012 Article ID 513829 6 pages

Hindawi Publishing CorporationInternational Journal of Antennas and PropagationVolume 2013 Article ID 376531 2 pageshttpdxdoiorg1011552013376531

EditorialAdvances in Antenna Technology for Wireless Handheld Devices

Jaume Anguera12 Aurora Anduacutejar1 Minh-Chau Huynh34 and Charlie Orlenius5

1 Technology and Intellectual Property Rights Department Fractus Barcelona Spain2 Electronics and Communications Department Ramon Llull University Barcelona Spain3 Systems and Concept Sony Mobile Redwood City CA USA4Communications Systems Group LitePoint Corporation CA USA5 Bluetest AB Gothenburg Sweden

Correspondence should be addressed to Jaume Anguera jaumeanguerafractuscom

Received 16 December 2012 Accepted 16 December 2012

Copyright copy 2013 Jaume Anguera et alThis is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Communication between two distant points has been aconstant challenge for mankind from ancient smoke signalsto telegraph to finally wireless communication throughelectromagnetic signals This evolution represents a constanteffort to improve the quality and effectiveness of distancecommunication with ever-evolving techniques to enhancethe delivery of contents from voice to data Wireless hand-held devices are the most representative paradigm of theseefforts Since they first appeared their size has continuouslybeen shrinking while their functional capabilities have beenincreasing hence creating the never-ending challenge inantenna design In this regard the antenna community oftenhas the important role of designing low-profile small andmultiband antennas capable of being integrated within thehandset platform as well as capable to coexist with multipleantenna systems in order to satisfy the strict demands ofemergent multifunction wireless devices Furthermore thecomplexity of handheld antenna design is continuouslyincreasing not only by the pressure of the market needs butalso by the duty of safety regulations which require efficientantennas capable of radiating as much power as possible infree-space conditions while minimizing the power radiatedtowards the human head

Antenna modeling in handset devices using electro-magnetic simulation software has improved significantlythanks to the progress of computing hardware Complexenvironments surrounding the antenna such as a handsetdevice held beside a human head and precise details onnearby components including the presence of a loudspeaker

in the antenna volume can be modeled accurately to predictantenna performance that is closer to reality without sacrific-ing simulation speedThis evolution considerably contributesto simplify the antenna design process

Not only the simulation tools have considerably evolvedin the latest years but also the measurement systems havebeen forced to evolve for satisfying the emergent commu-nication systems requirements These recent advances inmeasurement systems and methodologies have been hottopics in the antennameasurement community for capturingfor instance radiated performance in the emergent LTE andMIMO antenna systemsThese next generation systems havealready started to appear in wireless handheld devices inthe consumer market However new measurement methodsneed to be developed as these antenna systems are to be usedas well as tested in fading environments

Finally the commercial success of wireless handhelddevices leads to an improvement of the manufacturing tech-niques and processesThis is not only important for reducingthe cost ofmass production but also for enhancing the designperformance and size in a controlled fashion

This special issue contains five papers that gather someof the recent advancements in handset antenna design In thepaper entitled ldquoBand-notched ultrawide band planar inverted-F antennardquo an ultrawide planar inverted-F antenna coveringthe 34GHzndash112 GHz band with a band-notch at 508GHzndash6GHz is presented The wideband behavior is obtained byparasitic elements whereas the band-notch is achieved by aW-shaped slot on the top radiating element of the antenna

2 International Journal of Antennas and Propagation

The paper ldquoCompact dual-band dual-polarized antennafor MIMO LTE applicationsrdquo proposes an antenna sys-tem operating in the LTE bands 700MHzndash862MHz and25GHzndash269GHz The design is composed of two compactorthogonal monopoles to perform diversity in mobile termi-nals such as tablets or laptops

In the paper ldquoPrinted internal pentabandWWANantennausing chip-inductor-loaded shorting strip for mobile phoneapplicationrdquo a compact size on-Board printed antenna usingcapacitive coupled-fed excitation to generate multiple reso-nant modes for pentabandWWAN operation (GSM850900GSM18001900 UMTS2100) is designed

The paper ldquoEvaluation of SAR distribution in six-layerhuman head modelrdquo numerically analyzes a single layer anda six-layer human head model for SAR computation at the900MHz frequency

Finally in the paper ldquoAdvances in antenna technology forwireless handheld devicesrdquo the evolution of wireless handhelddevices regulations and challenges in todayrsquos smartphonesand handset characterization are reviewed Finally recentadvancements in antenna technology for wireless handheldor portable devices are presented

Jaume AngueraAurora Andujar

Minh-Chau HuynhCharlie Orlenius


Review ArticleAdvances in Antenna Technology forWireless Handheld Devices

Jaume Anguera1 2 Aurora Anduacutejar1 Minh-Chau Huynh3 Charlie Orlenius4

Cristina Picher1 and Carles Puente1 5

1 Technology and Intellectual Property Rights Department Fractus 08190 Barcelona Spain2 Electronics and Communications Department Universitat Ramon Llull 08022 Barcelona Spain3 Systems and Concept Sony Mobile Redwood City CA 94085 USA4 Bluetest AB Lindholmsalleacuten 10 417 55 Gothenburg Sweden5Department of Signal eory and Communications Universitat Politegravecnica de Catalunya 08034 Barcelona Spain


Received 24 August 2012 Accepted 27 November 2012

Academic Editor Mandeep Singh Jit Singh

Copyright copy 2013 Jaume Anguera et alis is an open access article distributed under the Creative CommonsAttribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

e constant evolution of wireless handheld devices together with the apparition of multiple wireless communication systemsfosters the antenna community to design new radiating and measurements systems capable of satisfying the market demands It isan object of the present paper to provide an overview of the evolution that wireless handheld technology has experienced in thelast years In this sense a description of the evolution of wireless handheld devices regulations challenges in todayrsquos smartphonesand handset characterization is reviewed Finally recent advances in antenna technology for wireless handheld or portable devicesare presented

1 Introduction

Communication between two distant points has been aconstant challenge for mankind from ancient smoke signalsto telegraph to nally wireless communication throughelectromagnetic signals is evolution represents a constanteffort to improve the quality and effectiveness of distancecommunication with ever-evolving techniques to enhancethe delivery of contents from voice to data Wireless hand-held devices are the most representative paradigm of theseefforts In this regard the antenna community oen hasan important role focused on designing low-prole smalland multiband antennas together with multiple antennasystems capable of satisfying the strict demands of emergentmultifunction wireless devices Furthermore the complexityof handheld antenna design is continuously increasing notonly by the pressure of the market needs but also by theduty of safety regulations which require efficient antennascapable of radiating as much power as possible in free-spaceconditions while minimizing the power radiated towards thehuman head

Antenna modeling in handheld devices using electro-magnetic simulation soware has improved signicantly byallowing the simulation of the antenna behavior in com-plex environments surrounding the antenna us currentelectromagnetic soware allows the simulation of handheldantennas regarding not only the human presence (such ashuman head and hand) but also the presence of nearby com-ponents (such as cameras batteries displays and speakers)

At the same time recent advances in measurementsystems and methodologies have become hot topics in theantenna measurement community for capturing radiatedperformance in emergent LTE and MIMO antenna systems

Finally with the commercial success of wireless handhelddevices comes the important role of good manufacturingtechniques is is not only important for reducing thecost of mass production but also for enhancing the designperformance and size in a controlled fashion

e paper is divided into the following sections Section2 describes the evolution of handheld mobile telephonesand generations the apparition of new frequency bands theindustrial design inuence on antennas requirements and


regulations and nally antenna design challenges in todayrsquossmartphones Section 3 explains the most relevant electro-magnetic parameters to characterize antennas for wirelesshandheld devices such as radiation efficiency impedancemismatch signal branch correlation diversity gain MIMOcapacity Total Radiated Power (TRP) Specic AbsorptionRate (SAR) Total Isotropic Sensitivity (TIS) or Total RadiatedSensitivity Average Fading Sensitivity (AFS) and Data bitthroughput (TPUT) In addition Section 3 shows how theseparameters can bemeasured in a reverberation chamber Sec-tion 4 summarizes recent advances in the eld of antennas forwireless handheld devices In particular Section 4 describesantenna technology for designing antennas at low frequenciessuch as FM for short-range wireless applications and nallyfor mobile communications For this last section severalantenna design techniques are explained such as coupledmonopoles and PIFAs combined with slots In addition atechnique robust to human loading is presented based on anarray of small monopoles Section 4 further discloses the useof broadband matching networks to enhance the bandwidthof an antenna element in order to increase the number ofoperating bands It is also focused on techniques to add intel-ligence in the ground plane for enhancing bandwidth andefficiency Finally a novel antenna technology based on smallnonresonant ground plane boosters is described e pro-posal is focused on exciting the groundplane radiationmodesthat the inherent ground plane of any handset platformperforms at mobile frequencies is technology removes theneed of including large antenna elements featuring quarter-wavelength dimensions thus enabling the integration ofmultiple antenna elements and multiple functionalities andservices in the wireless platform

2 Evolution of HandheldMobile Telephones

e evolution of handheld mobile telephones throughouthistory has been captivating e rst telephone call using ahandheld device dates back to the 1970s [1] Since the 1980shandheld telephone devices have become a commodity foreveryone and the mobile market has not stopped expandingsince then e exponential increase in the number ofsubscribers pushes research and development in wirelesscommunication to deliver technologies capable of accommo-dating that growthese technologies have evolved to a greatextent and have included going from analog to digital andgoing from using one frequency band to multiple frequencybands as well as many others is constant evolution ledto the recent deployment of the latest generation radiosonto the consumer market the Long-Term Evolution (LTE)technology

Operators of consumer wireless handheld devicesrecently started to deploy the LTE wireless technology forthe next-generation smartphones Before going through thechallenges engineers have to face in developing antennasfor LTE-capable phones it is important to look at theprevious generations of mobile handheld devices to describethe general challenges in antenna design some of whichstill remain in the current design challenges e following

sections talk about the challenges that exist in antenna designfor wireless mobile handsets

21 Wireless Mobile Generations e rst generation (1G)wireless communication technology was introduced back inthe early 1980s It used an analog standard A few com-mercially used 1G standards included NMT (Nordic MobileTelephone) and AMPS (Advanced Mobile Phone System)NMT network rst used a frequency band in the 450-MHzregion called NMT-450 Due to the subscribersrsquo demandit expanded its network to the 900-MHz region (NMT-900) since it could carry more channels at that frequencyband than its previous band e AMPS standard used inthe United States was deployed in the 800-MHz frequencyregion e subsequent generation radios for example 2G3G and 4G started in the 1990s ese newer generationswere drastically different in the sense that they were all usingdigital standards ere were many advantages to replacinganalog with digital standards One of the advantages is thatdigital standards could accommodate more users which wasnecessary

Even though the 2G standards such as GSM D-AMPSand CDMAOne have been superseded by their newer gener-ations they still remain widely used networks in all the partsof the world e third generation (3G) network appearedon the market in early 2000 and the latest LTE networkwas offered in 2010 ese later standards were tailoredto improve data services e following sections describewhat inuences antenna design and what challenges antennaengineers have to face in the development of mobile handsetdevices

22 Increase in the Number of Frequency Bands Wirelesscommunication standards sometimes come with a newset of frequency bands Fortunately some bands of newergenerations overlap previous generations which releasessome of the burden on the antenna design when a newgeneration standard comes into the picture Looking backfrom the rst generation to the current generation thenumber of frequency bands kept increasing Antennas forthe rst generation handheld devices were designed backin the 1980s to work in one frequency band As the num-ber of frequency bands increased with newer generationsthe need for multiband antenna designs became necessaryFurthermore as the mobile market became more and morepopular and global travel became more accessible to thegeneral population there was a need for making deviceswith roaming capability is was necessary in order forsubscribers of one market region to be able to use the samedevice in other regions with similar standards but differentoperating frequency bands As an example of todayrsquos USmobile devices a phone operating in North America has themain bands operating from 824MHz to 894MHz and from1850MHz to 1990MHz for both GSM (2G) and UMTS (3G)standards Furthermore an additional band is now neededfor the LTE standards in the 700-MHz band e phonewould generally have roaming capability at operating bandsused in the rest of the world precisely GSM 900 GSM 1800

International Journal of Antennas and Propagation 3

F 1 Mobile handheld phone examples through all the gener-ations

UMTS B I and B VIII e frequency band of coverage ofthese roaming bands are from 880MHz to 960MHz (GSM900 andUMTSBVIII) 1710MHz to 1880MHz (GSM1800)and 1920MHz to 2170MHz erefore there is a need fordesigningmultiband antennas that can operate in these bandswith good performance

2 ndustrial esign nuene on ntennas For some peo-ple a mobile telephone handset is a device that serves only asaway of communication and they do not carewhether it is bigor small thin or thick shiny or mat For some other peopleindustrial design is an important factor when it comes tousing consumer electronics devicese look and feel of theirphones are important factors in making their purchasingdecision

Prior to the early 2000s antennas in mobile handsetswere designed externally ey were mostly monopole-typeretractable or not or helical stub antennas protruding fromthe top of the phones (Figure 1) Industrial design did nothave much inuence or impose great limitations for antennadesign In the early 2000s antenna design formobile handsetscompletely changed its course and internal design becamethe next design evolution as it was very appealing in termsof industrial design However new design challenges startedto haunt engineers from many disciplines including RFaudio and of course antenna engineers As expected theintegration of antennas inside the phone created interferenceand noise problems that had to be controlled Furthermoreantenna design was now limited within the shape of thephone Nonetheless these challenges were surpassed with thehelp of new technologies and the fantastic creative mind ofantenna engineers

24 Requirements and Regulations Requirements are animportant part of mobile handset designs Operators rely ontheir sets of specications to make sure that the phones theysell work well in their network Phone manufacturers haveto make sure that they meet operatorrsquos requirements Up tothe 3rd generation wireless standards antenna performanceonly was measured by two quantities TRP and TIS TRP is ameasure of how much power is radiated by the antenna when

F 2 A model of the specic anthropomorphic mannequin(SAM) head

it is connected to a transmitter TIS is dened as a measureof the smallest power that can be input to the receiver so thatthe receiver can still maintain a reliable communication linkFor example the communication link reliability for the GSMstandard is dened using a bit-error-rate (BR) level at 2

Operatorrsquos requirements have evolved over the yearsere are several reasons why this evolution occurred Ulti-mately operators as well as phone manufacturers know thatthe phones need to perform well under the real environmentcondition of the user holding the phone against his or herhead However it is not possible for operators to rely onperformance measurement from phone manufacturers usinga human head and hand grip of a real person as each personrsquoshead and hold would differ from one to another A focusgroup was needed to investigate on how to come up with astandardized model of a human head and hand One suchorganization is the CTIAmdashe Wireless Association [2] Asubgroup in this organization was created to come up witha set of a standardized head and hand for the purpose ofobtaining consistent and reliable performance measurementin a controlled lab environment While this work was understudy operators had to rely on measured TRP and TIS in afree-space condition

e phantom head model called SAM (Specic Anthro-pomorphicMannequin) was rst introduced in 2002 (Figure2) e material inside the plastic shell has specic electricalproperties that is dielectric constant and conductivity thatare modeled closely to the real human head Modeling thehand was more dicult and it took longer to get to the nalset of phantom hands (Figure 3)

Operators from around the world had different require-ments and when they decided to adopt new measurementconditions for their requirements it was not at the same timeAntenna designers had to face the challenge of designingantennas with performance that had to meet various oper-atorsrsquo requirements with different environment conditions


F 3 Examples of phantom hand models

In certain phone designs where the antenna is externalor when there is enough antenna volume for the internalantenna design it is not a problem to meet all operatorsrsquorequirements When the design is limited due to industrialor mechanical designs then antenna variants for differentmarkets are needed each one of them meeting the operatorrsquosrequirements of their market while the over-the-air (OTA)performance in the roamingmarket can be relaxed a little bit

A good example of antenna design change due toa requirement modication is when the operator ATTchanged the cellular antenna requirements from free-spaceto talk position (with the phone placed against the phantomhead) In order to come up with an attractive handset designand still meet operatorrsquos OTA performance requirements andother regulations Motorola came out with a thin phonewith the cellular antenna in the bottom of the phonePlacement of the antenna in the bottom of the phone allowedthem to design a thin form-factor handset and still meetthe operatorrsquos requirement with great performance in thelow band (824ndash894MHz) at year marked the change inantenna location in antenna design

Requirements are specic to operators Handset manu-facturers must also meet the broadcasting and RF emissionregulations that are specic to countries For example theFederal Communication Commission (FCC) [3] has dutiesof regulating RF emissions in the United States A fewregulations pertaining to mobile phone radiated emissionand antennas include SAR (Specic Absorption Rate) com-pliancy HAC (Hearing Aid compliancy) and GPS E911requirements

SAR relates to the near E-eld effects of the antennas(Figure 4) FCC regulations mandate that all phones usedin the United States must meet a SAR limit of 16WKgaveraged over a volume of 1 gram of tissue [4] In someother countries the SAR limit is 2Wkg averaged over avolume of 10 grams of tissue [4] e SAR requirementcan be a show stopper for phone manufacturers ey mustmeet the regulations or else the phones cannot be releasedto the market Antenna designers have to make sure thatsuch regulations are met One way to reduce the SAR valueis to decrease radiated power is is done by reducing thetransmit power or detuning the antenna impedance so thatantenna performance is degraded However this techniqueof SAR reduction would impact the OTA performance andmay cause a failure to meet the operatorsrsquo OTA requirements

F 4 SAR measurement system e wireless handheld deviceradiating RF power is attached to a phantom cheek A probe mea-sures the electrical eld generated by the device inside a phantomlled with liquids emulating the human tissue at the frequencies ofinterest

Fortunately there are other techniques e general idea is toreduce the E-eld towards the head One example that helpsreduce SAR in the low band (850MHz band) is moving theantenna located on the top of the phone to the bottom Aphone with good OTA performance and a thin form factordesign would have a very difficult time to meet the SAR limitif the cellular antennawas placed on the top of the phoneisis another important factor of the antenna location

Regulations in the United States for interference withhearing aid devices due to wireless mobile handsets wereimposed on phonemanufacturers and operators around 2006[5] ere are two kinds of interference related to HAC T-Coil and RF emission Interference due to T-coil is takencare of by acoustics engineers and relates to the couplingeffect between the coil in the handset earspeaker and thatof the hearing aid Antenna engineers have to deal with theRF emission interference precisely the near E- and H-eldsemanating from the cellular antenna around the earspeaker ofthe phone ese elds are measured within a 5 cm by 5 cmsquared area centered 15mm above the phone earspeaker[6] ey are required to be below a certain strength level inorder to be compliant Just like the SAR problem antennaengineers have to nd ways to reduce the near elds aroundthe earspeaker without affecting the OTA performance of thephones

Another antenna challenge relating to regulations per-tains to the Enhanced 911 or E911 is mandate from theFCCorganizationwas created to assure that when calling 911for an emergency the user can be geographically located witha certain amount of accuracy within 30 seconds aer dialing911 in the United States In order to locate a user this fast astandaloneGPS system is not enoughe systemneeds someassistance from the network to acquire the required locationaccuracy within a small amount of time is system is calledassisted-GPS (aGPS) Regardless of whether the system isstandalone or assisted the most important parameter in thesystem is antenna performance e GPS antenna has to bedesigned in such a way that under the use-case condition


its radiation pattern has a good coverage of the sky wherethe GPS satellites are Even though the aGPS system generallyworks with ease under the open-sky environment that is noobstruction between the sky and the system the differencebetween good and bad antenna design can be seen when itcomes to testing it in the urban and indoor environments

ere are other requirements and regulations specicto operators and countries but the ones just previouslydescribed are the challenging ones that antenna engineershave to deal with during the concept design phase anddevelopment of antenna systems in wireless mobile handsets

25 Antenna Design Challenges in Todayrsquos Smartphones eprevious sections highlighted some challenges that antennaengineers have had to face in antenna design for phonesese challenges are not getting easier in todayrsquos mobilehandheld devices Smartphones are becoming a universaldevice that subscribers want to have ese devices arepacked with a great amount of applications ey are nolonger just a simple phone Examples of such applicationsinclude data communications such as internet browsingmovie streaming email access navigation system remotecontrol geotagging in photoshooting and a payment systemAll these applications need the use of an antenna built in thesmartphone whether it is a cellular Bluetooth WiFi GPSNFC or FM antenna e obvious challenge is to design allthe necessary antennas inside a compact device Placementof these antennas is crucial to the design as coupling betweenantennas needs to be minimized Another design challengein compact devices is the additional NFC antenna neededfor near-eld communication such as the payment systemFeliCa in Japanis NFC antenna is conventionally made ofa coil resonating at 1356MHz e coil is generally designedon a ferrite sheet to minimize Eddie current created by thecoil on any metal surface underneath the NFC antennaBig coils and ferrite materials can oen cause performancedegradation in other antennas located nearby which cancomplicate other antenna designs as space can become morelimited

As mentioned in the previous section creating a con-trolled environment for testing over-the-air performance inlabs needs to be close to the real use-case environmentOperators are now starting to adopt and create requirementsfor OTA phone performance testing in the talk positionincluding the phantom hand (Figure 5) Even though onehand-grip testing does not represent the entire spectrum ofhand grips from real users it is still one step closer to cap-turing performance effects of a real use-case condition isnew requirement forces antenna engineers to pay attention tothe effect of the hand on the antenna performance so that asystem can be designed to satisfy the requirement

e next-generation smartphones that are LTE-capablefurther increase the level of challenge involved in antennadesign For an LTE system in phones a second antenna forreceive diversity is needed along with a primary cellularantenna Both antennas are operating in the same frequencyband at is an additional antenna to design in a smalldevice that is already populated with multiple antennas For

F 5 Over-the-air phone testing in the talk position includingthe phantom hand

MIMO design antenna efficiency antenna isolation gainimbalance and correlation between the two antennas areimportant parameters in designing antennas for the LTEsystems In MIMO systems optimal system gain is obtainedif the two antennas are totally uncorrelated have similargain performance and are uncoupled Increasing the antennaspace can help reach optimal performance However inphone design space is limited Fortunately operator require-ments can tolerate the secondary receive antenna having anantenna efficiency level of about 3 to 6 dB below that of theprimary is is helpful for the secondary antenna designas its efficiency does not have to be as good as that of theprimary antenna However isolation and correlation remainthe challenging tasks to work on

Correlation is mostly dependent upon the far-eldantenna pattern Radiation pattern characteristics at frequen-cies of 1500MHz and higher are generally dependent uponthe antenna location is means that at higher frequencythe radiation patterns of the two antennas can be verydifferent with enough distance separation between themand therefore it is generally not an issue in meeting theoperatorsrsquo requirements at LTE bands higher than 1500MHze challenge still remains for LTE bands at frequenciesbelow 1000MHz is is because the radiation patternsat these frequencies have somewhat similar characteristicsno matter where the antennas are placed within the realestate of the phone design e reason to this similarity incharacteristics is because the PCB or ground of the antennais the main radiating element at frequency below 1000MHzfor a typical phone length Operators target an envelopecorrelation coefficient (ECC) of 05 as their requirement

Isolation is also a challenge in smartphones at frequenciesbelow 1000MHz due to antenna small electrical distanceseparation If not designed well the overall efficiency of bothantennas can degrade dramatically and instead of designinga system that gives additional processing diversity gainperformance one can end up with a system that has a similaror worse performance to a conventional system with oneantenna

LTE systems are data centric At this stage voice is notsupported on the LTE network Voice-over-LTE (VoLTE) isstill in the test phase and is not yet deployed ereforethere is no simultaneous data communication over LTE andvoice communication For a 3G UMTS smartphone that has


additional LTE bands simultaneous data and voice can onlybe done in 3G So if a phone call is received and a useranswers during a data connection over the LTE network thendata connection has to fall back to a slower speed in the 3GUMTS network One operator that is VerizonWireless in theUnited States takes it one step further to have a design that iscapable of having simultaneous voice in the CDMA networkand data communication over their LTE network e reasonfor this design is that their CDMA network does not allowsimultaneous voice and data communication One antennais designed for voice in the 850- and 1900-MHz bands andfor the receive diversity for the LTE band at 750MHz eother antenna is designed to be the primary transmitreceiveantenna for data communication at the LTE band and EVDOCDMA bands is is a complex and challenging systemto design for a smartphone and to meet not only all theoperatorsrsquo OTA requirements but also the SAR limit forsimultaneous transmission which is still at 16Wkg averageover 1 gram of tissue

An overview of the challenges and issues antenna engi-neers have to face during the concept and developmentphases of wireless mobile handsets was discussed From thebeginning of the history of mobile phones the challengelevel for designing antennas has never decreased ere hasalways been a constant increase in the number of challengesfrom one generation to the next In the midst of all thisthe extraordinary creativity of the antenna designers hashelped them overcome all the challenges that have led fromthe design of a wireless mobile device with a large externalsingle-band antenna design to a small and slim device withmultiband and multiantenna systems

3 Verifying Designed PerformanceHandset Antenna Characterization

Antenna characterization has experienced a rapid devel-opment through the last couple of decades and a largepart of antenna measurement development has been causedby the introduction of handset antennas For traditionalantennas such as those used for radars point-to-point linksor macrocell base stations the radiation pattern is of greatimportanceose types of antennas are specically designedto direct energy in a certain direction and avoid spillingenergy in other directions

Handset types of antennas are by nature electrically smallwhich means that they exhibit more or less omnidirectionalradiation patterns due to the small size of the radiatingelementis is not necessarily a bad thing handset antennasare used in an arbitrary orientation with signals arriving tothe device from arbitrary directions and there is a benetin collecting as much of this energy as possible ereforedesigning handset antennas towards a specic radiationpattern is of less interest Hence the parameters used tocharacterize handset antennas have somewhat different focusthan those used for the traditional types of antennas men-tioned above

Another shi in antenna characterization is ongoing rightnow is shi is caused by the introduction of multielement

antennas which are used to facilitate antenna diversity orMIMO communication Still the same basic characteristicsas for single-element handset antennas are important butthese are complemented with additional parameters to val-idate the antennas functionality in the modern communica-tion system

31 Figure-of-Merits forWireless Handheld Devices ere areseveral gure-of-merits (FOMs) which are interesting forcharacterization of wireless handheld devices

e FOMs can be divided into passive and active param-eters where the former are antenna only parameters and thelatter include radio circuitry is division reects anotherfundamental difference between the two groups of FOMswhich is that the passive antenna parameters are componentvalues whereas the active parameters are composite valuescombining performance of several components into a singlevalue

311 Passive Antenna Parameters Commonly used passiveantenna parameters are

(a) radiation efficiency [7](b) impedance mismatch [7](c) signal branch correlation [8](d) diversity gain [8](e) MIMO capacity [8]

e rst two are traditional antenna parameters applica-ble to all types of small antennas whereas the latter three arerelevant for multielement antennas (MEAs) is does notmean that the two former parameters are less important forMEAs On the contrary radiation efficiency is still the mostimportant design parameter for electrically small antennas

Radiation efficiency of an antenna is basically the ratio ofpower radiated from the antenna to the delivered power tothe antenna feed which means that it is a description of theinternal losses of the antenna element is means that theradiation efficiency goes directly into the link budget of thecommunication system and therefore has a direct impact onthe performance of the system

Radiation efficiency is oen paired with impedance mis-match as the most useful design parameters for antennas inwireless handheld devices Total radiation efficiency (some-times also called antenna efficiency) is a combination of thesetwo dened as the product of the radiation efficiency and theefficiency due to mismatch

It is applicable to talk about radiation efficiency alsoin the case of MEAs e most proper way to characterizethe efficiency of each element of an MEA is to look at itsperformance when the other elements are present in orderto fully account for loss due to mutual coupling betweenelements Such radiation efficiency that accounts for mutualcoupling can be referred to as Embedded Element Efficiencywhere the embedded prex denotes the presence of othernearby antenna elements

Signal branch correlation is applicable to antennas withtwo or more branches and is a measure of how uncoupled the


antenna elements are It is calculated as the cross correlationbetween the signals received on two separate antenna portse signal branch correlation as well as radiation efficiencyand impedance mismatch is example of component param-eters that is parameters directly showing the performance ofa certain part of the communication system

Diversity gain and MIMO capacity the two latter pas-sive parameters in the list above are actually compositeparameters determined by the rst three passive antennaparameters just mentioned radiation efficiency mismatchand correlation In the literature there are a few denitionsof diversity gain to be found and it is important to applythese denitions in a correct way in order to draw justiedconclusions from a set of data e basic difference betweendifferent diversity gain denitions is how the radiationefficiency is embedded in the parameter e three basicdenitions of diversity gain are Apparent Diversity GainEffective Diversity Gain and Actual Diversity Gain wherethe difference is the reference fromwhich the diversity gain iscalculated [8] e reference can either be one of the diversitybranches (Apparent Diversity Gain) an ideal single referenceantenna (Effective Diversity Gain) or any practical antennato be replaced (Actual Diversity Gain)

Note that the passive parameters discussed here areintegral quantities based on the assumption of a statisticallyisotropic multipath environment surrounding the antennais type of environment is especially useful for handsetantenna characterization not only due to the similarity tothe environment where most handsets are used but also dueto that a handset is arbitrarily oriented due to individualpreferences of the users is environment can be referred toas Rich Isotropic MultiPath environment (RIMP) [9]

In some cases there is interest in creating the integratedparameters over other types of spatial distributions Anexample of this is the Mean Effective Gain parameter whichcan be described as radiation efficiency weighted with respectto a certain angular distribution of incoming waves to theantenna under test [10]

An extreme in the sense of spatial distributions is thepure Line-of-Sight environment where there is a singlesignal component arriving at the antenna under test isis the direct opposite of the RIMP environment mentionedabove meaning that these two environments complementeach other e difference between these two environmentsis how they impact amultiantenna system such as diversity orMIMO An example of a LOS parameter is the LOS diversitygain [11]

312 Active Antenna Parameters Commonly used activeantenna parameters are

(f) Total Radiated Power (TRP) [7](g) Specic Absorption Rate (SAR)(h) Total Isotropic Sensitivity (TIS) or Total Radiated

Sensitivity (TRS)(i) Average Fading Sensitivity (AFS) [12](j) Data bit throughput (TPUT) [13]

Device under test

F 6 Anechoic chamber having a gate with 31 probes toelectricallymeasure the radiation in one plane radiated by the deviceunder test e device under test is rotated so as to have the full 3Ddata

e three rst parameters of active antenna parameterslisted above can at this point all be considered traditionalcharacterization parameters for wireless devices Both TRPand TIS can be directly related to the total radiation efficiencyof the device antenna and are therefore commonly usedparameters to characterize the radiation efficiency of deviceswithout a direct external cable connection to its antenna SARis a bit different from other antenna parameters described inthis section of the paper since it is not a pure over-the-airparameter but a measure of the absorption rate of power insimulated human brain tissue

TIS is originally a single antenna parameter but it ispossible when measuring TIS in a multipath scatteringenvironment as the reverberation chamber to extend themeasurement to include multielement antenna performanceat is exactly the same measurement procedure as used forsingle element TIS will include the performance improve-ment offered by the multielement implementation as long asthe measurement is performed in a multipath scattering andwith the multiple signal combination activated in the device

e last parameter data bit throughput has attractedconsiderable interest in MIMO-OTA discussions in theantenna community over the past few years mainly becauseof its close link to end-user experience e basic principlebehind this type of throughput measurement is to create ascattering environment in which the unit experiences fadingand sample the data throughput over time to get a statisticalvalue of what data bit rate the unit can support given acertain average available power e measurement chamberneeds in this case to work as a spatial channel emulatorand there are several ways of achieving this either withexistingmeasurement setup (like reverberation chambers) ormodications of existing chambers (like anechoic chambers)

Data bit throughput is essentially equal to an error ratemeasurement taken over a fading sequence whether it is bit


Mode stirrers

(moves during measurement)

Turntable


Calibration antenna

Walls of reflective

material

Test object (DUT)

(USB modem on laptop)

Access panel

3 x fixed measurement

antennas with different

polarization connected

to a network analyzer

or a communication

tester

F 7 Reverberation chamber congured for measurements of antennas for wireless handheld devices

error rate (BER) packet error rate (PER) frame error rate(FER) or block error rate (BLER) e process of samplingthe error rate specically during a fading sequence hasbeen referred to as Average Fading Sensitivity (AFS) andis then very similar to how data throughput measurementsare performed today It is interesting to note that there is arelationship between the AFS and TIS value of a device

32 Measurement Methods of Antennas for Wireless HandheldDevices ere are two dominating range types for smallantenna measurements anechoic chambers and reverbera-tion chambers Although many of the parameters accessiblethrough measurements in these two chamber types areidentical the methods themselves work in diametricallyopposite ways

In an anechoic chamber everything but the direct signalfrom the measurement antenna to the antenna under test isremoved hence the name of anechoic chamber no echoesexist in the measurement setup To measure any integralparameter the antenna under test is rotated to cover all dif-ferent angles of arrival at the antennae integral parametersdescribed above are then calculated from the informationgiven in each angular direction (Figure 6)

e reverberation chamber on the other hand is fullyreective and creates a eld with many angles of arrivalpresent at the same time that is a lot of echoes but nodirect signal path As the so-called mode stirrers are movedsignals will combine in different ways and over a full stirringsequence all angles of arrival will be equally probable Hencethe integral parameters described above can be extracted asa direct result of a measurement sequence Figure 7 shows anexample of how a reverberation chamber looks like

Figure 8 shows the schematic setup for anechoic andreverberation chamber measurements respectively Notethat the instrumentation is similar between the two methods

With the current trend of creating fading channels to testhandset antennas there is much work ongoing to modifythe anechoic chamber to facilitate multipath fading in the

originally pure LOS environment e proposed methodmeans placing a ring or sphere of probes in the anechoicchamber and feed signals through these antennas so thata specic fading prole is created in the center of the testvolume e drawback with this modication is that thechamber has to be converted back to a normal anechoicchamber that is removing the additional probes beforetraditional antenna parameters can be measured so most ofMIMO-enabled anechoic chambers are likely to be dedicatedto MIMO testing only

Reverberation chambers have an inherent multipathfading due to its reective nature and therefore MIMOOTA measurements can be performed without any othermodications than adding xed measurement antennas tofacilitate the MIMO signaling

Figure 9 shows the schematic setups for MIMO OTAmeasurements in reverberation and modied anechoicchambers Note that both measurement setups are equippedwith channel emulator to control the fading In modiedanechoic chamber the channel emulator is essential in orderto create the fading and it is done by feeding prefaded signalson each of the probes in the chamber In the reverberationchamber the channel emulator is optional due to its inherentfading but the channel emulator gives a wider range ofpossible power delay proles in the measurement setup

Table 1 shows a compilation of the differentmeasurementmethods and which gures of merit used for design of smallantennas are applicable for each method

4 Antenna Technology forWirelessHandheld Devices

e massive incorporation of wireless handheld devices suchas mobile phones in our lives has changed their functionalityconception Nowadays mobile phones are not only used tocommunicate but they also offer a big range of servicessuch as digital camera video player internet connectivitygeolocalization TV services or FM radio In this regard


Anechoic chamber

DUT

Absorbers

Measurement antenna

VNA

(a)

Reverberation chamber

DUT

Mode stirrer

Fixed antenna

VNA

(b)

F 8 Example measurement setups for passive (cable-fed) testing of antennas for wireless handheld devices For active device testing theDUT is replaced by a functional handset and the vector network analyzer (VNA) is replaced by a base station simulator

Modified anechoic chamber

DUT

Absorbers

Measurement antennas

Channel

emulator Base station

(a)


MIMO link

DUT

Mode stirrer

Channel


Fixed antennas

(b)

F 9 Example measurement setups for active MIMO testing of antennas for wireless handheld devices

antenna industry as well as academic areas are being forcedto evolve constantly to obtain small and multiband antennascapable of radiating efficiently in such a hostile environmentOn one hand the volume constraints in wireless handhelddevices produced by the reduction of the available space dueto the existence of multiple components (such as displaysbatteries speakers and shieldings) must be considered foroptimizing the antenna performance On the other handuser interaction also needs to be taken into account fromtwo perspectives Firstly the amount of power absorbed bythe human body especially the head and hand has to beminimized Secondly the antenna needs to be robust to such

human interaction which causes power absorption andordetuning effects Minimizing power losses is an importantaspect since they produce higher battery consumption andeventually call drops

With the objective of reviewing several antenna applica-tions that can be found in current or emergent wireless hand-held devices this section is divided into three main partsFirstly antennas for reception applications are discussed inparticular for FM reception (88ndash108MHz) Secondly a briefdiscussion on antennas for short-range wireless applicationsis presented and nally a summary of some advances in theeld of handset antennas is disclosed


T 1 Measurement methods for characterization of antennas for wireless handheld devices and applicable gures of merit for respectivemethod

FOM Table ref Reverberation chamber Anechoic chamber Multi-probe MIMO setupin anechoic chamber

Radiation efficiency a Yes Yes NoImpedance mismatch b Yes Yes No

Signal branch correlation cYes calculated direct from

received signalsYes calculated fromradiation patterns

Yes calculated direct fromreceived signals

Diversity gain dYes direct from received

signal distributionsYes calculated fromradiation patterns

Yes direct from receivedsignal distributions

MIMO capacity e Yes from received signalstatistics Yes from radiation patterns Yes from received signal

statisticsTRP f Yes Yes NoSAR g No No NoTIS or TRS h Yes Yes No

TISTRS including diversity reception h YesNo no multipath fading in

anechoic chamber No

Average fading sensitivity (AFS) i YesNo no multipath fading in

anechoic chamber Yes

Data bit throughput (TPUT) j Yes No no multipath fading inanechoic chamber Yes

41 Broadcast Antennas FM e main challenge of design-ing antennas for providing operation in the FM servicemainly relies on size limitations Regarding the FM servicea conventional monopole antenna (1205821205824) operating at FMfrequencies is 75 cm length which is too long for beingintegrated in a handset phone In order to overcome thislimitation some mobile phone manufacturers incorporatethe FM antenna in the wire of the headsets but this solutiongoes against having a fully integrated wireless handhelddevice Other solutions found in the literature propose theuse of active schemes [13] thus resulting in an undesiredincrement of the battery consumption In order to solvethe aforementioned shortcomings this section explains twotechniques for designing internal antennas at the FM bandbased on

(i) nonresonant elements [14ndash16](ii) reusing a PIFA antenna operating at mobile commu-

nication services [17 18]

411 Nonresonant Elements eauthors of [15 16] describethe problem of designing a resonant antenna such as a spiralat the FM band taking into account the reduced space ofa PCB (Printed Circuit Board) Since the available space islimited coupling between antenna tips forces the need ofincreasing the total length in order to attain the desiredresonance thus resulting in a length larger than 1205821205824 Forexample to attain resonance at 100MHz in a 40mm times20mm times 5mm antenna volume a length of 2262mm isneeded which becomes larger than a quarter of a wavelengthat this operating frequency (1205821205824 = 750mm) [16] Moreoverdue to the aforementioned volume constraints the width ofthe antenna has to be thin Such constraint in the design

width can considerably increase ohmic losses thus producinga poor radiation In order to solve these limitations theproposed idea substitutes a resonant antenna by a nonreso-nant antenna inspired in the Hilbert geometry with a high-Qinductive element that brings the antenna to resonance Withthis approach better efficiency is obtained (around 20 dBmore) Although the efficiency for the nonresonant elementis around 1 this result is still acceptable for FM receptionfor two reasons First the transmit power for FM broadcasttower is in the order of KW Second the free-space loss forFM is not as critical as other telecommunication servicessuch as cellular communications (GSM) for example at100MHz the free-space loss is approximately 20 dB less thanat 900MHz As a result more power is available in theair With this condition a small compact antenna for FMreception inspired in the fractal geometry of theHilbert curveis proposed which becomes suitable for being integratedin current wireless handheld devices thanks to its reduceddimensions of just 30mm times 10mm times 1mm (Figure 10)

Besides the common electromagnetic parameters suchas SWR (Standing Wave Ratio) radiation patterns andefficiency another gure of merit is proposed to evaluatethe performance of antennas for FM reception It consistsof demodulating the RF signal to an audio signal isprocedure is presented in Section 412 where the perfor-mance of the proposed Hilbert antenna is compared to theperformance of a 1205821205824 monopole concluding that the Hilbertsolution offers a similar audio quality of the received signalwith the advantage of its reduced size and its integrationcapabilities

412 Reusing a Mobile Antenna is section introduces asolution for integrating an FM receiver antenna in a wireless


Zoom

Zoom times 3

F 10 External wire (75 cm length) and internal FM Chip Hilbert antennas (30mm times 10mm) integrated within a typical smartphoneplatform [19ndash21]

Feeding port

Shorting port

(a)

FM port

GSM port

Filter Switch

to the PIFArsquos

feeding port

to the PIFArsquos

shorting portL1

(b)

F 11 (a) 3D view of the PIFA Ground plane size is 100mm times 40mm and PIFA is 38mm times 15mm times 6mm (b) proposed matchingnetwork including a switching circuit a lter and a series inductor

handheld device that goes one step beyond e proposedtechnique is focused on reusing an existing antenna operatingat cellular bands In this sense a PIFA (Planar Inverted FAntenna) designed to operate at twoGSM standards (900 and1800MHz) (Figure 11(a)) can be reused to become operativeat the FM band [15] e PIFA behaves as a nonresonant ele-ment at FM frequenciese required 75 cm length needed tobehave as a 1205821205824 monopole is far from the PIFArsquos dimensionserefore a high series inductor is added in order to compen-sate for the capacitive behavior of the PIFA at FM frequencies(Figure 11(b))

e PIFA has a feeding port and a port which short-circuits the antenna with the ground plane In order toguarantee a good response in the FM band the shortingconnection must be removed because the distance betweenports is electrically small at these frequencies producing a

short-circuited antenna with poor electromagnetic perfor-mance at the FM band [18] To guarantee good radiationin the desired frequency bands (FM and GSM9001800) amatching circuit is needed (Figure 11(b))ePIFAused heredoes not need any matching network at GSM frequencies buta 1000 nH series inductor is required at FM Both ports areisolated by means of a lter and the series inductor e lteris designed to only reject the FM signal at the GSM portbecause the GSM signal in the FM port is already rejecteddue to the series inductor that presents high impedance atGSM frequencies Finally a switching circuit is needed inthe short port in order to disconnect the antenna from theground plane when it is operating at FM band

In [13] it was demonstrated that a high receivedpower does not mean necessarily a better signal quality Insome cases a low received power offers satisfactory audio


0

1

2

3

4

5

6

7

8

9

10

MonopoleHilbertPIFA

7374 72

Sign

al quality

F 12 Marks obtained through the quality evaluation aeraveraging 28 FM channels

reception whereas a high received power presents low SNR(Signal-to-Noise Ratio) leading to a decrement of the qualityaudio reception For this reason a subjective procedure [19]for evaluating the demodulated signal quality has been car-ried out regarding the PIFA the 75 cm length monopole aswell as the previous fractal-inspired Hilbert-based monopole[20 21]

is procedure consists in quantifying the quality of theFM signal received by the antenna being tested e signalquality indicator is ranked from 0 to 10 depending on thequality of the FM channel heard by the user [19]

Despite having the highest received power themonopolersquos nal evaluation does not differ from theother ones e nal mark for the 1205821205824 monopole is 74 thenal mark for the Hilbert antenna is 73 and nally thePIFArsquos mark is 72 (Figure 12) having the advantage thatthis antenna can also operate in the mobile communicationbands

It is interesting to outline that human body has been alsotaken into account concluding that in some position suchas holding the device with the hand the overall efficiency isimproved by 10 dB [22 23] is improvement is due to thefact that at this low frequency ranges the human body acts asa dielectric antenna with a size comparable to the wavelengthof operation thus becoming an efficient radiator (a humanbody of 17m at 100MHz is 056120582120582)

In conclusion the PIFA offers the same satisfactoryperformance as the reference monopole and it ensures theintegration of the FM antenna in wireless handheld devicesMoreover other handset antenna techniques such as theslotted ground planes (as described in the following sections)can be used in combination with the PIFA to obtain aheptaband antenna (FM GSM 85090018001900 UMTSand BluetoothWi-Fi)

One of themajor advantages of the proposed technique isthat no extra antenna is needed because the existing mobileantenna is reused

42 Short-Range Wireless Short-range wireless generallyrefers to those applications characterized in that they have

F 13 Current vector distribution of the antennas at theresonance frequency of 119891119891 = 845MHz

small transmitted power (order of mW) indoor operationrange of meters and limited bandwidth (about 4 for Blue-tooth application) Examples of short-range wireless systemsare Bluetooth WiFi ZigBee and RFID e vast majority ofwireless handheld devices incorporate a short-range wirelessantenna for BluetoothWLAN services Antenna size is againan important aspect to consider since the center frequencyof operation for Bluetooth is 245GHz meaning that a1205821205824 antenna is 30mm Such antenna size is still largeconsidering the devicersquos space limitation due to displaysbatteries speakers as well as the need of integrating othermultiple antennas such as the ones intended for mobilecommunication erefore the challenge relies on makingthe antenna as small as possible to simplify its integration in awireless handheld devicewhile preserving its electromagneticperformance

In order to face the challenge of antenna miniaturiza-tion for short-range wireless applications two categoriesdescribed extensively in the literature are proposed

(i) geometry based

(ii) material based

On one hand geometry-based antenna relies on design-ing antenna geometries capable of taking the maximumprot of the available space An example is found in space-lling geometries [24ndash36] On the other handmaterial-basedantennas are focused on using high dielectric materials suchas ceramics capable of providing the requiredminiaturization[37]

e suitability of space-lling geometries in the designof small antennas has been broadly investigated In thiscase small antennas like the Hilbert monopole are describedextensively in the literature [24ndash36] to demonstrate thatan antenna can become electrically smaller as the iterationincreases Using this type of miniaturization technique itis possible to reduce the electrical size of a conventionalquarter-wave monopole up to a factor of 11 [24]

To analyze the benets of the Hilbert curve in designingsmall antennas a comparison with a spiral antenna is carriedout [31 36] (Figure 13) Two antennas are designed toresonate at the same frequency of 845MHz occupying thesame footprint and having the same wire width Althoughthe spiral needs less wire for resonating at 845MHz the


Microstrip line

Clearance area

F 14 SMD space-lling-based antenna for 24-25GHz appli-cations Antenna is 41mm times 2mm times 1mm (41mm is 0033120582120582 at245GHz)

0

10

20

30

40

50

60

70

80

90

100

1

15

2

25

3

35

4

45

5

55

6

2 21 22 23 24 25 26 27 28 29 3

To

tal

effi

cie

ncy (

)

SW

R

Frequency (GHz)

VSWR

Total efficiency ()

F 15 Measured SWR and total efficiency for the 24-25GHzantenna shown in Figure 14

bandwidth of the Hilbert antenna is 162 larger for the sameradiation efficiency

anks to its miniaturization properties space-llingbased antennas are suitable to make efficient small andmultiband antennas Some examples for short-range wire-less applications (eg wireless headsets cellular handsetsBluetooth USB and serial Dongles) are already adopted inindustry (Figure 14)

Space-lling geometry-based antennas have been provento be efficient radiators showing that not only size and wirelength but also geometry plays a role in the performance of asmall antenna A small antenna featuring 41mm times 2mm times1mm for 24-25GHz operation shows a total efficiencymorethan 50 making it attractive for many wireless handhelddevices (Figure 15)

43 Mobile Communications is section discusses someantenna techniques for mobile communications In the rstpart some antenna types are presented based on monopolesand combination of PIFA (Planar Inverted F Antenna) andslots Second an antenna architecture robust to hand loadingis discussed ird the benet of manipulating the ground

plane is analyzed Fourth a particular matching networkfor enhancing the bandwidth is studied and nally a novelantenna technology based on the use of compact elementsfor exciting the ground plane of wireless handheld device ispresented

431 Radiators Nowadays internal antennas such aspatchPIFAs and monopoles are the most common designsfor handsets [37ndash42] For PIFAs several well-knowntechniques are used to provide dual-band or multibandoperations such as shaping the radiating path or usingslotted ground planes is fact increases the complexityof the design and makes difficult their integration in slimplatforms since to guarantee good performance the PIFAantenna has to be arranged at a certain height with respectto the ground plane hence occupying a considerable volume(asymp4500mm3) Monopole antennas are an alternative designto provide multiband operation in slim platforms mainlydue to its low prole characteristics [43] In this section twokinds of radiators are briey discussed e rst one employsmonopole antennas e mechanism to obtain multibandand enough bandwidth is achieved by a structure based ondriven parasitic elements e second radiator combines aPIFA with a slot to make a modular design in the sense thatthe number of bands is controlled independently from eachradiatorCoupled Monopoles e use of monopole antennas in wire-less handheld devices has increased in the recent years thanksto its low-prole characteristics that simplify their integrationin wireless platforms Many designs have appeared in theliterature and industry with the aim of covering the largestnumber of frequency bands as possible without reducing theantenna performance [44ndash50]

A multiband behavior (GSM85090018001900 andUMTS) is obtained with a technique using parasitic elementscoupled to a primary driven element At the same timethe proposal maximizes the space on the PCB to integrateother cellular components [51 52] e proposed antennahas also a planar prole which is attractive for slim platforms(Figure 16)edriven element is located closer to the groundplane separated at a distance from the parasitic elementseground plane area located at the right side of the antennaprovides a useful space to integrate some typical elements ofthis kind of devices such as a camera or a speaker On theother hand the design takes into account the most criticalvariables when dening the operating frequency rangesese variables are the element lengths and the gap betweenthem which determines their coupling effect Furthermorethe location of the elements determines the correct behaviorespecially at the low frequency bands (GSM850GSM900)

Coupling between the driven and a parasitic elementallows the apparition of an impedance loop in the Smithchart By properly controlling the coupling between bothelements the performance can be wideband or multibandElectrical models can be used to give a physical insightinto the coupling mechanism [52] In this particular casea rst parasitic element is tightly coupled to the drivenelement to obtain two separated bands (Figure 16) Another


Driven

Parasitic 1Parasitic 2

F 16 Prototypes regarding a ground plane of 45mmtimes90mm100mm and 110mm e driven element is fed through a 50Ωtransmission line

parasitic element (parasitic 2 Figure 16) is weakly coupledto the driven to obtain a wideband at the upper regionIt is interesting to outline that similar effect is found inmicrostrip antennas formed by a driven and a parasiticelement erefore the use of electric models is useful tounderstand the behavior of the impedance performance ofantennas

e design features a footprint of 35mm times 15mm and1mm height achieving pentaband behavior for GSM850GSM900 GSM1800 GSM1900 and UMTSCombination of PIFA and Slots PIFA and slots have beenwidely studied in the literature [38 53 54] Basically thePIFA needs a 3D volume to radiate efficiently whereas the slotantenna can be completely at However due to the groundplane the space underneath the antenna cannot be reusedto place other handset components (such as a speaker abattery and shieldings) since they would affect signicantlythe antenna performance In order to combine the benetsof PIFAs and slot antennas (planar structures) a concept thatcombines a PIFA with a slot antenna is discussed here Otherkinds of combination such as monopole and slot antennasusing a self-complementary structure have been proposed in[55]

An illustration on how the concept works is shown next[56 57] Figure 17(a) depicts a slot in a ground plane having100mm times 40mm In this case the slot is excited around1900MHz which results in a 1205821205824 slot antenna e obtainedbandwidth covers GSM1800-UMTS at SWR le 3 Figure17(b) shows a 900MHz PIFA on the same ground plane efeeding mechanism is in the same position used to excitethe previous slot Both designs are combined that is thePIFA and the slot share the same feeding mechanism (Figure17(c)) e antenna combines both reection coefficients(Figure 17(d)) To increase the bandwidth at the second bandslot width may be increased [58]

Since the PIFA has only one branch the space can bereused to allocatemore branches and therefore increasing thenumber of bands [56] For this technique it can be concludedthat

(a) number of bands = number of PIFA bands + numberof slot bands

(b) bands due to the PIFA and the slot can be adjustedindependently

is concept is based on a parallel excitation of a PIFA-slot that becomes particularly useful to design multibandhandset antennas where the number of frequency bandsis given by the sum of the bands given by each radiatorMoreover said bands can be controlled independently whichadds an additional degree of freedom to the design

anks to the slot radiator the PIFA volume can bereused to add more bands With this structure an extraband centered at S-DBM has been added to nally design apentaband prototype including GSM900 1800 1900 UMTSand S-DMB [56] e total antenna volume is 39mm times11mm times 2mm (h) Results for total efficiency taking intoaccount several components (battery display speaker cam-era and phone covers) are satisfactory andmake this conceptattractive for the new generation of low-prole multibandhandset phones

432 Robust Architectures to Hand Loading e challengefor the antenna community is not only to design small-multiband antennas but also make them robust to humaninteraction that is to minimize the radiation toward thehuman body and make the antenna behavior independentfor instance from the hand loading that detunes and absorbsthe radiated power [59ndash62]

Several techniques have appeared in the literature In[63] two strips are located at the edges of the PCB tomake the system robust to hand loading Some schemespropose the compensation of the nger effect by an antennaselectionwhich requires a switchingmechanism that involvesan increment in the battery consumption [64 65]

A technique named distributed antenna system is pre-sented here to provide robustness to the hand-loading effecte technique proposes a handset antenna architecture basedon an array of small monopoles strategically arranged alonga PCB in order to provide robustness to the human loadingeffect and in particular to the nger loading effect (Figure18) [66ndash68]

It is well know from microwave theory that an array ofin-phase radiating elements presents the same return loss atthe input port of the feeding system as the return loss of thesingle element However if a phase delay is introduced forexample to achieve a certain beam tilting the bandwidthmaybe enhanced at the input port due to the nonconstructive sumof all the reections coming from each radiatoris principleof array theory is applied here in order to obtain not onlya broadband antenna but also a more insensitive system tonger loading effect than the one using a single element

e proposed system is completely passive which interms of simplicity and battery consumption is considerablyadvantageous

Electric models have been used to give a physical insighton the broadbanding mechanism of the distributed antennasystems [69]

A prototype having a singlemonopole another prototypecomprising two monopoles and a third one integrating threesmall monopoles combined in a single port are built and


Slot

λ4 band 2

(a)

PIFA

λ4 band 1

(b)

PIFA + slot

(c)

08 09 1 11 12 13 14 15 16 17 18 19 2 21 22 23 24 25

Frequency (GHz)

Refl

ecti

on

co

effi

cie

nt

(d

B)

minus14

minus13

minus12

minus11

minus10

minus9

minus8

minus7

minus6

minus5

minus4

minus3

minus2

minus1

0

DB(|S(11)|)

PIFA band 1

DB(|S(22)|)

Slot band 2

DB(|S(33)|)

PIFA + slot

088 GHz

minus789 dB

096 GHz

minus601 dB 217 GHz

minus639 dB

171 GHz

minus643 dB

(d)

F 17 Sequence showing the antenna concept (a) A slot on the ground plane is tuned at 1900GHz (band 2) (b) PIFA is tuned at900MHz (band 1) (c) parallel ecitation of both antennas (PIFA slot) (d) reection coecient of the antenna system Ground plane is100mm times 40mm for all cases

Phase delay

Printed circuit

board

Antenna 1 Antenna 2

F 18 Illustration of a distributed antenna system having twoelements placed at different locations of a handset device

measured in order to demonstrate the effectiveness of theproposal (Figure 19) [68] e bandwidth (SWR le 3) for thesystem with three monopoles is broader than that attainedby the other prototypes e bandwidth is 156 236and 340 for the single two and three antenna casesrespectively It is worth to note that the three prototypes oper-ate across the GSM850-GSM900 mobiles services Howeverit should be taken into account that the array with three

F 19 Single monopole (le) an array of two monopoles(middle) an array of three monopoles (right) Ground plane is90mmtimes40mmprinted on an FR4 substrate 1mm thickMonopolesare 13mm times 11mm

antennas operates also from 700MHz to 824MHz whereneither the array of two antennas nor the single antennapresent a good reection coecient is is particularly use-ful for providing operation in the emergent communicationstandards such as LTE700

To determine the robustness to human loading a handphantom is used (Figure 20) e hand phantom is lled


(a) (b)

F 20 (a) Common holding position during a call (b) the hand phantom emulating the real situation illustrated in (a)

35

3

13

23

46

10

2021

48

19

25 25

0

5

10

15

20

25

30

35

40

45

50

Free space Left Center Right

Average a

nte

nn

a

effi

cie

ncy (

)

1 monopole

2 monopoles

3 monopoles

F 21 Comparisons of themeasured average antenna efficiency(824ndash960MHz) in free space and regarding hand loading for theproposed antenna systems depicted in Figure 19 and regarding thethree positions of the nger

with liquids emulating the electromagnetic properties of thehuman hand at the frequencies of interest [70] Differentexperiments with the nger located 1mm away from theantenna have been carried out considering three distinctpositions le middle and right e palm is 20mm spacedfrom the ground plane in order to characterize a realisticscenario when the user is holding the phone For the threemonopoles the same scheme is used (the bottom monopoledoes not suffer from the nger loading effect)

For the single antenna the nger in the right position iscritical since the nger totally covers the antenna whereas forthe le position the nger is far away (Figure 21) It should beoutlined that these experiments consider a critical scenario inwhich the nger is only 1mm above the antenna

For the array of two elements efficiency is better for allcases except for the le position where the single antennadoes not suffer from the nger effect since it is far awayHowever in the best case of the single antenna antenna

WPCB

LPCB

dgap

L

F 22 L-shaped monopole printed on a ground plane edimensions of the monopole antenna are 119871119871 = 23mm with a stripwidth of 2mm and it is located in the shorter edge of a PCB at adistance119863119863gap = 4mm from the ground plane e PCB dimensionsare 119871119871PCB = 90mm and119882119882PCB = 40mm

efficiencies for the single and the array of two elements arequite comparable e advantage of the array of two elementsis demonstrated for the other cases where the efficiency isabove the efficiency of the single antenna case

For the array of three elements the advantages are evenbetter since it presents the best results among the threeprototypes For example for the right case the efficiency inthe 824ndash960MHz frequency range is 25 dB higher than thearray using two elements and 79 dB higher than the singleantenna case showing that this technique may be useful tomitigate the efficiency drop due to the nger loading that canbe directly related to a decrement of the battery durationreduction of coverage and eventually call drops


0

10

20

30

40

50

60

70

80

90

100

Effi

cie

ncy (

)

Frequency (GHz)

Antenna efficiency with MN

16

2

16

6

17

1

17

5

17

9

18

3

18

7

19

1

19

5

19

9

20

4

20

8

21

2

21

6

22

22

4

22

8

23

2

23

7

24

1

24

5

24

9

25

3

25

7

26

1

26

5

27

F 23Measured antenna efficiency of the L-monopole shown in Figure 22 Broadbandmatching network consisting of a shunt119871119871 = 33nHand a shunt 119862119862 = 13pF

A distributed handset antenna system using three smallmonopoles has been described featuring enough bandwidthto cover the communication standards in the range of 686to 970MHz is concept uses an array of monopoles witha proper phase shi to improve the bandwidth comparedwith a single antenna element Moreover the proposedsystem is robust to the nger effect because when oneelement is interfered by the nger there are still two moreelements that efficiently contribute to the radiation Finally itshould be emphasized that the proposed distributed system iscompletely passive being advantageous in terms of simplicityand battery consumption

433 Matching Networks In combination with antennatechniques matching networks play a signicant role notonly in tuning the band location but also in providing greaterbandwidth [71ndash75] A technique consisting of a simple circuitis discussed to enhance the bandwidth of a simple antenna bya factor of about 245 times for SWR = 3 [72 73]

Matching networks using lumped components are widelyused in many commercial handset devices In many situ-ations the use of a matching network helps to ne tunethe operating bands Here a technique for broadening theinherent bandwidth of a handheld antenna is reviewedBasically the technique consists in adding an LC shunt circuitthat allows creating an impedance loop of proper size to beinscribed inside the circle of a given target SWR [73]

A circuit analysis shows that the bandwidth of an antennafeaturing an input impedance similar to that produced byan RLC series circuit around the central operating frequencycan be improved by a theoretical factor of 245 regarding anSWR = 3 [73] To demonstrate the potential of this tech-nique a single L-shaped monopole featuring an RLC seriesinput impedance along the central frequency of operation ismatched with a broadband matching network (Figure 22)Bandwidth and efficiency measurements demonstrate thatthis single element of reduced dimensions can be operativeat GSM1800 GSM1900 UMTS LTE2100 LTE2300 andLTE2500 (Figure 23)

Measured radiation patterns are stable across the fre-quency range of operation being omnidirectional and havinga minimum along the long axis of the PCB Measureddirectivities range from 28 to 44 dB As a result a BWenhancement of at least one half of Fanorsquos limit [76] isachieved with a simple two-stage matching network As apractical example a monopole with an inherent BW0 of1421 SWR le 3 has been improved to achieve a BW119891119891 of524 SWR le 3with an averagemeasured antenna efficiencyof 565

As a conclusion matching networks and in particular theproposed broadbandmatching network allows increasing thebandwidth of the antenna element without the necessity ofincreasing the antenna size

434 Intelligence in the Ground Plane e efforts on theantenna design have been mainly addressed to the antennageometry and not to the ground plane since its relevancein the radiation process was underestimated Accordinglythe antenna element was typically a self-resonant elementthat provided an efficient radiation independently from theground plane structure Nevertheless the ground plane isprogressively acquiring relevance since several studies havedemonstrated its strong contribution to the radiation prop-erties [77ndash90]

e future generations of mobile phones will need tooperate over as much frequency bands as possible suchas LTE700 GSM850 GSM900 DCS1800 PCS UMTSLTE2300 LTE2500 among others It has been shown thata ground plane length of 04120582120582 effectively excites the groundplane which improves bandwidth and efficiency [37]

us the antenna design is mainly determined by thePCB dimensions which are xed by the size of the handset orwireless device A further important limitation is the antennaheight which should be small enough as for allowing theemergent generation of ultraslim phones Moreover suchnew mobile phones also incorporate extra-large number ofextra services such as photo-video cameras big displays towatch television and several speakers for high-delity audio


(a) (b)

F 24 (a) Introducing slots in the ground plane to electrically lengthen the current path (b) Continuous arrows are a qualitativerepresentation of the main current distribution for 900MHz which is distributed along the long edges of the PCB PCB is 100mm times 40mm

(a)

(b) (c)

F 25 Manufactured antenna prototypes (a) dual-band PIFAand rear view of the (b) quad-band PIFA and (c) the hexabandPIFA In (a) the carrier to attach themetal plate and the plastic coverare also shown

which undesirably contribute to the reduction of the availablespace to t the antennaerefore new techniques are neededin order to attain themaximumperformancewith an antennathat occupies the smallest possible space ree techniques tomanipulate the ground plane are revisited

(i) use of slot to lengthen the ground plane(ii) use of a conductive strip to lengthen the groundplane(iii) use of traps to electrically reduce the ground plane

Lengthen the Ground Plane by Using Slots To effectivelyenlarge the ground plane slots can be used e idea is

illustrated in Figure 24 where the slot is used to tune theground plane mode (enlarging the current path) at the lowfrequency range (900MHz) while placed underneath theantenna area to act as a parasitic element at higher frequencies(1800ndash2100MHz)

Prototypes of three PIFA antennas namely a dual-bandPIFA without slots a quad-band PIFA with one slot and theproposed hexaband PIFA with multiple slots on the groundplane have been constructed and studied (Figure 25) [88]e simulation soware IE3D was used for optimizing thedesign parameters

In this concept a slotted ground plane is used to improvethe bandwidth at both low and high frequency regionswithout increasing the volume of the antenna On one handat low frequencies the slot is below resonance but forcesthe ground plane mode to be excited so as to increase thebandwidth at low frequencies on the other hand the slotsare comparable to 1205821205824 at high frequencies and thereforethey enhance the bandwidth (Figure 26) is solution doesnot excite directly the slots as the case with PIFA and slotsexplained in Section 431 but by coupling being the PIFA thedriven element

e placement of a component (speaker) over the slot(without any metallic contact between the speaker and theground plane) does not affect the antenna performance at lowfrequencies However it is critical at high frequencies whenthe component is close to the open edge of the slot [88] eeffect is minimized at the center and at the short end of theslot Also the SAR has been evaluated for this concept andthe ones using slots in the ground plane Results show thatthis concept presents a similar SAR to that of the PIFA onthe bare PCBwith the advantage that more bands are coveredwith the slotted ground plane solution [89]

is new design has been compared with the same designwithout the slots Results show that the bandwidth and as aconsequence the total efficiency are improved obtaining aradiator useful for multiband handset applications

Lengthen the Ground Plane by Using Conductive Strips Asdiscussed above the ground plane plays an important rolein the electromagnetic behavior of a handset antenna enext technique uses a conductive strip on the ground plane toeffectively produce an electromagnetic enlargement capable


minus14

minus12

minus10

minus8

minus6

minus4

minus2

0

S1

1(d

B)

07

08

09 1

11

12

13

14

15

16

17

18

19 2

21

22

23

24

25

Frequency (GHz)

Dual-band PIFA

Quad-band PIFA

Multiband PIFA

082 GHz

minus6 dB

103 GHz

minus6 dB

176 GHzminus6 dB

184 GHz

minus6 dB

197 GHz

minus6 dB

25 GHz

minus5 dB

212 GHz

minus61 dB

F 26 Measured reection coefficient for the three studiedprototypes It can be seen how the proposed multiband design canoperate at least over the GSM850 GSM900 DCS PCS UMTS andBluetooth bands

F 27 Conductive strip on a real handset PCB

of tuning the resonant frequency of the fundamental modeto lower values close to 900MHz (Figure 27) Basically tomake the ground plane larger a strip at the opposite edgeof the antenna location is used Such a strip is designed totune the ground plane mode [91] As a result the bandwidthand efficiency are increased e length of the strip canbe reduced by inductive loading andor dielectric loadingPhysical insight is given by electrical models [91] and usingradar cross-section analysis [92]

Other authors have used the strip to mitigate the handloading effect [63] In [93] a mechanism to control nearelectrical and magnetic elds is used for hearing-aid compat-ibility

To give a better perspective of the efficiency improve-ment four case studies are selected (Figure 27) handsetphone without strip with the strip with the strip lengthhaving 48mm and 23mm and with the respective loadinginductor On one hand it is clearly shown how the efficiencyis improved at the low frequency region (Figure 28) eunloaded strip and the inductive loaded strip having 48mmlength perform very similar demonstrating the benet of theinductance loading e 23mm case improves the efficiencypeak but the efficiency drops at 960MHz In summary thestrip with 48mm length improves the efficiency across the

0

5

10

15

20

25

30

35

40

45

50

55

820 890 920 960

Frequency (MHz)

Measu

red

to

tal

effi

cie

ncy (

)

Without strip

Strip 70 mm length Strip 23 mm length L = 47 nH

Strip 48 mm length L = 22 nH

F 28 Measured total efficiency without the strip and with thestrip considering loading inductors for the case shown in Figure 27e case having L of 22 nH and 48mm length effectively enhancesthe efficiency across the 820ndash960MHz band

band In particular the improvement at 960MHz is verysignicant 35 dB On the other hand the strip does not alterthe performance in the high frequency region [91]

As a conclusion this technique is useful to improve thebandwidth and efficiency at the low frequency region wherethe ground plane is smaller than 04120582120582 which is approximatelythe optimum length to excite the fundamental mode of theground plane and thus to maximize the bandwidth andefficiencyReducing the Ground Plane Using Stubs In some platformssuch as for example clamshell type handsets the groundplane is large in open position Moreover if the antennais placed at one edge instead of that in the hinge it mayexcite a particular mode that results in a radiation patternwith many lobes and a minimum in the horizontal plane Inthis regard the present technique consists in reducing theelectrical length of the ground plane by adding a trap (Figure29) [94] In [95] the technique of using traps increases thebandwidth at the high frequency region In effect at thisfrequency a typical length of a bar-type handset of 100mmis 0631205821205820 at 1900MHz being larger than 041205821205820 erefore thestrip forces the ground plane to be 041205821205820 in length at suchfrequencies Similar effects can be obtained by introducing aslot in the ground plane [96]

When the antenna is placed at one edge of a clamshellplatform the radiation in the horizontal plane does notpresent a maximum radiation due to a multilobe pattern Byadding the trap which is a short-ended 1205821205824 stub at the centralfrequency of operation the current is blocked due to the highimpedance of the stub In this way the current is minimizedAs a result the higher order mode has been removed atthe ground plane that supports a fundamental mode whichradiates with a maximum in the horizontal plane

435 Ground Plane Boosters Wireless device manufacturersregard the volume dedicated to the integration of the radiat-ing structure and in particular the antenna element as being


Main PCB

Upper PCB

x

y

z

(a)

Trap

x

z

y

(b)

F 29 Simulated current distribution at 182GHz without and with a shortening mechanism consisting of a short circuit plate of 1205821205824 at182GHz e continuous line is a qualitative approach of the currents on the ground plane For (a) a current mode having two sinusoids issupported causing a multi-lobe pattern For (b) the current in the upper PCB board has been mitigated due to the trap

F 30 Comparison of a PIFA antenna and the solution based on ground plane boosters for operation at GSM850900 DCS PCS andUMTS e volume of the PIFA is 4600mm3 whereas the compact solution is only 250mm3

a toll to pay in order to provide wireless capabilities to thehandheld or portable device

e new technique named ground plane booster antennatechnology provides very compact elements easy to inte-grate and able to be used as standard elements [97ndash106]is technique is based on the concept of using the groundplane as the main radiator An element called groundplane booster is in charge of properly exciting the efficientradiation modes that the inherent ground plane of anywireless platform features at mobile frequencies Its properlocation together with a radiofrequency system allows multi-band operation with signicant small dimensions (eg only250mm3 to obtain multiband performance at GSM850 9001800 1900 and UMTS) thus making the new architectureattractive to emergent multifunction wireless devices

Other different approaches have appeared in the liter-ature In [107] two antenna structures based on couplingelements designed to transfer energy to the ground planemode are presented ey are intended for covering thecommunication standards GSM900 andGSM1800 separatelyby means of a single-resonant matching circuit based on dis-tributed matching elements Other reference based on cou-pling elements is given in [108] where an antenna structureconsisting in two coupling elements and two resonant circuitsis proposed e proposal achieves a quad-band behavior

Nevertheless the coupling elements presented for coveringeach frequency region (624mm3 and 64mm3 resp) andespecially the one in charge of providing operability in thelow frequency region still present a considerable volumecompared to the 250mm3 disclosed herein for providingpentaband operation In [98 100] the pentaband behavioris achieved by means of two ground plane boosters and twomatching networks capable to provide multiband operationat each frequency region (Figure 30)

A wireless device employing very small elements wouldbe advantageous as it would make the integration of theradiating structure into the wireless handheld device easiere volume freed up by the absence of the antenna elementwould enable smaller andor thinner devices or even to adoptradically new form factors which are not feasible today due tothe presence of an antenna element Furthermore by elim-inating precisely the element that requires customizationa standard solution is obtained which only requires minoradjustments to be implemented in different wireless devices

Accordingly the present solution replaces the self-resonant antenna element by nonresonant ground planeboosters (Figure 31) In this case a challenge appears sincethe ground plane resonance is not coupled to the antennaresonance us the present technique is focused on provid-ingmultibandwireless handheld device architecture based on


F 31 Schematic of a handset phone including two groundplane boosters located at the short edge of the PCB

F 32 Single-band prototype including the reactance cancella-tion inductor and the broadband matching network

the proper excitation of the ground plane without the need ofan antenna element [97ndash100] is technique demonstratesthat no handset antenna is required for effectively exciting theradiation modes of the ground plane e novel architectureintroduced here only requires small ground plane boostersfeatured by a high quality factor (119876119876 119876 119876119876119876119876 for the lowfrequency region and119876119876 119876 119876119876119876 for the high frequency region)and extremely poor stand-alone radiation properties incombination with a matching network for providing simul-taneous operability in the main communication standards(GSM850900 DCS PCS and UMTS) [100]

However the proper excitation of the predominant modeis not enough for providing pentaband behavior and amatch-ing network is required in order to guarantee operabilityin the aforementioned communication standards For thepresent example each ground plane booster uses a reactanceelement to cancel out the reactance and a broadbandingcircuit as the one described in Section 433 to achieveenough bandwidth to cover the required standards Sucha broadbanding circuit follows the principles explained inSection 433 (Figure 32) Also a combiner is used to mergethe two port solution into a single inputoutput port (Figure33)

In this sense the conventional handset antenna featuredby a considerable volume (1198764550mm3) has been replacedby two low-volume nonresonant ground plane boosters(250mm3) and amatching topologywith a systematic designese elements are in charge of properly exciting the efficientradiation mode of the ground plane which presents highradiation efficiency and low 119876119876 at the frequencies of interestespecially in the low frequency region (GSM850900) e

High frequency region

Low frequency region

Transmission line

F 33 Pentaband prototype designed including the reactancecancellation inductor the broadband matching network and thenotch lters required for providing isolation between both fre-quency regions

systematic matching network design enables the operabilityin the desired frequency regions e radiation contributionprovided by such small boosters is negligible and theyshould not be considered as antennas Consequently theirintegration in the handset platform removes the need ofincluding a dedicated antenna in thewireless handheld device[97ndash106]

e effects of head absorption and SAR have beencompared to other technologies such as PIFA using slots inthe ground planes and the coupled monopoles presented inthis paper resulting in a technique more robust to the effectsof the head [109]

is proposal becomes an alternative to current antennatechnology and appears as a promising standard solution forbeing integrated in emergentmultifunctional wireless devicessince the available space in handset platforms for integratingnew functionalities is further increased while the radiatingperformance is preserved ew advances in this eld showthe possibility of adding new bands such as LTE700 andLTE210023002500

5 Conclusions

e apparition of newwireless communications systems withnew platforms makes the antenna design a difficult challengesince not only more antennas are needed to operate at newbands but also the antennas require multiband operation andsmall size to be integrated into the wireless handheld devices

However the characterization of the antennas is asimportant as their designe antennas integrated in wirelesshandheld devices operate in singular environments like forexample the presence of the human body and the multipathsignal propagation which add additional challenges eseparticular environments force the antenna community tocharacterize the integrated antennas in wireless handhelddevice to attain efficient antenna systems for this kind ofsituations On one hand head and hand phantoms are used toanalyze the effect that the human body has on the electromag-netic performance of the antennas and also how the radiation


of the antennas affects the human bodyis characterizationfacilitates the understanding of the antenna behaviorwhich atthe end serves to make robust antenna systems On the otherhand the multipath environment fosters new measurementssystems such as reverberation chambers which can emulate areal propagation environment

Finally smaller and multiband radiating systems arerequired to allow the integration of other handset com-ponents such as for example big displays which are acommon feature of current smartphones and an importantfactor for the nal user In this regard the ground planeboosters presented herein offer an alternative to currentantenna technologies since they signicantly reduce thevolume occupied by the radiating system while preservingthe electromagnetic performance An example of two groundplane boosters having a size of only 5mm times 5mm times 5mmhas been proved to operate at GSM850 GSM900 GSM1800GSM1900 and UMTS erefore the ground plane boostersbecome a promising technology for the new generation ofwireless handheld devices

Acknowledgments

e authors would like to thank the following institutionsfor their nancial support Spanish Ministry of IndustryCommerce and Tourism and ACC1Oacute

References

[1] ldquoMotorola Executive Helped spur Cellphone Revolutionrdquo WallStreet Journal p A10 2009

[2] httpwwwctiaorg[3] httpwwwfccgov[4] ldquoSafety levels with respect to human exposure to radio fre-

quency electromagnetic elds 3 kHz to 300GHzrdquo ANSIIEEEC95 1

[5] Guidelines for Limiting Exposure to Time-Varying ElectricMagnetic and Electromagnetic Fields (up to 300GHz) Inter-national Commission on Non-Ionizing Radiation Protection(ICNRP)

[6] ldquoAmerican national standard for methods of measurement ofcompatibility between wireless communications devices andhearing aidsrdquo ANSI C6319-2007

[7] M Andersson A Wolfgang C Orlenius and J CarlssonldquoMeasuring performance of 3GPPLTE terminals and small basestations in reverberation chambersrdquo in Long Term Evolution3GPP LTE Radio and Cellular Technology chapter 12 CRCPress New York NY USA 2009

[8] P S Kildal and K Rosengren ldquoCorrelation and capacity ofMIMO systems and mutual coupling radiation efficiency anddiversity gain of their antennas simulations and measurementsin a reverberation chamberrdquo IEEE Communications Magazinevol 42 no 12 pp 104ndash112 2004

[9] P S Kildal C Orlenius and U Carlberg ldquoMIMO LTE OTAmeasurements in reverberation chamber rich isotropic refer-ence environment makes agreement with theoretical systemmodelrdquo in Proceedings of the 6th European Conference on Anten-nas and Propagation (EuCAP rsquo12) Prague Czech RepublicMarch 2012

[10] T Taga ldquoAnalysis for mean effective gain of mobile antennasin land mobile radio environmentsrdquo IEEE Transactions onVehicular Technology vol 39 no 2 pp 117ndash131 1990

[11] J Carlsson U Carlberg and P S Kildal ldquoDiversity gains in ran-dom line-of-sight and rich isotropic multipath environmentrdquoin Proceedings of the Loughborough Antennas and PropagationConference (LAPCrsquo12) pp 1ndash4 Leicestershire UK November2012

[12] C Orlenius P S Kildal and G Poilasne ldquoMeasurementsof total isotropic sensitivity and average fading sensitivityof CDMA phones in reverberation chamberrdquo in Proceedingsof the IEEE Antennas and Propagation Society InternationalSymposium and USNCURSI Meeting pp 409ndash412 PiscatawayNJ USA July 2005

[13] A Skarbratt J Aringsberg and C Orlenius ldquoOver-the-air per-formance testing of wireless terminals by data throughputmeasurements in reverberation chamberrdquo in Proceedings of the5th European Conference onAntennas and Propagation (EUCAPrsquo11) pp 615ndash619 Rome Italy April 2011

[14] P Lindberg and A Kaikkonen ldquoBuilt-in handset antennasenable FM transceivers inmobile phonesrdquo RFDesignMagazine2007

[15] J Anguera D Aguilar J Vergeacutes M Riboacute and C PuenteldquoHandset antenna design for FM receptionrdquo in Proceedingsof the IEEE Antennas and Propagation Society InternationalSymposium San Diego Calif USA 2008

[16] D Aguilar J Anguera M Riboacute and C Puente ldquoSmall handsetantenna for FM receptionrdquo Microwave and Optical TechnologyLetters vol 50 no 10 pp 2677ndash2683 2008

[17] J Anguera C Borja C Picher and A Anduacutejar ldquoWire-less device providing operability for broadcasting standardsand method enabling such operabilityrdquo Patent applicationWO2010145825

[18] C Picher J Anguera A Anduacutejar C Borja C Puente and SKahng ldquoReuse of the mobile communication antenna for FMreceptionrdquo in Proceedings of the 5th European Conference onAntennas and Propagation (EuCAP rsquo11) pp 324ndash327 RomeItaly April 2011

[19] C Borja J Anguera C Puente and J Vergeacutes ldquoHow much canbe reduced the internal FM antenna of mobiles phonesrdquo inProceedings of the 4th European Conference on Antennas andPropagation (EuCAP rsquo10) Barcelona Spain April 2010

[20] J Anguera and A Sanz ldquoWireless portable device includ-ing internal broadcast receiverrdquo Patent application WO2007128340

[21] C Puente E Rozan and J Anguera ldquoSpace lling miniatureantennasrdquo Patent application WO 01 54225

[22] J Vergeacutes J Anguera C Puente and D Aguilar ldquoAnalysis ofthe human body on the radiation of FM handset antennardquoMicrowave and Optical Technology Letters vol 51 no 11 pp2588ndash2590 2009

[23] A Pladevall C Picher A Anduacutejar and J Anguera ldquoSomethoughts on human body effects on handset antenna at theFM bandrdquo Progress in Electromagnetics Research M vol 19 pp121ndash132 2011

[24] J Anguera C Puente E Martiacutenez and E Rozan ldquoe fractalHilbert monopole a two-dimensional wirerdquo Microwave andOptical Technology Letters vol 36 no 2 pp 102ndash104 2003

[25] C Puente E Rozan and J Anguera ldquoSpace lling miniatureantennasrdquo Patent application WO0154225

[26] D Gala J Soler C Puente C Borja and J Anguera ldquoMiniaturemicrostrip patch antenna loaded with a space-lling transmis-sion line based on the fractal Hilbert curverdquo Microwave andOptical Technology Letters vol 38 no 4 pp 311ndash312 2003


[27] J Anguera Fractal and broadband techniques on miniaturemultifrequency and high-directivity microstrip patch antennas[PhD thesis] Department of Signal eory and Communica-tions Universitat Politegravecnica de Catalunya 2003

[28] J Anguera C Puente C Borja and J Soler ldquoFractal-shapedantennas a reviewrdquo Wiley Encyclopedia of RF and MicrowaveEngineering vol 2 pp 1620ndash1635 2005

[29] K J Vinoy K A Jose V K Varadan and V V VaradanldquoResonant frequency of Hilbert curve fractal antennasrdquo inProceedings of the IEEE Antennas and Propagation SocietyInternational Symposium vol 3 pp 648ndash651 Boston MassUSA July 2001

[30] K J Vinoy K A Jose V K Varadan andV V Varadan ldquoHilbertcurve fractal antenna a small resonant antenna for VHFUHFapplicationsrdquoMicrowave andOptical Technology Letters vol 29no 4 pp 215ndash219 2001

[31] S R Best ldquoA comparison of the performance properties of theHilbert curve fractal and meander line monopole antennasrdquoMicrowave and Optical Technology Letters vol 35 no 4 pp258ndash262 2002

[32] S R Best ldquoA comparison of the resonant properties of smallspace-lling fractal antennasrdquo IEEE Antennas and WirelessPropagation Letters vol 2 pp 197ndash200 2003

[33] J M Gonzaacutelez-Arbesuacute S Blanch and J Romeu ldquoAre space-lling curves ecient small antennasrdquo IEEE Antennas andWireless Propagation Letters vol 2 pp 147ndash150 2003

[34] S R Best and J D Morrow ldquoe effectiveness of space-lling fractal geometry in lowering resonant frequencyrdquo IEEEAntennas and Wireless Propagation Letters vol 1 pp 112ndash1152002

[35] S R Best and J D Morrow ldquoOn the signicance of currentvector alignment in establishing the resonant frequency ofsmall space-lling wire antennasrdquo IEEE Antennas and WirelessPropagation Letters vol 2 pp 201ndash204 2003

[36] I Sanz J Anguera A Anduacutejar C Puente and C Borjaldquoe Hilbert monopole revisitedrdquo in Proceedings of the 4thEuropean Conference on Antennas and Propagation (EuCAPrsquo10) Barcelona Spain April 2010

[37] K L Wong Planar Antennas for Wireless CommunicationsWiley-Interscience New York NY USA 2003

[38] T Taga and K Tsunekawa ldquoPerformance analysis of a built-in planar inverted-F antenna for 800MHz band portable radiounitsrdquo IEEE Journal on Selected Areas in Communications vol5 no 5 pp 921ndash929 1987

[39] C R Rowell and R D Murch ldquoA compact PIFA suitable fordual-frequency 9001800-MHz operationrdquo IEEE Transactionson Antennas and Propagation vol 46 no 4 pp 596ndash598 1998

[40] D Manteuffel A Bahr and I Wolff ldquoInvestigation on inte-grated antennas for GSM mobile phonesrdquo in Proceedings of theESA Millennium Conference on Antennas amp Propagation (APrsquo00) Davos Switzerland April 2000

[41] C Puente C Borja J Anguera and J Soler ldquoMultilevelantennasrdquo Patent application WO0122528

[42] M Martiacutenez-Vaacutezquez O Litschke M Geissler D HeberlingAMMartiacutenez-Gonzaacutelez andD S Saacutenchez-Hernaacutendez ldquoInte-grated planar multiband antennas for personal communicationhandsetsrdquo IEEE Transactions on Antennas and Propagation vol54 no 2 pp 384ndash391 2006

[43] C Y Chiu P L Teng and K L Wong ldquoShorted folded planarmonopole antenna for dual-band mobile phonerdquo ElectronicsLetters vol 39 no 18 pp 1301ndash1302 2003

[44] K L Wong G Y Lee and T W Chiou ldquoA low-proleplanar monopole antenna for multiband operation of mobile

handsetsrdquo IEEE Transactions on Antennas and Propagation vol51 no 1 pp 121ndash125 2003

[45] K L Wong and C H Huang ldquoPrinted loop antenna with aperpendicular feed for penta-band mobile phone applicationrdquoIEEE Transactions on Antennas and Propagation vol 56 no 7pp 2138ndash2141 2008

[46] K L Wong and S C Chen ldquoPrinted single-strip monopoleusing a chip inductor for penta-band WWAN operation in themobile phonerdquo IEEE Transactions on Antennas and Propaga-tion vol 58 no 3 pp 1011ndash1014 2010

[47] H Kanj and S M Ali ldquoCompact multiband folded 3-Dmonopole antennardquo IEEE Antennas and Wireless PropagationLetters vol 8 pp 185ndash188 2009

[48] J Ma Y Z Yin J L Guo and Y H Huang ldquoMiniature printedoctaband monopole antenna for mobile phonesrdquo IEEE Anten-nas and Wireless Propagation Letters vol 9 pp 1033ndash10362010

[49] H W Hsieh Y C Lee K K Tiong and J S Sun ldquoDesignof a multiband antenna for mobile handset operationsrdquo IEEEAntennas and Wireless Propagation Letters vol 8 pp 200ndash2032009

[50] C T Lee and K L Wong ldquoPlanar monopole with a couplingfeed and an inductive shorting strip for LTEGSMUMTSoperation in the mobile phonerdquo IEEE Transactions on Antennasand Propagation vol 58 no 7 pp 2479ndash2483 2010

[51] J Anguera A Condes J Soler and C Puente ldquoCoupledmultiband antennasrdquo Patent application WO 04025778

[52] S Risco J Anguera A Anduacutejar A Peacuterez and C PuenteldquoCoupled monopole antenna design for multiband handsetdevicesrdquo Microwave and Optical Technology Letters vol 52 no2 pp 359ndash364 2010

[53] C I Lin and K L Wong ldquoPrinted monopole slot antenna forinternal multiband mobilephone antennardquo IEEE Transactionson Antennas and Propagation vol 55 no 12 pp 3690ndash36972007

[54] C H Wu and K L Wong ldquoHexa-band internal printed slotantenna for mobile phone applicationrdquo Microwave and OpticalTechnology Letters vol 50 no 1 pp 35ndash38 2008

[55] N Takemura ldquoInverted-FL antenna with self-complementarystructurerdquo IEEE Transactions on Antennas and Propagation vol57 no 10 pp 3029ndash3034 2009

[56] J Anguera I Sanz J Mumbruacute and C Puente ldquoMultibandhandset antenna with a parallel excitation of PIFA and slotradiatorsrdquo IEEE Transactions on Antennas and Propagation vol58 no 2 pp 348ndash356 2010

[57] J Anguera and C Puente ldquoShaped ground plane for radioapparatusrdquo Patent application WO 2006070017

[58] S K Sharma L Shafai and N Jacob ldquoInvestigation of wide-band microstrip slot antennardquo IEEE Transactions on Antennasand Propagation vol 52 no 3 pp 865ndash872 2004

[59] C H Li E Oi N Chavannes and N Kuster ldquoEffects ofhand phantom on mobile phone antenna performancerdquo IEEETransactions on Antennas and Propagation vol 57 no 9 pp2763ndash2770 2009

[60] M Pelosi O Franek M B Knudsen G F Pedersen and J BAndersen ldquoAntenna proximity effects for talk and data modesin mobile phonesrdquo IEEE Antennas and Propagation Magazinevol 52 no 3 pp 15ndash27 2010

[61] J Ilvonen O Kivekaumls J Holopainen R Valkonen K Rasi-lainen and P Vainikainen ldquoMobile terminal antenna perfor-mance with the userrsquos hand effect of antenna dimensioning andlocationrdquo IEEE Antennas and Wireless Propagation Letters vol10 pp 772ndash775 2011


[62] W Yu S Yang C L Tang and D Tu ldquoAccurate simulation ofthe radiation performance of a mobile slide phone in a hand-head positionrdquo IEEE Antennas and Propagation Magazine vol52 no 2 pp 168ndash177 2010

[63] J M Jung S J Kim K H Kong J S Lee and B LeeldquoDesigning ground plane to reduce hand effects on mobilehandsetsrdquo in Proceedings of the IEEE Antennas and PropagationSociety International Symposium Honolulu Hawaii USA June2007

[64] R Valkonen S Myllymaumlki A Huttunen et al ldquoCompensationof nger effect on a mobile terminal antenna by antennaselectionrdquo in Proceedings of the International Conference onElectromagnetics in Advanced Applications (ICEAA rsquo10) pp364ndash367 Sydney Australia September 2010

[65] J Ilvonen R Valkonen O Kivekaumls P Li and P VainikainenldquoAntenna shielding method reducing interaction between userand mobile terminal antennardquo Electronic Letters vol 47 no 16pp 896ndash897 2011

[66] J Anguera and C Puente ldquoDistributed antenna system robustto human loading effectsrdquo Patent application WO 2007141187

[67] J Anguera A Camps A Anduacutejar and C Puente ldquoEnhancingrobustness of handset antennas to nger loading effectsrdquo Elec-tronics Letters vol 45 no 15 pp 770ndash771 2009

[68] J Anguera A Andujar Y Cobo C Picher and C PuenteldquoHandset antenna array to mitigate the nger loading effectrdquoin Proceedings of the 5th European Conference on Antennas andPropagation (EUCAP rsquo11) pp 611ndash614 Rome Italy April 2011

[69] A Anduacutejar J Anguera Y Cobo and C Picher ldquoDistributedantenna systems for wireless handheld devices robust to handloadingrdquo IEEE Transactions on Antennas and Propagation vol60 no 10 pp 4830ndash4837 2012

[70] ldquoBasic standard for the measurement of specic absorptionrate related to human eposure to electromagnetic eldsfrom mobile phones (300MHzndash3GHz)rdquo CENELEC-EuropeanCommittee for Electrotechnical Standardization Std EN 50361 2001

[71] J S Lee G C Kang B Jung et al ldquoTriple band internalantenna using matching circuitsrdquo in Proceedings of the IEEEAntennas and Propagation Society International Symposium andUSNCURSI Meeting vol 1A pp 442ndash445 July 2005

[72] J Anguera C Puente C Borja G Font and J Soler ldquoA sys-tematic method to design single-patch broadband microstrippatch antennasrdquoMicrowave and Optical Technology Letters vol31 no 3 pp 185ndash188 2001

[73] A Anduacutejar J Anguera and C Puente ldquoA systematic methodto design broadband matching networksrdquo in Proceedings of the4th European Conference on Antennas and Propagation (EuCAPrsquo10) Barcelona Spain April 2010

[74] Y Li B Derat D Pasquet and J C Bolomey ldquoMatching limitsfor a dual-band mobile phone antennardquo in Proceedings of theIEEE International Symposium on Microwave Antenna Prop-agation and EMC Technologies for Wireless Communications(MAPE rsquo07) pp 656ndash659 Hangzhou China August 2007

[75] Y Li T Cantin B Derat D Pasquet and J C BolomeyldquoApplication of resonant matching circuits for simultaneouslyenhancing the bandwidths of multi-band mobile phonesrdquo inProceedings of the IEEE International Workshop on AntennaTechnology Small and Smart Antennas Metamaterials andApplications (iWAT rsquo07) pp 479ndash482 Cambridge UK March2007

[76] R M Fano ldquoeoretical limitations on the broadband match-ing of arbitrary impedancesrdquo Journal of the Franklin Institutevol 249 no 2 pp 139ndash154 1950

[77] T Y Wu and K L Wong ldquoOn the impedance bandwidth of aplanar inverted-F antenna for mobile handsetsrdquoMicrowave andOptical Technology Letters vol 32 no 4 pp 249ndash251 2002

[78] M C Huynh and W Stutzman ldquoGround plane effects on pla-nar inverted-F antenna (PIFA) performancerdquo IEE ProceedingsMicrowaves Antennas and Propagation vol 150 no 4 pp209ndash213 2003

[79] K L Wong J S Kuo and T W Chiou ldquoCompact microstripantennas with slots loaded in the ground planerdquo in Proceedingsof the 11th International Conference on Antennas and Propaga-tion (IEE Conference Publication No 480) vol 2 pp 623ndash626Manchester UK April 2001

[80] P Vainikainen J Ollikainen O Kivekaumls and I KelanderldquoResonator-based analysis of the combination of mobile hand-set antenna and chassisrdquo IEEE Transactions on Antennas andPropagation vol 50 no 10 pp 1433ndash1444 2002

[81] R Hossa A Byndas and M E Bialkowski ldquoImprovementof compact terminal antenna performance by incorporatingopen-end slots in ground planerdquo IEEE Microwave and WirelessComponents Letters vol 14 no 6 pp 283ndash285 2004

[82] A Byndas R Hossa M E Bialkowski and P Kabacik ldquoInvesti-gations into operation of single- and multi-layer congurationsof planar inverted-F antennardquo IEEE Antennas and PropagationMagazine vol 49 no 4 pp 22ndash33 2007

[83] M F Abedin and M Ali ldquoModifying the ground plane and itseffect on planar inverted-F antennas (PIFAs) for mobile phonehandsetsrdquo IEEE Antennas and Wireless Propagation Letters vol2 pp 226ndash229 2003

[84] B Sanz-Izquierdo J Batchelor and R Langley ldquoMultibandprinted PIFA antenna with ground plane capacitive resonatorrdquoElectronics Letters vol 40 no 22 pp 1391ndash1392 2004

[85] J Anguera I Sanz A Sanz et al ldquoEnhancing the performanceof handset antennas by means of groundplane designrdquo inProceedings of the IEEE International Workshop on AntennaTechnology Small Antennas and Novel Metamaterials (iWATrsquo06) pp 29ndash32 New York NY USA March 2006

[86] M Cabedo E Antonino V Rodrigo and C Suaacuterez ldquoAnaacutelisisModal de un Plano de Masa Radiante Doblado y con unaRanura para Terminales Moacutevilesrdquo in Proceedings of the 21stNational Symposium URSI rsquo06 Oviedo Spain 2006

[87] J Anguera I Sanz A Sanz T Condes C Puente and J SolerldquoMultiband PIFA handset antenna by means of groundplanedesignrdquo in Proceedings of the IEEE Antennas and PropagationSociety International Symposium Albuquerque NM USA July2006

[88] A Cabedo J Anguera C Picher M Riboacute and C PuenteldquoMultiband handset antenna combining a PIFA slots andground plane modesrdquo IEEE Transactions on Antennas andPropagation vol 57 no 9 pp 2526ndash2533 2009

[89] C Picher J Anguera A Anduacutejar C Puente and S KahngldquoAnalysis of the human head interaction in handset antennaswith slotted ground planesrdquo IEEE Antennas and PropagationMagazine vol 54 no 2 pp 36ndash56 2012

[90] C Picher J Anguera A Cabedo C Puente and S KahngldquoMultiband handset antenna using slots on the ground planeconsiderations to facilitate the integration of the feeding trans-mission linerdquo Progress in Electromagnetics Research C vol 7 pp95ndash109 2009

[91] J Anguera A Anduacutejar and C Puente ldquoA mechanism toelectrically enlarge the ground plane of handset antennas abandwidth enhancement techniquerdquo Microwave and OpticalTechnology Letters vol 53 no 7 pp 1512ndash1517 2011


[92] J Anguera and A Anduacutejar ldquoGround plane contribution inwireless handheld devices using radar cross section analysisrdquoProgress in Electromagnetics Research M vol 26 pp 101ndash1142012

[93] J Holopainen J Ilvonen O Kivekaumls R Valkonen C Ichelnand P Vainikainen ldquoNear-eld control of handset antennasbased on inverted-top wavetraps focus on hearing-aid compat-ibilityrdquo IEEE Antennas and Wireless Propagation Letters vol 8pp 592ndash595 2009

[94] J Anguera and C Puente ldquoHandset with electromagnetic brardquoPatent application WO 2005083833

[95] P Lindberg and E Oumljefors ldquoA bandwidth enhancement tech-nique for mobile handset antennas using wavetrapsrdquo IEEETransactions on Antennas and Propagation vol 54 no 8 pp2226ndash2233 2006

[96] C T Lee and K L Wong ldquoInternal WWAN clamshell mobilephone antenna using a current trap for reduced ground planeeffectsrdquo IEEE Transactions on Antennas and Propagation vol57 no 10 pp 3303ndash3308 2009

[97] J Anguera A Anduacutejar C Puente and JMumbruacute ldquoAntennalesswireless devicerdquo Patent application WO2010015365 2009

[98] J Anguera A Anduacutejar C Puente and J Mumbruacute ldquoAntenna-less wireless device capable of operation in multiple frequencyregionsrdquo Patent Application WO2010015364 2009

[99] J Anguera and A Anduacutejar ldquoAntennaless wireless devicecomprising one or more bodiesrdquo Patent application WO2011095330

[100] A Anduacutejar J Anguera and C Puente ldquoGround plane boostersas a compact antenna technology forwireless handheld devicesrdquoIEEE Transactions on Antennas and Propagation vol 59 no 5pp 1668ndash1677 2011

[101] A Anduacutejar J Anguera C Puente and C Picher ldquoWirelessdevice capable of multiband MIMO operationrdquo Patent applica-tion WO 2012017013

[102] A Anduacutejar and J Anguera ldquoCompact radiating array forwireless handheld or portable devicesrdquo Patent Application US61661 885 2012

[103] J Anguera C Picher A Anduacutejar and C Puente ldquoCon-centrated antennaless wireless device providing operability inmultiple frequency regionsrdquo Patent application US 616719062012

[104] A Anduacutejar and J Anguera ldquoOn the radiofrequency system ofground plane booster antenna technologyrdquo Electronics Lettersvol 48 no 14 pp 815ndash817 2012

[105] A Anduacutejar and J Anguera ldquoMultiband coplanar ground planebooster antenna technologyrdquo Electronic Letters vol 48 no 21pp 1326ndash1328 2012

[106] A Anduacutejar and J Anguera ldquoMagnetic boosters for multi-bandoperationrdquo Microwave and Optical Technology Letters vol 55no 1 pp 65ndash75 2013

[107] J Villanen J Ollikainen O Kivekaumls and P VainikainenldquoCoupling element based mobile terminal antenna structuresrdquoIEEE Transactions on Antennas and Propagation vol 54 no 7pp 2142ndash2153 2006

[108] S Ozden B K Nielsen C H Jorgensen J Villanen C Ichelnand P Vainikainen ldquoQuad-band coupling element antennastructurerdquo US Patent 7 274 340 2007

[109] A Anduacutejar J Anguera C Picher and C Puente ldquoGroundplane booster antenna technology Human head interactionfunctional and biological analysisrdquo in Proceedings of the 6thEuropean Conference on Antennas and Propagation (EuCAPrsquo12) pp 2745ndash2749 Prague Czech Republic 2012


Application ArticleEvaluation of SARDistribution in Six-Layer HumanHeadModel

Asma Lak1 and Homayoon Oraizi2

1 Young Researchers Club Bushehr Branch Islamic Azad University Bushehr Iran2 Iran University of Science and Technology Tehran Iran

Correspondence should be addressed to Asma Lak lakasmaegmailcom

Received 4 May 2012 Revised 2 August 2012 Accepted 16 October 2012

Academic Editor Aurora Anduacutejar

Copyright copy 2013 A Lak and H Oraiziis is an open access article distributed under theCreativeCommonsAttributionLicensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

e interaction between human head model and electromagnetic eld sources is studied e head models are composed of oneand six layerse six layers are skin fat bone dura (the outermembrane of brain and spinal cord) CSF (colony stimulating factor)and brain An antenna as a source of exposure is simulated too e E-eld strength distribution in both one- and six-layer humanmodels is shown to estimate the intensity of E-eld penetration in human head Like standard models the antenna is situated nearthe head model at a distance of 5mm e local and average SARs (specic absorption rates) are simulated at 00MHz in bothhuman head models e results are then compared between the two models e HFSS soware is used for all the simulations epaper wants to show that the initial model (one layer) is not a good model because the real human head tissue is not equivalentlymodeled It seems that the values of one-layer model are not reliable so the paper considers the better and more similar humanhead model and compares these two models

1 Introduction

It is well known that high frequency EM elds can damagehuman and other biological tissues by damaging molecularstructure and rising of body temperature e biologicaleffects of radiofrequency elds and living systems can beevaluated at various levels including the molecular subcellu-lar organ or whole body environments According to [1 2]bioeffects from radiofrequency elds are classied into threecategories that is high-level effects (thermal) intermediate-level effects (athermal) and low-level effects (nonthermal)ermal effects are energy depositions higher than thenatural human thermoregulatory capacity e studies showsome effects due to nonthermal and athermal sources suchas blood brain barrier morphology immune system geneand chromosomal morphology enzyme activity and tumourpromotion More information can be seen in [3ndash5] In thispaper dosimetry and SAR are dened So the human headmodel (one and six layers) and an antenna as an exposuresource are simulated in HFSS soware For validation ofresults two antenna types are used dipole and PIFA eresults for SAR and E-eld strength for these two models areshown and compared Because of some limitations the stan-dard phantom models are made of one layer For example

because of the gel or liquid materials it is not easy to modelall tissues For example the human head is amultilayer tissueand its modelling is very hard ese phantoms are not goodmodels for the human tissue because the real properties oftissues are different from each other Also the human headdoes not have equivalent electrical properties So the Six-layer phantom model is the better model of human head tosimulate

2 Measurements of EM Field Absorption

In this part the dosimetry is introduced Some parametershave been used to measure EM elds e SAR as a quantityfor EM measurement at radiofrequency spectrum is denedand nally the electrical properties of tissues that is conduc-tivity and permittivity are described

21 enition of osimetry According to [7] the relation-ship between exposure levels and electromagnetic energydeposited in the body is called ldquoelectromagnetic dosimetryrdquoOn the other hand the electromagnetic dosimetry describesthe relationship between the induced elds in biologicalbodies and distribution of an electromagnetic eld in free


1 +07

1 +06

1 +05

1 +04

1 +03

1 +02

1 +01

1 +00

1 +02

1 +01

1 +00

1 minus 01

1 minus 02

1+

2

1+

3

1+

4

1+

5

1+

6

1+

7

1+

8

1+

9

1+

10

1+

11

Permittivity

Conductivity

Frequency

F 1 Dielectric properties spectrum of a high water contenttissue [3]

space Dosimetry information is very important to protecthumans from probable electromagnetic eld health hazards

22 Sei bsortion ate By the widespread wirelessdevice applications such as mobile phones the personsand operators living and working in near electromagneticsources the biological effects of exposure to these elec-tromagnetic elds are an important subect e safetystandards such as the Federal Communication Commis-sion (FCC) the International Commissions on NonionizingRadiation Protection (ICNIRP) and National RadiologicalProtection Board (NRPB) are established for human pro-tection and safety from electromagnetic elds e specicabsorption rate (SAR) is used to quantify the energy absorbedin tissues at radiofrequency spectrum which is expressed inunits of watts per kilogram SAR is dened as the ratio of theabsorbed power to the absorbing mass [7 8] e total powerabsorbed in the human body is

119875119875abs = 10045601004560119881119881

12120590120590|119864119864|2119889119889119881119881119889 (1)

where 120590120590 is the conductivity of tissue 119864119864 is the electric eldintensity and 119881119881 is the volume of the biological tissue SARis dened as

SAR = 10076531007653 1205901205902120588120588100766910076691198641198642119894119894 119889 (2)

where 120590120590 is the conductivity of tissue 119864119864 is is the electric eldintensity and 120588120588 is the mass density of the tissue

23 Human Tissues Human body tissues have differentvalues of dielectric properties that is permittivity andconductivity [9] ese properties are functions of severalvariables such as frequency geometry and size of tissue andwater contents For example the dielectric constant of a highwater content tissue is shown in Figure 1 as a function offrequency

F 2 ne-layer human head model dened in HFSS soware

F 3 Six-layer human head model dened in HFSS soware

3 Simulations

Many researchers have simulated measured and evaluatedthe probable biological effects of M elds on the humanand other living systems Several researchers have simulatedthe human body models specially the human head and haveevaluated the effective parameters on SAR [11ndash14]

31 Phantoms Measurements of SAR and M elds in thehuman body are not possible consequently the phantomshave been designed tomodel the human body at normal bodytemperatures ey have many shapes such as spherical andhuman-like bodies e liquids or gels as materials to tissuesare placed in phantoms and exposure source is situated nearthem ne robot arm will then measure the or H eld by aprobe placed at various locations near themodel A computerprocessor calculates the SAR ese measurement systemshave several problems

32 Modelling by HFSS Soware Because of the availablecommercial systems the common models for SAR mea-surements are a thin bowl (a 5mm thickness shell with 46relative permittivity) containing fully the head (brain) tissueequivalent materials Figure 2 shows this commercial model


Brain

CSF

Dura

Bone

Fat

Skin

F 4 e perspective of Six-layer human head model [6]

F 5 Dipole antenna

T 1 Specications of one-layer head

Tissue Permittivity Conductivity(Sm)

ickness(mm)

Head equivalentmaterial 415 09 85

Shell 46 0 5

T 2 ther specications of model

Dipole length 149mmpower 1WattSAR linebull 180mmbullSAR line is a line that HFSS soware measures the SAR around it

en a six-layer humanheadmodel has been tried insteadof a one-layer common phantom model because it modelsthe real human head in a much better way as shown in Figure3is newmodel is composed of six spheres similar to a six-layer model for the human head that is skin fat bone duraCSF and brain as shown in Figure 4

Tables 1 2 and 3 show the model properties and dimen-sions in HFSS simulations

33 Source Exposure A dipole antenna has been used as theexposure source as shown in Figure 5 e antenna is situatedat 5mm distance from the head models in both one- and six-layer models e simulations have been done at frequency900MHz e length of antenna is 149mm its radius is18mm and the frequency of operation is 900MHz

34 Characteristics of Models All specications of the one-and six-layer human head model and dipole antenna areshown in Tables 1 to 3

19525 +002

18313 +002

171e+002

15887 +002

14674 +002

13462 +002

12249 +002

11036 +002

98234 +001

86107 +001

73979 +001

61852 +001

49724 +001

37597 +001

2547 +001

13342 +001

1215 +000

E field (Vm)

F 6 -eld strength distribution in one-layer human headmodel at 900MHz

20

18

16

14

12

10

8

6

4

2

00 10 20 30 40 50 60

Distance (mm)

Local SAR

Average SAR

SAR (wattkg)

F 7 Head model as a function of the distance between thedipole and the head model

T 3 Tissue properties and thickness of six-layer human headmodel [6]

Tissue Permittivity Conductivity (Sm) ickness (mm)Skin 407 065 1Fat 10 017 014Bone 209 033 041Dura 407 065 05CSF 791 214 02Brain 411 086 81

4 Results

In this section the results of human head model in oneand six layer and antenna model both dipole and PIFA areshown Also the results for SAR and electric elds strengthare compared


E field (Vm)

12126 +002

11368 +002

10611 +002

98528 +001

9095 +001

83372 +001

75795 +001

68217 +001

60639 +001

53062 +001

45484 +001

37906 +001

30329 +001

22751 +001

15173 +001

75958 +000

18184 minus 002

F 8 -eld strength in brain tissue at 900

E field (Vm)

13783 +002

12922 +002

12062 +002

11201 +002

10341 +002

94802 +001

86196 +001

7759 +001

68984 +001

60379 +001

51773 +001

43167 +001

34561 +001

25956 +001

1735 +001

8744 +00013818 minus 001

F 9 -eld strength in CSF tissue at 900

T 4 -eld strength in six-layer model dipole antenna

Tissue -eld strength (m)Skin 180times 102

Fat 177times 102

Bone 167times 102

Dura 140times 102

Csf 137times 102

Brain 121times 102

41 ldquoOne-Layerrdquo HumanHeadModel with Dipole Antenna at09 GHz is model consists of a shell with 5mm thicknessand a sphere with 85mm radius as the head equivalent mate-rials According to (2) the SAR has a direct relationship to

E field (Vm)

14096 +002

13216 +002

12336 +002

11457 +002

10577 +002

96977 +001

88181 +001

79385 +001

70589 +001

61793 +001

52997 +001

44201 +001

35405 +001

26609 +001

17813 +001

90171 +00022117 minus 001

F 10 -eld strength in Dura tissue at 900

E field (Vm)

1672 +002

15676 +002

14631 +002

13586 +002

12541 +002

11496 +002

10451 +002

94058 +001

83609 +001

73159 +001

6271 +001

5226 +001

41811 +001

31361 +001

20912 +001

10462 +00112991 minus 002

F 11 -eld strength in bone tissue at 900

the -eld strength Because of the importance of the electriceld in SAR calculations the -eld strength distribution inthe one-layer headmodel is shown in Figure 6emaximumvalue is at the nearest point to the source exposure (redcolour) e local and average SAR as a function of thedistance between the dipole and the head model is shown inFigure 7

42 ldquoSix-Layerrdquo Human Head Model with Dipole Antennaat 09 GHz e -eld distribution in the six-layer modelis simulated and shown in Figures 8 9 10 11 12 and 13e -eld strength is simulated in all the six layers by theaforementioned procedure

e result of simulations is shown in the Table 4 Asshown the -eld strength is decreasing by the distance


T 5 Compression between SAR in two models (one- and six-layer model by Dipole antenna)

Max SAR (WattKg) Layer Conductivity Local SAR Average SAR

One layer + dipole Shell 0 0 0Head equivalent material 09 18times 10+1 12times 10+1

Six layers + dipole

Skin 065 139times 10+2 22times 10+1

Fat 017 3 57times 10+1 777Bone 033 506times 10+1 205times 10+1

Dura 065 66times 10+1 783CSF 214 216times 10+2 775times 10+1

Brain 086 88times 10+1 45 times 10+1

T 6 -eld strength in six layers (PIFA antenna)

Tissue -eld strength (m)Skin 799times10minus1

Fat 78times 10minus1

Bone 74times 10minus1

Dura 64times 10minus1

CSF 62times 10minus1

Brain 52times 10minus1

E field (Vm)

17717 +002

16612 +002

15507 +002

14401 +002

13296 +002

12191 +002

11086 +002

99804 +001

88752 +001

777e+001

66647 +001

55595 +001

44542 +001

3349e+001

22438 +001

11385 +00133291 minus 001

F 12 -eld strength in fat tissue at 900MHz

from the source consequently the maximum value of -eldstrength in the brain tissue as an internal layer is the lowest

e comparison of SAR between one- and six-layer headmodels are given in Table 5

e values show that the maximum of SAR strength inthe six layers is more than one layer It says that the standardmodel (that is used in standard systems) may be not suitableand complete and does not show the accurate model ofhuman tissues

43 ldquoSix-Layerrdquo Human Head Model with PIFA Antenna at09 GHz For further consideration of the problem simula-tion is repeated for head model but with PIFA antenna at

E field (Vm)

18019 +002

16896 +002

15774 +002

14651 +002

13529 +002

12407 +002

11284 +002

10162 +002

90396 +001

79172 +001

67948 +001

56724 +001

455 +001

34276 +001

23052 +001

11828 +00160453 minus 001

F 13 -eld strength in skin tissue at 900MHz

75 mm

10 mm Antenna patch

62 mm

33 mm

Ground plane

50 mm

92 mm

F 14 e structure of PIFA antenna at 09GHz [10] lowastHeightof the antenna patch from the ground plane is the 75mm

900MHz [10] e geometry of antenna is shown in Figures14 15 and 16 e other specications of the model areaccording to Table 1 e results for the -eld strength andSAR simulations are shown in Figures 17 18 19 20 21 and22

e results show that the -eld strength is decreasingby increasing the distance from the source consequentlythe maximum value of -eld strength in the brain tissueat the internal layer is the lowest e results for PIFAantenna are similar to those of dipole antenna with regards


T 7 Compression between SAR in two models (one- and six-layer model by PIFA antenna)

Max SAR (wattKg) Layer Conductivity Local SAR Average SAR

One layer + PIFA Shell 0 0 0Head equivalent material 09 127times10minus4 083times10minus4

Six layers + PIFA

Skin 065 24times10minus3 127times10minus3

Fat 017 14times10minus3 09times10minus3

Bone 033 1times10minus3 099times10minus3

Dura 065 15times10minus3 052times10minus3

CSF 214 5times10minus3 32times10minus3

Brain 086 2times10minus3 147times10minus3

F 15 PIFA antenna at 900MHz

F 16 Six-layer human head model with PIFA antenna

to the decreasing values of SAR with increasing the distancebetween the source and head Table 6 shows the results forPIFA antenna

Table 7 shows the SAR maximum strength values in one-and six-layer model by PIFA antenna as a source exposureAccording to these results it has been seen that the SAR valuesin these simulation depend on the distance from antennaand conductivity value of tissues For example in one-layermodel shell is the nearer layer to the exposure source soalthough it has lower conductivity the SAR ismore than headequivalent material Also in six-layer model the SAR is variedby conductivity and distance from the exposure source too

E field (Vm)

52151 minus 001

48999 minus 001

45847 minus 001

42695 minus 001

39542 minus 001

3639 minus 001

33238 minus 001

30086 minus 001

26934 minus 001

23782 minus 001

2063 minus 001

17478 minus 001

14326 minus 001

11174 minus 001

80214 minus 002

48693 minus 002

17172 minus 002

F 17 -eld strength in brain tissue at 900MHz

E field (Vm)

62004 minus 001

58225 minus 001

54445 minus 001

50665 minus 001

46886 minus 001

43106 minus 001

39327 minus 001

35547 minus 001

31768 minus 001

27988 minus 001

24209 minus 001

20429 minus 001

1665 minus 001

1287 minus 001

90907 minus 002

53112 minus 002

15317 minus 002

F 18 -eld strength in CSF tissue at 900MHz


E field (Vm)

64315 minus 001

60437 minus 001

56559 minus 001

52682 minus 001

48804 minus 001

44926 minus 001

41048 minus 001

3717 minus 001

33292 minus 001

29415 minus 001

25537 minus 001

21659 minus 001

17781 minus 001

13903 minus 001

10025 minus 001

61475 minus 002

22696 minus 002

F 19 -eld strength in ura tissue at 900MHz

E field (Vm)

7482 minus 001

70223 minus 001

65625 minus 001

61027 minus 001

56430 minus 001

51832 minus 001

47234 minus 001

42637 minus 001

38039 minus 001

33441 minus 001

28844 minus 001

24246 minus 001

19648 minus 001

15051 minus 001

10453 minus 001

58554 minus 002

12578 minus 002

F 20 -eld strength in bone tissue at 900MHz

For example dura and skin have same conductivity but theskin is the nearer layer to antenna so it has more SAR

e penetration of elds on human body for examplehuman head has been considered bymany researchers Someof these researches are in simulation by soware and someof them are done by measurement systems (human phantommodels) For more information the references [15ndash18] canbe seen So at the same frequency of exposure source theresults may be different because of the difference in humanbody model

5 Conclusions

e simulations are done at 900MHz because it is the stan-dard for the mobile communication systems e resulting

E field (Vm)

78242 minus 001

73529 minus 001

68816 minus 001

64103 minus 001

5939 minus 001

54678 minus 001

49965 minus 001

45252 minus 001

40539 minus 001

35826 minus 001

31114 minus 001

26401 minus 001

21688 minus 001

16975 minus 001

12262 minus 001

75495 minus 002

28367 minus 002


E field (Vm)79925 minus 001

75133 minus 001

70341 minus 001

65549 minus 001

60757 minus 001

55965 minus 001

51173 minus 001

46381 minus 001

41589 minus 001

36798 minus 001

32006 minus 001

27214 minus 001

22422 minus 001

17630 minus 001

12838 minus 001

80463 minus 002

32544 minus 002


data show that when the human body tissue especially thehuman head (because of placing the cell phone near thehead) is exposed to M elds the elds penetrate in allthe human head tissues e -eld strength penetrationand SAR deposition depend on tissue material properties(conductivity permittivity and permeability) By increasingthe conductivity SAR increases too In this paper two humanhead models have been used one and six layer e one-layer human head model is simple and easy to use forsimulation and measurement system because the humanbody equivalent materials are gel or liquid in commercialSAR measurement system But it has no assurance and themeasurements are not reliable because it does not show agood and real model of human headSo Six-layer head modelhas been used In Tables 5 and 7 the local and average SARfor one and six layer is shown e compression of both local


and average SAR in six-layer models is more than one layerFurthermore the commercial models (viz one layer model)may not be good models for the human body because theresults show that the EM eld penetration is higher in a six-layer model So the design of a better model is unavoidablefor the study of the exposure of human body to EM eldsources e result for another mobile antenna that is PIFAis repeated and showes similar results Results show that byPIFA antenna the E-eld strengths are very lower Accordingto (2) the SAR is lower too

References

[1] M H Repacholi ldquoLow-level exposure to radiofrequency elec-tromagnetic elds health effects and research needsrdquo Bioelec-tromagnetics vol 19 no 1 pp 20ndash32 1998

[2] A G Pakhomov Y Akyel O N Pakhomova B E Stuck andM R Murphy ldquoCurrent state and implications of research onbiological effects of millimeter waves a review of the literaturerdquoBioelectromagnetics vol 19 no 7 pp 393ndash413 1998

[3] F S Barnes and B Greenebaumby Bioengineering and Biophys-ical Aspects of Electromagnetic Fields Handbook of BiologicalEffect of Electromagnetic Fields Taylor amp Francis Group Lon-don UK 3rd edition 2007

[4] P Vecchia R Matthes G Ziegelberger James Lin and RSaunders Exposure To High Frequency Electromagnetic FieldsBiological Effects and Health Consequences (100KHzndash300GHz)International Commission on Non-Ionizing Radiation Protec-tion Oberschleissheim Germany 2009

[5] J C Lin ldquoEffects of microwave and mobile telephone exposureon memory processrdquo IEEE Antenna and PropagationMagazinevol 42 no 3 pp 118ndash120 2000

[6] H Khodabakhshi and A Cheldavi ldquoIrradiation of a six-layeredspherical model of human head in the near eld of a half-wavedipole antennardquo IEEE Transactions on Microwave eory andTechniques vol 58 no 3 pp 680ndash690 2010

[7] D A Saacutenchez-Hernaacutendez High Frequency ElectromagneticDosimetry 2009

[8] J C Lin Advances in Electromagnetic Fields in Living Systemsvol 4 Springer New York NY USA 2005

[9] C Gabriel ldquoe dielectric properties of tissuesrdquo in Radiofre-quency Radiation Dosimetry and Its Relationship To the Biolog-ical Effects of Electromagnetic Fields B J Klauengerg and DMiklavic Eds vol 82 of Nato Science Series pp 75ndash84 HighTechnology London UK 2000

[10] C W Khoo Multi-band antenna for handheld transceivers[PhD thesis] 2002

[11] A Lak H Oraizi and F Mohsenifard ldquoRisk from electromag-netic eldsrdquo in Proceedings of the 3rd International Conferenceon Mechanical and Electrical Technology (ICMET rsquo11) DalianChina August 2011

[12] L Asmae and O Homayoon ldquoSimulation and evaluation ofspecic absorption rate in human body in high frequencyelectromagnetic eldsrdquo in Advanced Materials Research vol433ndash440 pp 5489ndash5493 Trans Tech Publications ZurichSwitzerland 2012

[13] M R I Faruque M T Islam and N Misran ldquoAnalysis of SARlevels in human head tissues for four types of antennas withportable telephonesrdquo Australian Journal of Basic and AppliedSciences vol 5 no 3 pp 96ndash107 2011

[14] ldquoInternational Standard IEC 62209-1 human exposure to radiofrequency elds from hand-held and body-mounted wirelesscommunication devices-human models instrumentation andprocedures-Part 1 procedure to determine the specic absorp-tion rate (SAR) for hand-held devices used in close proximityto the ear (frequency range of 300MHz to 3GHz)rdquo IECpublication 2005

[15] A Anduacutejar J Anguera C Picher and C Puente ldquoHuman headinteraction over ground plane booster antenna technologyfunctional and biological analysisrdquo Progress in ElectromagneticsResearch vol 41 pp 153ndash185 2012


[17] S Risco J Anguera A Anduacutejar C Picher and J PajaresldquoComparison of a monopole and a PIFA handset antennain the presence of the human headrdquo Microwave and OpticalTechnology Letters vol 54 no 2 pp 454ndash459 2012

[18] A Lak and H Oraizi ldquoe effect of distance of human headmodel from EM sources on SARrdquo Journal of Basic and AppliedScientic Research vol 2 no 9 pp 9446ndash9453 2012

Hindawi Publishing CorporationInternational Journal of Antennas and PropagationVolume 2012 Article ID 516487 7 pagesdoi1011552012516487

Research Article

Printed Internal Pentaband WWAN AntennaUsing Chip-Inductor-Loaded Shorting Strip forMobile Phone Application

Yong-Ling Ban1 Shun Yang1 Joshua Le-Wei Li1 and Rui Li2

1 Institute of Electromagnetics University of Electronic Science and Technology of China 2006 Xi-Yuan AvenueWestern High-Tech District Sichuan Chengdu 611731 China

2 College of Software Engineering Chengdu University of Information Technology Chengdu 610225 China

Correspondence should be addressed to Yong-Ling Ban byluestceducn

Received 10 July 2012 Accepted 17 September 2012

Academic Editor Minh-Chau Huynh

Copyright copy 2012 Yong-Ling Ban et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

A compact size on-board printed antenna using capacitive coupled-fed excitation to generate multiple resonant modes for penta-band WWAN operation (GSM850900GSM18001900UMTS2100) is presented in this paper The proposed antenna occupiesonly a small footprint of 15 times 25 mm

2on one corner of the circuit board and a protruded ground of 10 times 15 mm

2is displaced

with close proximity to the antenna portion The proposed antenna has a very simple structure which is composed of two separatestrips a loop strip with an inserted chip inductor and an L-shaped feeding strip The loop strip is shorted to the ground andgenerates a resonant mode at 890 MHz to cover the GSM850900 band (824ndash960 MHz) while the feeding strip contributes to theGSM18001900UMTS210 band (1710ndash2170 MHz) operation With such a small size the proposed antenna can achieve compactintegration on the circuit board of the mobile phone thus the proposed scheme is quite suitable for the slim mobile phoneapplication Good agreements between simulations and measurements are obtained Details of proposed antenna are presentedand some key parameters are studied

1 Introduction

Mobile phone antennas with compact size low profile andwide operation band characteristics have attracted greatattention both in academic and industrial field In recentyears a variety of small size and broadband antennas excitedby the capacitive coupled-fed scheme to achieve multibandoperation have been reported [1ndash4] These reported anten-nas can be configured to occupy a compact volume inside themobile phone for multiband operation However most of thereported designs did not consider the integration of internalantenna with the system ground plane because an isolationdistance is often needed to guarantee the performance ofwideband operation As a result these antennas often occupythe whole edge of the system circuit ground plane whichis not so suitable for the practical application [1 2] Thisphenomenon is also common in some traditional three-dimensional antenna designs [3] It limits the integrationof the internal antenna with the associated electronic

components Recently several novel designs with protrudedground are proposed [5 6] which integrate the antennawith the system board well furthermore it has been shownthat protruded ground can effectively suppress the surfacecurrent distribution on the ground plane away from theedge where the antenna is mounted [7] Thus decreased nearfield emission can be achieved if the antenna is placed atthe bottom of the mobile handset But most of them arethree-dimensioned or have a large area occupation [5 6] Tominiature the antenna size chip inductors are widely usedto reduce the length of the strip for a special frequency soas to realize compact designs [8ndash10] A penta-band solutionwith protruded ground plane is reported in [11] which ispromising to be implemented in the slim mobile phonedesigns In this paper we presented a WWAN internalantenna suitable to be disposed at a small corner on thecircuit board to achieve compact integration The proposedantenna has a size of 15times 25 mm2 which only requires a small


Anetnna part

No-groundboard space

1-mm thick plastic casing(εr = 3 tan = 002)

θ ϕ

z

x

y

100 times 60 mm2

main ground

50Ω microstrip feedlinefor the testing antenna

25

B

A Via to a 50Ω SMA

10

08-mm thick FR4 substrateas system circuit

board (115 times 60 mm2)Protruded

ground

(a)

L = 12 nH

65

7

3

2

1

23

15

45

m = 12

t = 225 05

A

B

Gap= 15

xy

z

(b)

Figure 1 (a) Overall structure of the proposed antenna (b) Dimensions of the proposed antenna (unit mm)

foot print of the system circuit plane A protruded groundarea of 10 times 15 mm2 is left at the center part of the edgeto accommodate associated electronic component such as auniversal serial bus (USB) connector and another no-groundportion at the other side of the circuit board edge nearthe protruded ground is promising for a another internalantenna deposition Detailed configurable illustrations andradiation characteristics of the proposed antenna are givenin the following sections

2 Proposed Antenna Configuration

Figure 1(a) shows the geometry of the on-board printedcoupled-fed compact antenna The proposed antenna isprinted on a small no-ground board portion of 15 times 25 mm2

and it only occupies a small part of the edge on the mobilephone system circuit board A 08 mm thick FR4 substrate ofrelative permittivity 44 loss tangent 0024 length 115 mmand width 60 mm is used in this study 1 mm thick plastic cas-ing of relative permittivity 30 and loss tangent 002 enclosesthe whole substrate to simulate the mobile phone casing Asit shows in the figure a protruded ground portion is placedclosely to the antenna and connected to the main groundThe protruded ground has a size of 10times 15 mm2 and it is justsuitable to accommodate a USB connector Also note thaton the other side of the protruded ground there is anotherno-ground board space of size 15 times 25 mm

2which can be

used to accommodate other internal antenna or electroniccomponents Furthermore such a small and symmetricalscheme is promising to develop a compact MIMO (multipleinput multiple output) antenna designs [12ndash14] Figure 1(b)shows prototype of the proposed antenna which is mainlycomposed of two parts an L-shaped feeding strip and a loopshorting strip The L-shaped feeding strip is directly fed frompoint A which is further connected to a 50-Ω transmissionline as shown in the Figure 1(a) The loop shorting strip iscoupled-fed by the L-shaped feeding strip and shorted to theground plane through a via-hole at point B A chip inductor

Figure 2 The photos of the proposed antenna

of L = 82 nH is inserted at the corner of the loop strip toshorten the path for GSM850900 band operation The totallength of loop strip is about 60 mm that is much shorterthan a quarter of the wavelength for 850 MHz which is about90 mm As the lower band is mainly generated by the loopshorting strip the front portion of the loop strip is set to bea variable of m as shown in Figure 1(b) The length of the L-shape monopole is also set as a variable t to tune the upperband operation Detailed effects of the parameters on theantenna performance will be shown in the following sections

3 Result and Discussion

Figure 2 shows the fabricated antenna with rulers to demon-strate the antenna size The simulation is done usingthe high frequency simulation software (HFSS) version 12and the measurement in conducted by Agilent N5247Avector network analyzer Good match between measuredand simulated reflection coefficient of the proposed antennadesign is shown in Figure 3 The impedance matching forfrequencies over the two operating bands is better than6-dB return loss which is widely used as the design spec-ification for the internal WWAN mobile phone antennasAccording to this criterion both the simulation and themeasured results cover the operation bands (GSM850900GSM18001900UMTS2100) perfectly The simulated inputimpedance of the proposed antenna on the smith chart isshown in Figure 4 to provide more impedance information

International Journal of Antennas and Propagation 3S1

1 (d

B)

500 1000 1500 2000 2500

SimulatedMeasuredminus6 dB

0

minus5

minus10

minus15

minus20

minus25

minus30

Frequency (MHz)

Figure 3 Measured and simulated S11 for the proposed antenna

A dashed-line circle is drawn in the smith chart to demon-strate the region in which the impedance is well matched

In order to classify the function of different parts ofproposed antenna the operating principles are analyzedFigure 5 shows simulation reflection coefficient of thecomparison between the proposed antenna and referenceantennas The corresponding Ref1 antenna has only the feedstrip while the Ref2 case has no inserted inductor For theRef1 case there is no resonance near 900 MHz and theresonance of the upper band is also some kind of weak dueto absence of the bended monopole And for the Ref2 caseit is seen that both the lower band and the upper band areshifted towards higher frequencies and this phenomenon isespecially obvious in the lower band These results supportthe idea that the higher band is generated by the L-shapedfeeding strip and also slightly affected by the loop strip whilethe lower band operation is dominantly decided by the loopstrip It also indicates that the inductor can effectively reducethe length of strip for specific resonance which helps torealize the miniature of the antenna design

Simulated current distributions on the antenna part andground plane at 890 MHz and 1940 MHz are shown inFigure 6 The directional arrows show the current flowingwith varied colors Corresponding magnitude of differentcolor is shown as label on the left At the frequency of890 MHz the current flows along the loop strip and themagnitude of the current are smallest at the front part whileit increases to the largest value at the shorting point At thesame time the current on the ground plane is well alignedin the same direction which also contributes to the lowerband radiation While at the frequency of 1940 MHz there isstrong current distribution on the feeding strip which showsthat the feeding strip is the main radiator at 1940 MHz forthe antenna

Simulated reflection coefficient and current distributionsfor the case with and without USB mounted on theprotruded ground are presented in Figures 7 and 8 toexplore the possibility of integrating a USB for the practical

824

960

1710

2170

180

170

160

150

140

130

120110

100 90 8070

60

50

40

30

20

10

0

minus170

minus160

minus150

minus140

minus130

minus120

minus110minus100 minus90 minus80

minus70minus60

minus50

minus40

minus30

minus20

minus10

00 02

02

05

05

1

2

1

2 5

5

minus02

minus05

minus1

minus2

minus5

500ndash2500 MHz824ndash960 MHz1710ndash2170 MHz

Figure 4 Simulated input impedance on the Smith chart

applications To simulate the influence of the USB connectoron the antenna performance a cubic conductor with a sizeof 9 lowast 8 lowast 4 mm3 is placed under the protruded groundThe simulated reflection coefficient demonstrates a slightfrequency shifting at the upper band but still covers the upperoperation band from 1710 MHz to 2170 MHz meanwhilethere is almost no change to the reflection coefficient inthe lower band The simulated frequency range is set to befrom 500 to 3000 MHz on purpose to show the change moreclearly As it can be observed from the reflection coefficientresult in Figure 7 there is another resonant mode around2700 MHz this resonant mode is generated by the stripbetween grounding point B and the inserted inductor L asthe high frequency current is blocked by the inductor Asthe protruded ground is placed closely to the strip whichgenerates the 2700 MHz resonant mode a USB connectoraffects the 2700 MHz resonant mode while causing smallvariance in other bands This analysis can be further verifiedin the current distribution at the higher band in Figure 8It is observed from the simulated current distribution withand without the presence of USB connector there is a verylittle change of the current distribution on the antenna andsurrounding area The high frequency current along the loopstrip is effectively confined between the grounding point Band the inserted inductor L

A parametric study of the major parameters on tun-ing the antennarsquos lower and upper bands is conductedFigure 9(a) shows the simulated reflection coefficient of theproposed antenna when the value of the inserted inductor isselected to be 39 nH 82 nH and 15 nH which are availablein the lab In Figure 9(a) it is found that the excited resonant


500 1000 1500 2000 2500

Frequency (MHz)

ProposedRef1

Ref2minus6 dB

0

minus5

minus10

minus15

minus20

minus25

minus30S1

1 (d

B)

Figure 5 Comparison of the proposed antenna with two reference antennas

50000e+001

46500e+001

43000e+001

39500e+001

36000e+001

32500e+001

29000e+001

25500e+001

22000e+001

18500e+001

15000e+001

80000e+001

45000e+001

10000e+001

Jsurf (A per m)

(a) (b)

Figure 6 Current distribution at frequency of (a) 890 MHz and (b) 1940 MHz

500 1000 1500 2000 2500 3000

Frequency (MHz)

Proposed Ref antenna with USB

minus30

minus25

minus20

minus15

minus10

minus5

0

S11

(dB

)

Figure 7 Comparison of reflection coefficient between the antenna with and without USB connector


(a) (b)

Figure 8 Simulated current distribution with (a) and without (b) the presence of USB

S11

(dB

)

500 1000 1500 2000 2500

0

minus5

minus10

minus15

minus20

minus25

minus30

Frequency (MHz)

L = 39 nHL = 82 nH

L = 15 nHminus6 dB

(a)

Frequency (MHz)

500 1000 1500 2000 2500

S11

(dB

)

0

minus5

minus10

minus15

minus20

minus25

minus30

m = 7 nH

m = 12 nHm = 18 nHminus6 dB

(b)

Frequency (MHz)

500 1000 1500 2000 2500

S11

(dB

)

0

minus5

minus10

minus15

minus20

minus25

minus30

t = 39 nHt = 82 nH

t = 15 nHminus6 dB

(c)

Figure 9 Simulated reflection coefficient for the proposed antenna as a function of (a) the value of the inserted inductor (b) the length ofthe coupling strip and (c) the length of the feeding strip Other dimensions are the same as in Figure 1


minus50minus40minus30minus20minus10

minus10

0

10

10

0

0

90

180

270

minus40minus30minus20

xz-plane

(a)


minus10

0

10

10

0

0

90

180

270


yz-plane

(b)

Figure 10 Measured radiation pattern of the proposed antenna at the frequencies of 890 MHz and 1940 MHz (line with rectangle E-philine with cross E-theta)

800 820 840 860 880 900 920 940 960 9800

10

20

30

40

50

60

70

80

90

100

GSM850900

EfficiencyGain

Frequency (MHz)

Rad

iati

on E

ffici

ency

(

)

0

1

2

3

4

5

6

An

tenn

a Gain

(dBi)

minus2

minus1

(a)

1700 1800 1900 2000 2100 22000

10

20

30

40

50

60

70

80

90

100

DCS1800PCS1900UMTS2100

EfficiencyGain

Frequency (MHz)

Rad

iati

on E

ffici

ency

(

)

0

1

2

3

4

5

6

An

tenn

a Gain

(dBi)

minus2

minus1

(b)

Figure 11 Measured antenna radiation efficiency and antenna gain for the proposed antenna at (a) lower band and (b) upper band


mode is shifted to the lower frequencies in the lower bandwhen the value of the inductor L is increased Meanwhileonly small variations are found in the upper band whichverifies that the inserted inductor mainly affects the lowerband and is effective to reduce strip length for a specialresonant mode Effects of the length of the front portion ofthe loop strip are studied in Figure 9(b) where simulatedreflection coefficient of the proposed antenna is presentedwhen the strip lengthm varied from 7 mm to 18 mm There isgreat similarity between Figures 9(a) and 9(b) that along thechange of m there is significant effect on the lower band whilesubtle effect on the upper band This is reasonable as theloop is the main radiator of the lower band operation in thisscheme From Figures 9(a) and 9(b) it can be concluded thatthe inserted inductor and the loop strip jointly determine theresonant mode of the lower band On the other hand thereis significant change on the upper band and little variationis found in the lower band when the length of the feedingstrip is increased from 195 mm to 235 mm as shown inFigure 9(c) This is also reasonable because the feeding stripcontrols the upper band resonant mode

The radiation characteristics of the proposed antenna arealso studied Figure 10 shows the two-dimensional radiationpattern of the presented antenna at the frequencies of890 MHz and 1940 MHz For each frequency it is observedfrom three different planes namely xz-plane yz-plane andxy-plane Dipole-like radiation pattern is found at 890 MHzwhich means a dumbbell-like shape radiation pattern at theE-plane and a circle one in the H-plane However whenit comes to the upper band the radiation characteristicvaries more quickly in different directions due to surfacecurrent of the ground plane As it is shown in Figure 6the current distribution on the ground is quite uniformand contributes to the radiation of the 900 MHz But at thefrequency of 1900 MHz the length of the system groundplane is comparable to the resonant wavelength so thereare current nulls excited on the system ground plane whichresults in nulls and dips in the obtained radiation patternsradiation Figure 11 shows the measured antenna radiationefficiency and antenna gain The efficiency ranges from 42to 63 over the GSM850900 band and the efficiency variesfrom 57 to 75 for the GSM18001900UMTS2100 bandThe efficiency over the five operation bands is all above40 which is acceptable for the practical mobile antennaapplication The measured gain is about 0 to 12 dBi and16ndash25 dBi over the lower and upper bands respectivelyGood radiation characteristics are generally obtained for theproposed antenna

4 Conclusion

In this paper a compact penta-band antenna design formobile phone application is presented With the presenceof the chip inductor the resonant strip length for theGSM850900 band operation is significantly reduced Due toits small size and simple structure it is promising to be imple-mented in the slim smart mobile phone designs by usingPCB fabrication techniques with low cost Moreover Goodimpedance match and radiation characteristics are found

in the five operation bands making it preponderant forapplication in the small-size mobile phone for WWANLTEoperating communication

References

[1] C T Lee and K L Wong ldquoUniplanar coupled-fed printedPIFA for WWANWLAN operation in the mobile phonerdquoMicrowave and Optical Technology Letters vol 51 no 5 pp1250ndash1257 2009

[2] K L Wong M F Tu T Y Wu and W Y Li ldquoSmall-sizecoupled-fed printed pifa for internal eight-band ltegsmumtsmobile phone antennardquo Microwave and Optical TechnologyLetters vol 52 no 9 pp 2123ndash2128 2010

[3] Y L Ban C Q Lei J H Chen S C Sun Z X Xie and F YeldquoCompact coupled-fed PIFA employing T-shaped monopolewith two stubs for eight-band LTEWWAN internal mobilephonerdquo Journal of Electromagnetic Waves and Applications vol26 pp 973ndash985 2012

[4] W Y Chen and K L Wong ldquoWideband coupled-fed PIFAfor HAC penta-band clamshell mobile phonerdquo Microwave andOptical Technology Letters vol 51 no 10 pp 2369ndash23742009

[5] S C Chen and K L Wong ldquoLow-profile small-size wirelesswide area network handset antenna close integration with sur-rounding ground planerdquo Microwave and Optical TechnologyLetters vol 54 pp 623ndash629 2012

[6] F H Chu and K L Wong ldquoInternal coupled-fed loop antennaintegrated with notched ground plane for wireless wide areanetwork operation in the mobile handsetrdquo Microwave andOptical Technology Letters vol 54 pp 599ndash605 2012

[7] S C Chen and K L Wong ldquoHearing aid-compatible internalLTEWWAN bar-type mobile phone antennardquo Microwave andOptical Technology Letters vol 53 no 4 pp 774ndash781 2011

[8] K L Wong and S C Chen ldquoPrinted single-strip monopoleusing a chip inductor for penta-band WWAN operationin the mobile phonerdquo IEEE Transactions on Antennas andPropagation vol 58 no 3 pp 1011ndash1014 2010

[9] Y L Ban J H Chen J L W Li and Y J Wu ldquoPrintedultrawideband antenna for LTEGSMUMTS wireless USBdongle applicationsrdquo IEEE Antennas and Wireless PropagationLetters vol 11 pp 403ndash406 2012

[10] K L Wong and C T Lee ldquoSmall-size wideband monopoleantenna closely coupled with a chip-inductor-loaded shortedstrip for 11-band WWANWLANWiMAX operation in theslim mobile phonerdquo Microwave and Optical Technology Lettersvol 53 no 2 pp 361ndash366 2011

[11] Y W Chi and K L Wong ldquoInternal compact dual-bandprinted loop antenna for mobile phone applicationrdquo IEEETransactions on Antennas and Propagation vol 55 no 5 pp1457ndash1462 2007

[12] J Zhang J Ou Yang K Z Zhang and F Yang ldquoA noveldual-band MIMO antenna with lower correlation coefficientrdquoInternational Journal of Antennas and Propagation vol 2012Article ID 512975 7 pages 2012

[13] C Yang Y Yao J S Yu and X D Chen ldquoNovel compactmultiband MIMO antenna for mobile terminalrdquo InternationalJournal of Antennas and Propagation vol 2012 Article ID691681 9 pages 2012

[14] Q H Zeng Y Yao S H Liu JS Yu P Xie and X D ChenldquoTetraband small-size printed strip MIMO antenna for mobilehandset applicationrdquo International Journal of Antennas andPropagation vol 2012 Article ID 320582 8 pages 2012


Research Article

Compact Dual-Band Dual-Polarized Antenna forMIMO LTE Applications

Lila Mouffok Anne Claire Lepage Julien Sarrazin and Xavier Begaud

Department Comelec Institut Mines Telecom Telecom ParisTech LTCI CNRS UMR 5141 46 Rue Barrault75634 Paris Cedex 13 France

Correspondence should be addressed to Lila Mouffok lilamouffoktelecom-paristechfr

Received 15 May 2012 Revised 18 July 2012 Accepted 6 September 2012


Copyright copy 2012 Lila Mouffok et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

A system of two dual-band dual-polarized antennas is proposed It operates in two bands 700 to 862 MHz and 25 to 269 GHzthereby making it suitable for LTE applications The design is composed of two compact orthogonal monopoles printed closeto each other to perform diversity in mobile terminals such as tablets or laptops For each band two orthogonal polarizationsare available and an isolation higher than 15 dB is achieved between the two monopoles spaced by λ010 (where λ0 the centralwavelength in free space of the lower band) A good agreement is observed between simulated and experimental results Theantenna diversity capability is highlighted with the calculation of envelope correlation and mean effective gain for several antennasrsquopositions in different environment scenarios

1 Introduction

Deployment of existing and emerging wireless communica-tion systems require a high-data-rate transmission in orderto satisfy the needs of multimedia applications on terminalsMultiple Input Multiple Output (MIMO) applications havebeen suggested as an effective way to increase the channelcapacity by exploiting multipath scattering effects

MIMO technology is present in many recent wirelessstandards such as Long Term Evolution (LTE) and willbe implemented in mobile devices [1] Several researchworks have proven the efficiency of two-antenna diversityon mobile terminals [2 3] However when the availablespace is limited the use of a dual-polarized antenna ismore suitable than two separated antennas [4] A varietyof dual-polarized antennas have been reported recently inwhich good dual-polarized radiation over a wide bandwidth[5] and high isolation between the feeding ports [6] havebeen achieved However these antennas are mainly designedfor single-band operation [7] or for frequencies above800 MHz [8] Most of the dual-band dual-polarized antennasproposed in literature exploit harmonics frequencies [9]or use techniques to generate additional resonances such

as insertion slot [10] But generally it leads to a ratiobetween frequency bands below or equal to 2 and impliesa dependence between the two frequency bands Todayvery few designs are reported for dual-band dual-polarizedoperations for the following bands 700ndash862 MHz and 25ndash269 GHz In this paper we firstly present the design of adual-band antenna which can provide a dual-polarizationfor each band for LTE devices such as a tablet or a laptopThen we introduce an enhanced design in which the lowerbandwidth has been increased and the mutual couplingbetween ports has been reduced in the two bands Thelower band is extended towards TV White Space (TVWS)band to provide radio-cognitive capabilities to the terminal[11]

Finally the diversity performances of the proposed dual-band dual-polarized antennas are evaluated through theenvelope correlation (ρe) and the mean effective gain inisotropic indoor and outdoor environments

2 Antenna Design

As shown in Figure 1 the proposed structure is com-posed of two orthogonal monopoles with dimensions of


Lm

dWm

l

1 2

45XY

Z

Via holes

L2L1

(a)

Lgd2

Lgd1

Wgd1

Lgd3

Wgd2

Ls

Ws

Added ground plane

Removed corners

Wgd3

(b)

Figure 1 (a) Front view (b) Back view of one meander bend ending antennas with added ground plane and slot

LmtimesWm = 275times15 mm2 The two monopoles are identicaland chosen for their omni-directional radiations patternenabling them to receive signals whatever their orientationThey are printed on a 140 times 83 mm2 low cost substrate(FR4 εr = 38 plusmn 01 tan δ = 002 thickness of 07 mm)Each monopole is connected to two bend endings one bendending is a meander line whose length is L1 = 433 mmoperating at 790ndash862 MHz and the small one whose lengthis L2 = 235 mm operates at 25ndash269 GHz The distancebetween the two bend endings is l = 14 mm This designallows to obtain independent frequency bands The two

monopoles are spaced by d = 36 mm which correspondsto λ0110 for the lower band and λ023 for the higher bandwhere λ01 is the free-space wavelength of the lower bandcentral frequency ( f01 = 826 MHz) and λ02 the free-spacewavelength of the higher band central frequency ( f02 =259 GHz)

The monopoles are fed by two 50 ohms coplanar waveg-uides (CPW) directly etched in the ground plane as shownin Figure 1(b) in order to distance the connectors andto avoid perturbations on the measured radiation patternsEach CPW has a line width of 18 mm and a gap of 033 mm


07 075 08 085 09 095 1minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S11| without added ground plane|S11| with added ground plane|S21|without added ground plane|S21| with added ground plane

|S ij|(

dB)

(a)

2 22 24 26 28 3minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)


|S ij|(

dB)

(b)

Figure 2 Simulated |Si j| parameters of one meander bend ending antennas without slot with and without added ground plane (a) lowerband (b) higher band

with the ground plane Monopoles are connected to CPWthanks to metallic via holes located as the extremity of eachmonopole as shown in Figure 1(a)

21 Ground Plane Geometry Since the small bend ending isclose to the ground plane extremities it is sensitive to thepath taken by currents along the ground plane Thereforea study of the upper part of the ground plane geometryis relevant It is found that removing corners (shaded partin Figure 1(b)) provides an improvement of higher bandmatching leading to optimized dimensions Wgd1 = 25 mmLgd1 = 25 mm and Lgd2 = 46 mm

Coupling between the two antennas occurs via currentsflowing from one antenna to the other one through theground plane It can be reduced by altering the ground planeto modify currentsrsquo path Thus the ground plane is extendedwithout increasing the overall structure size by adding on thelower part of the substrate two rectangular shapes on eitherside with dimensions of each one Wgd3 = 40 mm and Lgd3 =17 mm (framed part in Figure 1(b)) Simulations have beenperformed with Transient Solver of CST Microwave StudioFigure 2 shows a comparison between |Si j| parameters fordesigns without slot with and without added ground planein each band Because of the structurersquos symmetry only |S11|and |S21| are plotted The matching bandwidth criterion istaken for a return loss less than minus10 dB With added groundplane a shift of the lower band towards lower frequencies(from 09 to 085 GHz) is observed in Figure 2(a) withoutincreasing the structure size The bandwidths of the structure

without added ground plane are 837ndash957 MHz (134)235ndash286 GHz (196) and for the structure with addedground plane are 796ndash914 MHz (138) 238ndash278 GHz(155) Regarding the isolation it is largely reduced thanksto the added ground plane |S21| becomes below minus20 dB inthe lower band Indeed a resonance has been introduced atthe frequency where coupling occurs However the couplingremains high (|S21| lt minus7 dB) in the higher band as shown inFigure 2(b)

To improve isolation between ports in the higher banda slot is etched in the ground plane while keeping the samedistance between ports (d) as shown in Figure 1(a) Theintroduction of the slot produces an open circuit which stopsthe circulation of current from one radiating element tothe other one [12] The optimized structure has a lengthLs = 34 mm and a width Ws = 14 mm Figure 3 shows acomparison of simulated |Si j| parameters of one meanderbend ending antennas with added ground plane with andwithout slot in the higher band The introduction of the slotachieves an isolation improvement of 10 dB in the higherband while it has no effect in the lower band The bandwidthis slightly reduced but still covers the desired band Thusoptimization of the two degrees of freedom which arethe slot dimensions and rectangular shapes ground planedimensions leads to a high isolation in the two frequencybands

22 Radiating Element In order to increase the bandwidth ofthe lower band towards the TVWS band two bend endings


|S11| without slot

|S11| with slot|S21| without slot|S21| with slot

2 22 24 26 28 3minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)

Figure 3 Simulated |Si j| parameters of one meander bend endingantennas with added ground plane with and without slot in thehigher band

are added below the initial meander line to provide addi-tional resonances close to each other These two meanders areout of sync to provide a single wide band Moreover the threelines are connected to each other to extend the bandwidthtowards lower frequencies After optimization with TransientSolver of CST Microwave Studio the distance between eachmeander is s = 7 mm as shown in Figure 5 and the overallsize of three bend endings antennas with added ground planeand slot becomes 150 times 90 mm2

Figure 4 shows the comparison between S-parameters ofone and three bend endings antennas with added groundplane and slot Matching bandwidth criterion is taken for|S11| lt minus10 dB It is seen that the bandwidth is enhancedtowards lower frequencies Indeed the relative bandwidth forthe structure with one bend ending is 98 (786ndash867 MHz)and 219 (692ndash862 MHz) for the structure with 3 bendendings While keeping almost the same electrical lengthof the structure the relative bandwidth has been improvedby 12 Indeed the overall size is 035 λlow times 021 λlow forthree meander bend ending antennas (λlow the free spacewavelength at 692 MHz) when it is 037 λprimelowtimes 022 λprimelow forone meander bend ending antennas (λprimelow the free spacewavelength at 786 MHz)

3 Prototype and Measurement

A prototype of three bend endings antennas with addedground plane and slot described previously has been realizedMonopoles and the ground plane with CPW are locatedon opposite sides of the same substrate and can be seensimultaneously on Figure 5 because of the transparency ofthe FR4 substrate Simulated and measured S-parameters are

1 bend ending

3 bend endings

05 06 07 08 09 1minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)

Figure 4 Simulated |S11| parameters of one and three meanderbend endings antennas with added ground plane and slot in thelower band

compared in Figure 6 Simulations results are in good agree-ment with measurement The measurement results show thatthe antenna operates in two bands (|S11| lt minus10 dB) thelower band extends from 700ndash880 MHz (219) and thehigher one from 251ndash272 GHz (8) In these two bandsthe two monopoles are satisfactorily uncoupled with anisolation |S21| below minus15 dB within the higher band andfrom 770 to 880 MHz At the beginning of the lower bandthe isolation remains acceptable and is below minus10 dB Thesimulated total efficiency of the structure which takes intoaccount all losses has been evaluated it varies from 83 to97 in the lower band and from 74 to 87 in the higherband as shown in Figures 7 and 8

Figure 9 compares the simulated and measured copolarand cross-polar radiation patterns in the E plane (YZ plane)and H plane (XZ plane) respectively Because both portsare symmetrical we only represent radiation patterns forport number 1 while port 2 is loaded by 50 ohms For bothplanes and both bands it is found that the simulated and themeasured co-polar radiation patterns are in good agreementThe maximum simulated realized gain is 25 dB at 778 MHzand 5 dB at 26 GHz The measured cross-polar level is about10 dB lower than the copolar level in the lower band but inthe higher one the polarization purity is deteriorated It isprobably due to the proximity of the meander bend endingsto the small one

To further investigate the diversity the simulated radia-tion patterns of each radiating element in the XY plane forthe two bands are plotted in Figure 10 (one port is excitedwhile the other one is loaded by 50 ohms) Thanks to agood agreement observed in Figure 9 between simulationsand measurement only simulations results are presentedAs it can be observed for the lower band the directions


Ls

Ws

S

45XY

Z

1 2

Figure 5 A photograph of the prototype with the three meander bend endings antennas

|S11| simulation

|S21| simulation

|S11|measurement

|S21|measurement

1 15 2 25minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)

Figure 6 Simulated and measured |Si j| parameters of the opti-mized three bend endings antennas with added ground plane andslot

of the pattern maxima are close to orthogonal leading togood pattern diversity Each antenna presents monopole-like radiation patterns Indeed surface currents are weakon the bend endings For the higher band even if patternsare not orthogonal one monopole presents minimum gaindirections where the other one has a maximum gain exceptfor the directions θ = plusmn45 This is well-suited to providehigh diversity capabilities

07 075 08 08550

60

70

80

90

100

Frequency (GHz)

Figure 7 Simulated total efficiency in the lower band for theoptimized three bend endings antennas with added ground planeand slot

4 Evaluation of the Diversity Performance

The diversity performance of a mobilersquos antenna systemcan be affected by the environment in which the deviceis located [13] Therefore in this section we evaluate thediversity performance of the proposed three bend endingsantennas with added ground plane and slot by calculatingthe envelope correlation coefficient (ρe) and the meaneffective gain (MEG) taking into account the propagationenvironment


25 255 26 265 27

Frequency (GHz)

50

60

70

80

90

100

Figure 8 Simulated total efficiency in the higher band for the optimized three bend endings antennas with added ground plane and slot

The envelope correlation ρe quantifies the similaritybetween the radiation patterns of the two monopoles Thelower the correlation the better the diversity performance

Vaughan and Andersen show in [13] that the coefficient canbe expressed by

ρe =∣∣∣

int

Ω

(

XPDE1θElowast2θ pθ + E1ϕE

lowast2ϕpϕ

)

dΩ∣∣∣

2

int

Ω

(


lowast1ϕpϕ

)

dΩint

Ω

(


lowast2ϕpϕ

)

dΩ (1)

E1θ(Ω) E1ϕ(Ω) E2θ(Ω) E2θ(Ω) are simulated complexelectric fields along θ and ϕ radiated by the antenna fed bytwo different ports The solid angle Ω is defined by θ[0 π]in elevation and ϕ[0 2π] in azimuth pθ(Ω) and pϕ(Ω) arethe Angle-of-Arrival (AoA) distributions of incoming wavesThe parameter XPD is the cross-polarization discriminationof the incident field and is defined as XPD = SθSϕ (whereSθ and Sϕ represent the average power along the sphericalcoordinates θ and ϕ)

The environment depends strongly on the angles ofarrival distribution and on XPD The most common dis-tributions proven by measurements are Gaussian (G) andLaplacian (L) distributions [14] Thus we consider differentdistributions in elevation while in azimuth plane (XYplane) the distribution is uniform as demonstrated by twomeasurement campaigns in the literature [14 15]

To obtain more realistic results different environmentsare considered Each environment is characterized by typicalvalues of XPD mean angle of incident wave distribution (θi)and standard deviation of wave distribution (σ) [16] Thesevalues were deduced from several measurements [14ndash16] fordifferent environments isotropic indoor and outdoor Theisotropic environment is defined by XPD = 0 dB pθ(Ω) =pϕ(Ω) = 1 the indoor (In) environment by XPD = 1 dBθi = 20 σ = 30 and the outdoor (Out) environment byXPD = 5 dB θi = 10 σ = 15

As antennas will be implemented on a mobile terminala study of the effect of the antennas orientation on the

correlation has been done Three configurations of rotationshave been studied rotation of antenna around axis A andaround axis B for two initial positions horizontal andvertical as shown in Figure 11

For each configuration the envelope correlation coef-ficient for the three meander bend endings antennas withadded ground plane and slot has been calculated fromsimulated radiation patterns Minimum and maximumvalues at center frequencies of the two bands 777 MHz and26 GHz are reported in Table 1

For isotropic environment a very low correlation isobserved in the two bands as a result of good matching(|S11| lt minus10 dB) a high isolation level (|S21| lt minus10 dB)and orthogonality between radiation patterns especiallyin the lower band In addition polarization diversity isnaturally achieved because of the orthogonal positions ofboth antennas

For the other cases maximum values of the correlationenvelope coefficient ρe are close to 05 for outdoor environ-ment whatever the distribution Indeed the incoming wavesare mainly along Eθ which implies less diversity in someantennarsquos position

When XPD gets close to 0 dB (indoor environmentXPD = 1 dB) Eθ and Eϕ values are almost the same Becausethese two components are uncorrelated by definition andbecause each antenna receives preferentially one of eachcomponent the correlation is getting low


05

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

(a)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 0

5

(b)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn5

φ = 90

(c)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 0

5

Copolar simulationCopolar measurementCross-polar simulationCross-polar measurement

(d)

Figure 9 Simulated and measured radiation patterns for port 1 (dB) (a) in the E plane (YZ plane) at 778 MHz (b) in the H plane (XZplane) at the 778 MHz (c) in the E plane (YZ plane) at 26 GHz and (d) in the H plane (XZ plane) at 26 GHz

For rotation around axis A minimum values of ρe areobtained for position at which one antenna receives only Eθcomponent of the incoming waves while the other one onlyEϕ component

For rotation around axis B for both configurations(b and c) minimum values are obtained when the tworadiating elements are positioned on AB plane Indeed at

these positions the radiation diversity is exploited as shownin Figure 10 and thus a low correlation is obtained

Finally for most configurations envelope correlationcoefficient is less than 05 which provides high diversitycapabilities [13] This result has been achieved thanksto the two orthogonal and identical antennas which arespatially separated It can provide for either or both spatial


0

5

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

(a)

05

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

Port 1Port 2

(b)

Figure 10 Simulated realized gain (dB) on the XY plane for the twobands (a) at 778 MHz (b) at 26 GHz

and pattern diversity In addition polarization diversity isavailable in the Z-direction

In the following part we evaluate the MEG which wasintroduced by Taga [17] It is defined as the ratio between themean received power of antennas over the random route andthe total mean incident power When each monopole receivesthe same quantity of power the MEG ratio (R) of the twoantennas is equal to one which means that no performancedeterioration is expected due to some power imbalance [18]

A

B

C

(a)

A

B

C

(b)

A

B

C

(c)

Figure 11 Rotation of antenna (a) around axis A (b) around axisB (horizontal antenna position) and (c) around axis B (verticalantenna position)

The mathematical expression is given by the followingequation

MEG =int

Ω

(XPD

XPD + 1GθPθ +

1XPD + 1

GϕPϕ

)

dΩ (2)

where Gθ and Gϕ are the θ and ϕ components of theantenna power gain pattern respectively The calculatedmean effective gains of the monopoles from simulatedradiation patterns at 777 MHz and 26 GHz are presented inTable 2

The Maximum values of the ratio (R) of MEG1 deter-mined at port 1 over MEG2 determined at port 2 areequal to 1 which satisfy an equal contribution of thetwo monopoles to receive the same quantity of powerThe proposed structure is completely symmetric and theGaussian and Laplacian angular distributions are taken onlyalong the elevation as presented in [15] In addition theincident power in the outdoor environment (or indoor) isconcentrated around 10 (or 20) above the horizon withan aperture of 30 (or 60) and for these directions bothantennas receive an equal amount of power

Minimum values of ratio (R) are obtained for positionsat which the Eθ (or Eϕ) components of the two antennashave different levels in the directions of incident power


Table 1 Coefficients of correlation for the two bands for all environments of the proposed structure

Rotation Distribution777 MHz 26 GHz

ρemin ρemax ρemin ρemax

Whatever Isotropic 710minus5 410minus3

around A

G-In 002 010 10minus3 008

G-Out 020 042 710minus4 039

L-In 007 016 10minus5 022

L-Out 026 049 310minus4 051

around B (horizontal position)

G-In 710minus4 010 510minus4 810minus3

G-Out 10minus3 042 410minus5 110minus2

L-In 210minus3 016 10minus5 310minus3

L-Out 310minus3 046 10minus4 510minus3

around B (vertical position)

G-In 510minus5 005 210minus4 007

G-Out 10minus4 040 410minus3 039

L-In 210minus4 010 810minus5 022

L-Out 610minus4 049 710minus4 054

Table 2 MEG ratio (R) for the two bands for all environments ofthe proposed structure


Rmin Rmax Rmin Rmax

Whatever Isotropic 1 1

around A

G-In 070 1 078 1

G-Out 035 1 071 1

L-In 063 1 046 1

L-Out 030 1 058 1

G-In 099 1 099 1

around B G-Out 097 1 098 1

(horizontal position) L-In 099 1 099 1

L-Out 096 1 098 1

G-In 094 1 080 1


(vertical position) L-In 085 1 033 1

L-Out 070 1 048 1

For example if antenna 1 presents a low Eθ component whereantenna 2 a high one an unbalanced power is obtained

For most configurations ratio (R) is greater than 05which is acceptable to provide high diversity capabilities [18]

5 Conclusion

In this paper a compact dual-band dual-polarized antennafor LTE applications is proposed with an extension of thelower band towards TV White Space band to provide radio-cognitive capabilities to the terminal A design provides dualpolarizations in both of the bands 700ndash862 MHz and 25ndash269 GHz with good impedance matching (|S11| lt minus10 dB)

Measurement results are in good agreement with sim-ulated ones In addition good performances are obtainedby calculating the envelope correlation coefficient and theMEG ratio for several antennasrsquo positions in different

environments isotropic indoor and outdoor For mostconfigurations it is found that the system satisfies thecondition ρe lt 05 and MEG1MEG2 gt 05 Thus thepresented design is suitable for MIMO communicationapplications and thus enables the SNR value at the terminalside to be maximized

Acknowledgment

The research leading to these results has received fundingfrom the European Communityrsquos Seventh Framework Pro-gram (FP72007ndash2013) under Grant agreement SACRA no249060

References

[1] 3rd Generation Partnership Project Technical SpecificationGroup Radio Access Network Evolved Universal TerrestrialRadio Access (E-UTRA) Radio Resource Control (RRC)Protocol Specification 3GPP TS 36 331

[2] R G Vaughan ldquoPolarization diversity in mobile communica-tionsrdquo IEEE Transactions on Vehicular Technology vol 39 no3 pp 177ndash186 1990

[3] K Ogawa and T Uwano ldquoDiversity antenna for very small800-MHz band portable telephonesrdquo IEEE Transactions onAntennas and Propagation vol 42 no 9 pp 1342ndash1345 1994

[4] J W Wallace M A Jensen A L Swindlehurst and B DJeffs ldquoExperimental characterization of the MIMO wirelesschannel data acquisition and analysisrdquo IEEE Transactions onWireless Communications vol 2 no 2 pp 335ndash343 2003

[5] S Hienonen A Lehto and A V Raisanen ldquoSimple broad-band dual-polarized aperture-coupled microstrip antennardquoin Proceedings of the IEEE Antennas and Propagation SocietyInternational Symposium vol 2 pp 1228ndash1231 Orlando FlaUSA August 1999

[6] P Brachat and J M Baracco ldquoPrinted radiating element withtwo highly decoupled input portsrdquo Electronics Letters vol 31no 4 pp 245ndash246 1995

[7] Y L Kuo and K L Wong ldquoDual-polarized monopole antennafor WLAN applicationrdquo in Proceedings of the IEEE Antennas


and Propagation Society International Symposium vol 4 pp80ndash83 June 2002

[8] C Yang Y Yao J Yu and X Chen ldquoNovel compact multibandMIMO antenna for mobile terminalrdquo International Journalof Antennas and Propagation vol 2012 Article ID 691681 9pages 2012

[9] T W Chiou and K L Wong ldquoA compact dual-band dual-polarized patch antenna for 9001800-MHz cellular systemsrdquoIEEE Transactions on Antennas and Propagation vol 51 no 8pp 1936ndash1940 2003

[10] K S Kim T Kim and J Choi ldquoDual-frequency aperture-coupled square patch antenna with double notchesrdquoMicrowave and Optical Technology Letters vol 24 no 6 pp370ndash374 2000

[11] SACRA European Project (FP7 2007ndash2013) httpwwwict-sacraeu

[12] K J Kim W G Lim and J W Yu ldquoHigh isolation internaldual-band planar inverted-F antenna diversity system withband-notched slots for MIMO terminalsrdquo in Proceedings of the36th European Microwave Conference (EuMCrsquo06) pp 1414ndash1417 Manchester UK September 2006

[13] R G Vaughan and J B Andersen ldquoAntenna diversityin mobile communicationrdquo IEEE Transactions on VehicularTechnology vol 36 no 4 pp 149ndash172 1987

[14] K Kalliola K Sulonen H Laitinen O Kivekas J Krogerusand P Vainikainen ldquoAngular power distribution and meaneffective gain of mobile antenna in different propagationenvironmentsrdquo IEEE Transactions on Vehicular Technology vol51 no 5 pp 823ndash838 2002

[15] F Adachi M T Feeney A G Williamson and J D ParsonsldquoCrosscorrelation between the envelopes of 900 MHz signalsreceived at a mobile radio base station siterdquo Proceedings of IEEon Communications Radar and Signal Processing Part F vol133 no 6 pp 506ndash512 1986

[16] Z Ying T Bolin V Plicanic A Derneryd and G KristenssonldquoDiversity antenna terminal evaluationrdquo in Proceedings ofthe IEEE Antennas and Propagation Society InternationalSymposium and USNCURSI Meeting pp 375ndash378 July 2005


[18] W C T Brown Antenna diversity for mobile terminal[PhD thesis] University of Surrey 2002 httpepubssurreyacuk2125


Research Article

Band-Notched Ultrawide Band Planar Inverted-F Antenna

H T Chattha1 M K Ishfaq2 Y Saleem3 Y Huang4 and S J Boyes4

1 Department of Electrical Engineering University of Engineering and Technology Lahore Faisalabad Campus Faisalabad Pakistan2 Department of Electrical Engineering GC University Faisalabad Pakistan3 Department of Computer Science and Engineering University of Engineering and Technology Lahore Pakistan4 Department of Electrical Engineering and Electronics University of Liverpool Liverpool L69 3GJ UK

Correspondence should be addressed to H T Chattha chattha43hotmailcom

Received 25 February 2012 Accepted 9 April 2012


Copyright copy 2012 H T Chattha et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

A novel ultrawide band planar inverted-F antenna with band-notched characteristics is presented in this paper The planarinverted-F antenna uses two parasitic elements to enhance the bandwidth to cover the ultrawide band The band-notched featureis added by inserting a W-shaped slot on the top radiating element of the antenna with a band rejection from 508 to 6 GHz(measured) Both the measured and simulated results are obtained to draw the conclusions

1 Introduction

A considerable amount of research has been conducted fordeveloping the ultrawide band (UWB) antennas for its char-acteristics such as high data rate being low power and havingwide bandwidths and simple hardware structure in manyreal world applications In 2002 Federal CommunicationsCommission (FCC) of USA allocated a bandwidth from31 GHz to 106 GHz to ultrawide band [1] This band alsocovers the bands of the previously present wireless networkswith standards such as IEEE 802 11a in USA (515ndash535 GHz5725ndash5825 GHz) HIPERLAN2 in Europe (515ndash535 GHz547ndash5725 GHz) and Microwave Access (WiMAX) system(525ndash5825 GHz) [2 3] To avoid the interference betweenthese UWB systems and the nearby communication systemssuch as wireless (WLAN) there is a need to employ someform of filter In order to save the space and cost and reducethe complexity of the UWB system this filter should ideallybe integrated into the radiating element of the antennaTo tackle this issue many printed type of antennas withband-notched characteristics have been presented [2ndash10]All these antennas have almost omnidirectional radiationpatterns however some UWB applications require antennaswith comparatively higher directivity

The planar inverted-F antenna (PIFA) is now widelyused in mobile and portable radio applications due toits simple design lightweight low cost conformal naturereliable performance and attractive radiation pattern [11ndash14] The PIFA has higher directivity as compared to theplanar monopole antennas which makes it more suitable forcertain UWB applications [15] PIFA was previously knownas an antenna having narrow-band characteristics and areasonable research is already done to enhance its impedancebandwidth [16ndash18] Feik et al have shown in [19] that thefractional impedance bandwidth up to about 25 can beobtained by having different shapes of feed plates Recentlysome UWB PIFA antennas are introduced [20 21] and oneband-notched UWB PIFA is introduced using a spiral slot[22] on the feed plate However the UWB PIFAs presented inthe [21 22] have height h = 75 mm which is relatively highand also it has two PIFA antennas (one on each edge of theground plane) to cover the whole UWB band which makes itvery difficult to integrate with other PCB components Thispaper presents a single-element band-notched UWB PIFAfor height h = 45 mm by introducing a W-shaped slot onthe top radiating plate


Dc1 Dc

tc1

tc2

tc3

Feed plateh

Feed

d

W

L

X

YZ

t

Ground plane

Wg

Top plate

Lb

Lg

Ws

Wf

CL

x2

x3x3

y2y2y3

x1

x2x2

y1

Figure 1 PIFA geometry

Figure 2 The built PIFA with SMA connector

3 4 5 6 7 8 9 10 11 12minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

SimulatedMeasured

Figure 3 Ref coefficients S11 (dB) versus frequency (GHz)

2 Antenna Configurations

The structure of the designed PIFA as shown in Figure 1 has aradiating top plate with dimensions of width W and length Land Wg and Lg are width and length of the ground plane The

3 4 5 6 7 8 9 10 11 12minus20

minus15

minus10

minus5

0

5

10

Frequency (GHz)

Peak

gai

n (

dB)

SimulatedMeasured

Figure 4 Simulated peak gain (dB) versus frequency (GHz)

dimensions of the shorting and feeding plates are Wstimes(h+t)and Wf times h respectively having a horizontal distance ofLb between them and h is the height of the antenna havingair in the space between the top plate and the substrateThe distance of the parasitic element having a shape ofan inverted-L from the feeding plate is Dc This parasiticelement has a thickness of tc1 The horizontal extension ofthis element is CL having a thickness of tc2 Second parasiticelement rectangular in shape is inserted at the upper edge ofthe ground plane at a distance Dc1 from the shorting plateThe width of this element is tc3 The heights of both theparasitic elements are the same and is equal to (h + t) minus dhaving a vertical distance of d between the elements and theradiating plate The W-shaped slot on the top plate is insertedat a distance x1 from the side edge and at a distance of y1

from the upper edge of the top plate The W-shaped slot isshown separately in Figure 1 to highlight its dimensions Thethickness of the slot is 05 mm The feeding to the PIFA isprovided by a coaxial cable with an SMA connector as shownin Figure 2


3 4 5 6 7 8 9 10 11 12

050

100150200250

Frequency (GHz)Im

peda

nce

Z

Real componentImaginary component

200minus

150minus

100minus

50minus

Figure 5 Impedances Z(Ω) versus frequency (GHz)

Gain total

42075e+00039516e+00036958e+00034399e+00031841e+00029283e+00026724e+00024166e+00021608e+00019049e+00016491e+00013933e+00011374e+00088158eminus00162574eminus00136991eminus00111407eminus001

Z

Y

Φ

X

θ

Figure 6 Simulated 3D radiation pattern of PIFA at 75 GHz

3 Results

The optimization of the entire antenna parameters isperformed through parametric study in high frequencystructure simulator (HFSS) in order to get the maximumimpedance bandwidth and feed is provided at the upperedge of the ground plane The optimized values of all theparameters are found as follows Wg = 185 mm h =45 mm Lg = 28 mm W = 185 mm L = 10 mm Wf =85 mm Ws = 05 mm Lb = 55 mm Dc = 05 mm Dc1 =007λ = 3 mm tc1 = tc2 = tc3 = 05 mm d = 05 mm(h + t) minus d = 5 mm CL = 25 mm x1 = 1 mm y1 = 1 mmx2 = 25 mm x3 = 2 mm y1 = 1 mm y2 = 75 mm andy3 = 42 mm

The simulated and experimental results of the reflectioncoefficient are shown in Figure 3 It is evident that thebandwidth achieved by these techniques of inserting parasiticelements for S11 lt minus10 dB is extremely broad from about34 to 112 GHz The lower frequency and first resonance iscontrolled by the main structure of PIFA whereas the inser-tion of inverted-L-shaped parasitic element creates a secondresonance at 65 GHz and the presence of rectangular-shaped parasitic element produces a third resonance around

107 GHz (simulated) Due to the insertion of W-shaped slotband-notched characteristics are introduced with a bandrejection from 508 to 6 GHz (measured) The simulatedand measured results are generally in good agreement Theirdifferences are mainly due to the cables and connectorswhich are not being involved in the simulations but existin the measurements and the manufacturing tolerance ingetting the accurate parameters in the manual fabricationof this antenna Figure 4 shows the simulated and measuredpeak gain of the band-notched PIFA as a function offrequency in GHz A sharp decrease in peak gain is observedin the notched frequency band centered at around 53 GHz(measured) which confirms that this antenna provides agood level of rejection to signals at frequencies within thenotched band The impedance Z of this PIFA versus thefrequency in GHz is shown in Figure 5 to get a betterunderstanding of this antenna

The simulated 3D radiation pattern (polar plot) ofthe band-notched PIFA at 75 GHz is shown in Figure 6and the measured 2D radiation patterns of this antennaare shown in Figure 7 Figure 8 shows the simulated time-domain response of the PIFA to an input pulse which affirmsthe suitability of the PIFA for UWB applications


015

30

45

60

75

90

105

120

135

150

165plusmn180

minus15

minus30

minus45

minus60

minus75

minus90

minus10 5

minus120

minus135

minus150

minus1650

minus10

minus20

Φ = 0 for f = 45GHzΦ = 0 for f = 75GHz

Φ = 0 for f = 105GHz

(a)

0

15

30

45

607590105

120

135

150

165

plusmn180

minus15

minus30

minus45

minus60


minus120

minus135

minus150

minus165

5minus5

minus15

=90 for f = 45 GHzθ


=90 for f = 105GHzθ

(b)

Figure 7 (a) 2D rad pattern with total gain in dB for elevation XZ plane (Φ = 0) for diff frequencies (b) 2D rad pattern with total gainin dB for azimuth XY plane (θ = 90) for different frequencies

0 05 1 15 2 25 3 35 4minus08minus06minus04minus02

002040608

1

Time (ns)

Mag

nit

ude

Input signalOutput signal

Figure 8 The time domain response of PIFA

4 Parametric Study

The parameters of the W-shaped slot are varied to observeits effects on the band-notched characteristics of the PIFAantenna The distance x1 from the side edge of the top plateis varied from 1 mm to 10 mm while all other parametersare held constant Figure 9 shows the variation of x1 versusthe frequency in GHz which makes it obvious that positionof the W-slot on the top plate does not significantly affectsthe notched band of the PIFA but significantly affects theperformance of PIFA over the UWB band

Similarly the length of the outer legs of the W-shapedslot y2 is varied from 5 mm to 7 mm to observe its effectsFigure 10 shows that varying the length y2 changes the band

3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

x1 = 1 mmx1 = 4 mm

x1 = 7 mmx1 = 10 mm

Figure 9 Simulated S11 (dB) for values of x1 versus frequency(GHz)

which is notched by the insertion of W-shaped slot whereasit does not significantly affect the performance of the PIFAover the UWB band Therefore we can vary the length y2

to change the band to be notched In the similar way thelength of the inner legs of the W-shaped slot y3 is also variedfrom 2 mm to 6 mm to observe its effects on the performanceof the W-shaped slot and on the overall performance of thePIFA It is obvious as shown in Figure 11 that the length y3 isvery critical parameter to decide which band is exactly to benotched Varying the length y3 also varies the performanceof the PIFA over the UWB band Therefore an appropriateand optimized value of y3 is required to get the exact band


3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

y2 = 5 mmy2 = 6 mmy2 = 7 mm

Figure 10 Simulated S11 (dB) for values of y2 versus frequency(GHz)

3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

y3 = 2 mmy3 = 3 mmy3 = 42 mm

y3 = 5 mmy3 = 6 mm


to be notched by the W-shaped slot and also to achieve thereflection coefficient below minus10 dB over the UWB band

5 Conclusions

A band-notched UWB PIFA antenna is presented in thispaper It has been shown that a very wide bandwidth isachieved which almost covers the whole UWB band from34 to 112 GHz and a band rejection from 508 to 6 GHz isachieved by inserting a W-shaped slot on the top plate

References

[1] ldquoFCC first report and order on ultra-wideband technologyrdquoFebruary 2002

[2] Y D Dong W Hong Z Q Kuai et al ldquoDevelopment ofultrawideband antenna with multiple band-notched charac-teristics using half mode substrate integrated waveguide cavitytechnologyrdquo IEEE Transactions on Antennas and Propagationvol 56 no 9 pp 2894ndash2902 2008

[3] Q X Chu and Y Y Yang ldquoA compact ultrawideband antennawith 3455 GHz dual band-notched characteristicsrdquo IEEE

Transactions on Antennas and Propagation vol 56 no 12 pp3637ndash3644 2008

[4] J R Kelly P S Hall and P Gardner ldquoPlanar band-notchedUWB antennardquo in Proceedings of the 3rd European Conferenceon Antennas and Propagation (EuCAP rsquo09) pp 1636ndash1639March 2009

[5] Y J Cho K H Kim D H Choi S S Lee and S O Park ldquoAminiature UWB planar monopole antenna with 5-GHz band-rejection filter and the time-domain characteristicsrdquo IEEETransactions on Antennas and Propagation vol 54 no 5 pp1453ndash1460 2006

[6] A J Kerkhoff and H Ling ldquoDesign of a band-notched planarnonopole antenna using genetic algorithm optimizationrdquoIEEE Transactions on Antennas and Propagation vol 55 no3 pp 604ndash610 2007

[7] S J Wu C H Kang K H Chen and J H Tarng ldquoStudyof an ultrawideband monopole antenna with a band-notchedopen-looped resonatorrdquo IEEE Transactions on Antennas andPropagation vol 58 no 6 pp 1890ndash1897 2010

[8] J Qiu Z Du J Lu and K Gong ldquoA planar monopole antennadesign with band-notched characteristicrdquo IEEE Transactionson Antennas and Propagation vol 54 no 1 pp 288ndash292 2006

[9] W S Lee W G Lim and J W Yu ldquoMultiple band-notchedplanar monopole antenna for multiband wireless systemsrdquoIEEE Microwave and Wireless Components Letters vol 15 no9 pp 576ndash578 2005

[10] S W Qu J L Li and Q Xue ldquoA band-notched ultrawidebandprinted monopole antennardquo IEEE Antennas and WirelessPropagation Letters vol 5 no 1 pp 495ndash498 2006

[11] K Hirasawa and M Haneishi Analysis Design and Measure-ment of Small and Low-Profile Antennas Artech House 1992

[12] K L Virga and Y Rahmat-Samii ldquoLow-profile enhanced-B and width PIFA antennas for wireless communicationspackagingrdquo IEEE Transactions on Microwave Theory andTechniques vol 45 no 10 pp 1879ndash1888 1997

[13] P S Hall E Lee and C T P Song ldquoPlanar inverted-F antennas chapter 7rdquo in Printed Antennas for WirelessCommunications R Waterhouse Ed John Wiley amp Sons2007

[14] Y Huang and K Boyle Antennas from Theory to Practice JohnWiley amp Sons 2008

[15] H T Chattha Y Huang M K Ishfaq and S J Boyes ldquoA com-prehensive parametric study of planar inverted-F antennardquoScientific Research Wireless Engineering and Technology vol 3no 1 pp 1ndash11 2012

[16] D Liu and B Gaucher The Inverted-F Antenna Height Effectson Bandwidth IEEE IBM T J Watson Research CentreYorktown Heights NY USA 2005

[17] F Wang Z Du Q Wang and K Gong ldquoEnhanced-bandwidthPIFA with T-shaped ground planerdquo Electronics Letters vol 40no 23 pp 1504ndash1505 2004

[18] P W Chan H Wong and E K N Yung ldquoWidebandplanar inverted-F antenna with meandering shorting striprdquoElectronics Letters vol 44 no 6 pp 395ndash396 2008

[19] R Feick H Carrasco M Olmos and H D Hristov ldquoPIFAinput bandwidth enhancement by changing feed plate silhou-etterdquo Electronics Letters vol 40 no 15 pp 921ndash923 2004

[20] H T Chattha Y Huang Y Lu and X Zhu ldquoAn ultra-wideband planar inverted-F antennardquo Microwave and OpticalTechnology Letters vol 52 no 10 pp 2285ndash2288 2010

[21] C H See R A Abd-Alhameed D Zhou H I Hraga P SExcell and M B Child ldquoUltra-wideband planar inverted FFantennardquo Electronics Letters vol 46 no 8 pp 549ndash550 2010


[22] H I Hraga C H See R A Abd-Alhameed et al ldquoPIFAantenna for UWB applications with WLAN band rejectionusing spiral slotsrdquo in Proceedings of the 5th European Confer-ence on Antennas and Propagation (EUCAP rsquo11) pp 2226ndash2229 April 2011







Editorial Board




Contents





































1 Introduction









































































Device under test







Mode stirrers


Turntable


Calibration antenna

Walls of reflective

material

Test object (DUT)


Access panel





or a communication

tester















Anechoic chamber

DUT

Absorbers

Measurement antenna

VNA

(a)


DUT

Mode stirrer

Fixed antenna

VNA

(b)



DUT

Absorbers


Channel


(a)


MIMO link

DUT

Mode stirrer

Channel


Fixed antennas

(b)


















anechoic chamber No












Zoom

Zoom times 3


Feeding port

Shorting port

(a)

FM port

GSM port

Filter Switch

to the PIFArsquos

feeding port

to the PIFArsquos

shorting portL1

(b)







0

1

2

3

4

5

6

7

8

9

10

MonopoleHilbertPIFA

7374 72

Sign

al quality












(i) geometry based

(ii) material based





Microstrip line

Clearance area


0

10

20

30

40

50

60

70

80

90

100

1

15

2

25

3

35

4

45

5

55

6

2 21 22 23 24 25 26 27 28 29 3

To

tal

effi

cie

ncy (

)

SW

R

Frequency (GHz)

VSWR

Total efficiency ()











Driven



















Slot

λ4 band 2

(a)

PIFA

λ4 band 1

(b)

PIFA + slot

(c)

08 09 1 11 12 13 14 15 16 17 18 19 2 21 22 23 24 25

Frequency (GHz)

Refl

ecti

on

co

effi

cie

nt

(d

B)

minus14

minus13

minus12

minus11

minus10

minus9

minus8

minus7

minus6

minus5

minus4

minus3

minus2

minus1

0

DB(|S(11)|)

PIFA band 1

DB(|S(22)|)

Slot band 2

DB(|S(33)|)

PIFA + slot

088 GHz

minus789 dB

096 GHz

minus601 dB 217 GHz

minus639 dB

171 GHz

minus643 dB

(d)


Phase delay

Printed circuit

board

Antenna 1 Antenna 2







(a) (b)


35

3

13

23

46

10

2021

48

19

25 25

0

5

10

15

20

25

30

35

40

45

50


Average a

nte

nn

a

effi

cie

ncy (

)

1 monopole

2 monopoles

3 monopoles





WPCB

LPCB

dgap

L





0

10

20

30

40

50

60

70

80

90

100

Effi

cie

ncy (

)

Frequency (GHz)


16

2

16

6

17

1

17

5

17

9

18

3

18

7

19

1

19

5

19

9

20

4

20

8

21

2

21

6

22

22

4

22

8

23

2

23

7

24

1

24

5

24

9

25

3

25

7

26

1

26

5

27












(a) (b)


(a)

(b) (c)












minus14

minus12

minus10

minus8

minus6

minus4

minus2

0

S1

1(d

B)

07

08

09 1

11

12

13

14

15

16

17

18

19 2

21

22

23

24

25

Frequency (GHz)

Dual-band PIFA

Quad-band PIFA

Multiband PIFA

082 GHz

minus6 dB

103 GHz

minus6 dB

176 GHzminus6 dB

184 GHz

minus6 dB

197 GHz

minus6 dB

25 GHz

minus5 dB

212 GHz

minus61 dB






0

5

10

15

20

25

30

35

40

45

50

55

820 890 920 960

Frequency (MHz)

Measu

red

to

tal

effi

cie

ncy (

)

Without strip









Main PCB

Upper PCB

x

y

z

(a)

Trap

x

z

y

(b)

















Transmission line





5 Conclusions






Acknowledgments


References


























































































































1 Introduction







1 +07

1 +06

1 +05

1 +04

1 +03

1 +02

1 +01

1 +00

1 +02

1 +01

1 +00

1 minus 01

1 minus 02

1+

2

1+

3

1+

4

1+

5

1+

6

1+

7

1+

8

1+

9

1+

10

1+

11

Permittivity

Conductivity

Frequency




119875119875abs = 10045601004560119881119881

12120590120590|119864119864|2119889119889119881119881119889 (1)


SAR = 10076531007653 1205901205902120588120588100766910076691198641198642119894119894 119889 (2)





3 Simulations





Brain

CSF

Dura

Bone

Fat

Skin


F 5 Dipole antenna



ickness(mm)


Shell 46 0 5







19525 +002

18313 +002

171e+002

15887 +002

14674 +002

13462 +002

12249 +002

11036 +002

98234 +001

86107 +001

73979 +001

61852 +001

49724 +001

37597 +001

2547 +001

13342 +001

1215 +000

E field (Vm)


20

18

16

14

12

10

8

6

4

2

00 10 20 30 40 50 60

Distance (mm)

Local SAR

Average SAR

SAR (wattkg)




4 Results



E field (Vm)

12126 +002

11368 +002

10611 +002

98528 +001

9095 +001

83372 +001

75795 +001

68217 +001

60639 +001

53062 +001

45484 +001

37906 +001

30329 +001

22751 +001

15173 +001

75958 +000

18184 minus 002


E field (Vm)

13783 +002

12922 +002

12062 +002

11201 +002

10341 +002

94802 +001

86196 +001

7759 +001

68984 +001

60379 +001

51773 +001

43167 +001

34561 +001

25956 +001

1735 +001

8744 +00013818 minus 001




Fat 177times 102

Bone 167times 102

Dura 140times 102

Csf 137times 102

Brain 121times 102


E field (Vm)

14096 +002

13216 +002

12336 +002

11457 +002

10577 +002

96977 +001

88181 +001

79385 +001

70589 +001

61793 +001

52997 +001

44201 +001

35405 +001

26609 +001

17813 +001

90171 +00022117 minus 001


E field (Vm)

1672 +002

15676 +002

14631 +002

13586 +002

12541 +002

11496 +002

10451 +002

94058 +001

83609 +001

73159 +001

6271 +001

5226 +001

41811 +001

31361 +001

20912 +001

10462 +00112991 minus 002









Six layers + dipole












E field (Vm)

17717 +002

16612 +002

15507 +002

14401 +002

13296 +002

12191 +002

11086 +002

99804 +001

88752 +001

777e+001

66647 +001

55595 +001

44542 +001

3349e+001

22438 +001

11385 +00133291 minus 001






E field (Vm)

18019 +002

16896 +002

15774 +002

14651 +002

13529 +002

12407 +002

11284 +002

10162 +002

90396 +001

79172 +001

67948 +001

56724 +001

455 +001

34276 +001

23052 +001

11828 +00160453 minus 001


75 mm

10 mm Antenna patch

62 mm

33 mm

Ground plane

50 mm

92 mm








Six layers + PIFA











E field (Vm)

52151 minus 001

48999 minus 001

45847 minus 001

42695 minus 001

39542 minus 001

3639 minus 001

33238 minus 001

30086 minus 001

26934 minus 001

23782 minus 001

2063 minus 001

17478 minus 001

14326 minus 001

11174 minus 001

80214 minus 002

48693 minus 002

17172 minus 002


E field (Vm)

62004 minus 001

58225 minus 001

54445 minus 001

50665 minus 001

46886 minus 001

43106 minus 001

39327 minus 001

35547 minus 001

31768 minus 001

27988 minus 001

24209 minus 001

20429 minus 001

1665 minus 001

1287 minus 001

90907 minus 002

53112 minus 002

15317 minus 002



E field (Vm)

64315 minus 001

60437 minus 001

56559 minus 001

52682 minus 001

48804 minus 001

44926 minus 001

41048 minus 001

3717 minus 001

33292 minus 001

29415 minus 001

25537 minus 001

21659 minus 001

17781 minus 001

13903 minus 001

10025 minus 001

61475 minus 002

22696 minus 002


E field (Vm)

7482 minus 001

70223 minus 001

65625 minus 001

61027 minus 001

56430 minus 001

51832 minus 001

47234 minus 001

42637 minus 001

38039 minus 001

33441 minus 001

28844 minus 001

24246 minus 001

19648 minus 001

15051 minus 001

10453 minus 001

58554 minus 002

12578 minus 002




5 Conclusions


E field (Vm)

78242 minus 001

73529 minus 001

68816 minus 001

64103 minus 001

5939 minus 001

54678 minus 001

49965 minus 001

45252 minus 001

40539 minus 001

35826 minus 001

31114 minus 001

26401 minus 001

21688 minus 001

16975 minus 001

12262 minus 001

75495 minus 002

28367 minus 002



75133 minus 001

70341 minus 001

65549 minus 001

60757 minus 001

55965 minus 001

51173 minus 001

46381 minus 001

41589 minus 001

36798 minus 001

32006 minus 001

27214 minus 001

22422 minus 001

17630 minus 001

12838 minus 001

80463 minus 002

32544 minus 002





References




















Research Article











2is displaced


1 Introduction




Anetnna part



θ ϕ

z

x

y

100 times 60 mm2

main ground


25

B

A Via to a 50Ω SMA

10



ground

(a)

L = 12 nH

65

7

3

2

1

23

15

45

m = 12

t = 225 05

A

B

Gap= 15

xy

z

(b)






2which can be







1 (d

B)

500 1000 1500 2000 2500


0

minus5

minus10

minus15

minus20

minus25

minus30

Frequency (MHz)






824

960

1710

2170

180

170

160

150

140

130

120110

100 90 8070

60

50

40

30

20

10

0

minus170

minus160

minus150

minus140

minus130

minus120


minus70minus60

minus50

minus40

minus30

minus20

minus10

00 02

02

05

05

1

2

1

2 5

5

minus02

minus05

minus1

minus2

minus5






500 1000 1500 2000 2500

Frequency (MHz)

ProposedRef1

Ref2minus6 dB

0

minus5

minus10

minus15

minus20

minus25

minus30S1

1 (d

B)


50000e+001

46500e+001

43000e+001

39500e+001

36000e+001

32500e+001

29000e+001

25500e+001

22000e+001

18500e+001

15000e+001

80000e+001

45000e+001

10000e+001

Jsurf (A per m)

(a) (b)


500 1000 1500 2000 2500 3000

Frequency (MHz)


minus30

minus25

minus20

minus15

minus10

minus5

0

S11

(dB

)



(a) (b)


S11

(dB

)

500 1000 1500 2000 2500

0

minus5

minus10

minus15

minus20

minus25

minus30

Frequency (MHz)

L = 39 nHL = 82 nH

L = 15 nHminus6 dB

(a)

Frequency (MHz)

500 1000 1500 2000 2500

S11

(dB

)

0

minus5

minus10

minus15

minus20

minus25

minus30

m = 7 nH


(b)

Frequency (MHz)

500 1000 1500 2000 2500

S11

(dB

)

0

minus5

minus10

minus15

minus20

minus25

minus30

t = 39 nHt = 82 nH

t = 15 nHminus6 dB

(c)




minus10

0

10

10

0

0

90

180

270


xz-plane

(a)


minus10

0

10

10

0

0

90

180

270


yz-plane

(b)


800 820 840 860 880 900 920 940 960 9800

10

20

30

40

50

60

70

80

90

100

GSM850900

EfficiencyGain

Frequency (MHz)

Rad

iati

on E

ffici

ency

(

)

0

1

2

3

4

5

6

An

tenn

a Gain

(dBi)

minus2

minus1

(a)

1700 1800 1900 2000 2100 22000

10

20

30

40

50

60

70

80

90

100


EfficiencyGain

Frequency (MHz)

Rad

iati

on E

ffici

ency

(

)

0

1

2

3

4

5

6

An

tenn

a Gain

(dBi)

minus2

minus1

(b)





4 Conclusion



References
















Research Article









1 Introduction





2 Antenna Design



Lm

dWm

l

1 2

45XY

Z

Via holes

L2L1

(a)

Lgd2

Lgd1

Wgd1

Lgd3

Wgd2

Ls

Ws

Added ground plane

Removed corners

Wgd3

(b)






07 075 08 085 09 095 1minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)


|S ij|(

dB)

(a)

2 22 24 26 28 3minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)


|S ij|(

dB)

(b)









|S11| without slot


2 22 24 26 28 3minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)






1 bend ending

3 bend endings

05 06 07 08 09 1minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)






Ls

Ws

S

45XY

Z

1 2


|S11| simulation

|S21| simulation

|S11|measurement

|S21|measurement

1 15 2 25minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)



07 075 08 08550

60

70

80

90

100

Frequency (GHz)





25 255 26 265 27

Frequency (GHz)

50

60

70

80

90

100




ρe =∣∣∣

int

Ω

(


lowast2ϕpϕ

)

dΩ∣∣∣

2

int

Ω

(


lowast1ϕpϕ

)

dΩint

Ω

(


lowast2ϕpϕ

)

dΩ (1)











05

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

(a)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 0

5

(b)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn5

φ = 90

(c)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 0

5


(d)







0

5

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

(a)

05

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

Port 1Port 2

(b)




A

B

C

(a)

A

B

C

(b)

A

B

C

(c)



MEG =int

Ω

(XPD

XPD + 1GθPθ +

1XPD + 1

GϕPϕ

)

dΩ (2)









around A

G-In 002 010 10minus3 008

G-Out 020 042 710minus4 039

L-In 007 016 10minus5 022

L-Out 026 049 310minus4 051













Rmin Rmax Rmin Rmax


around A

G-In 070 1 078 1

G-Out 035 1 071 1

L-In 063 1 046 1

L-Out 030 1 058 1

G-In 099 1 099 1



L-Out 096 1 098 1

G-In 094 1 080 1



L-Out 070 1 048 1



5 Conclusion




Acknowledgment


References






















Research Article









1 Introduction




Dc1 Dc

tc1

tc2

tc3

Feed plateh

Feed

d

W

L

X

YZ

t

Ground plane

Wg

Top plate

Lb

Lg

Ws

Wf

CL

x2

x3x3

y2y2y3

x1

x2x2

y1



3 4 5 6 7 8 9 10 11 12minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

SimulatedMeasured




3 4 5 6 7 8 9 10 11 12minus20

minus15

minus10

minus5

0

5

10

Frequency (GHz)

Peak

gai

n (

dB)

SimulatedMeasured





3 4 5 6 7 8 9 10 11 12

050

100150200250

Frequency (GHz)Im

peda

nce

Z


200minus

150minus

100minus

50minus


Gain total


Z

Y

Φ

X

θ


3 Results






015

30

45

60

75

90

105

120

135

150

165plusmn180

minus15

minus30

minus45

minus60

minus75

minus90

minus10 5

minus120

minus135

minus150

minus1650

minus10

minus20



(a)

0

15

30

45

607590105

120

135

150

165

plusmn180

minus15

minus30

minus45

minus60


minus120

minus135

minus150

minus165

5minus5

minus15




(b)



002040608

1

Time (ns)

Mag

nit

ude



4 Parametric Study



3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

x1 = 1 mmx1 = 4 mm

x1 = 7 mmx1 = 10 mm





3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

y2 = 5 mmy2 = 6 mmy2 = 7 mm


3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

y3 = 2 mmy3 = 3 mmy3 = 42 mm

y3 = 5 mmy3 = 6 mm



5 Conclusions


References

























Contents





































1 Introduction









































































Device under test







Mode stirrers


Turntable


Calibration antenna

Walls of reflective

material

Test object (DUT)


Access panel





or a communication

tester















Anechoic chamber

DUT

Absorbers

Measurement antenna

VNA

(a)


DUT

Mode stirrer

Fixed antenna

VNA

(b)



DUT

Absorbers


Channel


(a)


MIMO link

DUT

Mode stirrer

Channel


Fixed antennas

(b)


















anechoic chamber No












Zoom

Zoom times 3


Feeding port

Shorting port

(a)

FM port

GSM port

Filter Switch

to the PIFArsquos

feeding port

to the PIFArsquos

shorting portL1

(b)







0

1

2

3

4

5

6

7

8

9

10

MonopoleHilbertPIFA

7374 72

Sign

al quality












(i) geometry based

(ii) material based





Microstrip line

Clearance area


0

10

20

30

40

50

60

70

80

90

100

1

15

2

25

3

35

4

45

5

55

6

2 21 22 23 24 25 26 27 28 29 3

To

tal

effi

cie

ncy (

)

SW

R

Frequency (GHz)

VSWR

Total efficiency ()











Driven



















Slot

λ4 band 2

(a)

PIFA

λ4 band 1

(b)

PIFA + slot

(c)

08 09 1 11 12 13 14 15 16 17 18 19 2 21 22 23 24 25

Frequency (GHz)

Refl

ecti

on

co

effi

cie

nt

(d

B)

minus14

minus13

minus12

minus11

minus10

minus9

minus8

minus7

minus6

minus5

minus4

minus3

minus2

minus1

0

DB(|S(11)|)

PIFA band 1

DB(|S(22)|)

Slot band 2

DB(|S(33)|)

PIFA + slot

088 GHz

minus789 dB

096 GHz

minus601 dB 217 GHz

minus639 dB

171 GHz

minus643 dB

(d)


Phase delay

Printed circuit

board

Antenna 1 Antenna 2







(a) (b)


35

3

13

23

46

10

2021

48

19

25 25

0

5

10

15

20

25

30

35

40

45

50


Average a

nte

nn

a

effi

cie

ncy (

)

1 monopole

2 monopoles

3 monopoles





WPCB

LPCB

dgap

L





0

10

20

30

40

50

60

70

80

90

100

Effi

cie

ncy (

)

Frequency (GHz)


16

2

16

6

17

1

17

5

17

9

18

3

18

7

19

1

19

5

19

9

20

4

20

8

21

2

21

6

22

22

4

22

8

23

2

23

7

24

1

24

5

24

9

25

3

25

7

26

1

26

5

27












(a) (b)


(a)

(b) (c)












minus14

minus12

minus10

minus8

minus6

minus4

minus2

0

S1

1(d

B)

07

08

09 1

11

12

13

14

15

16

17

18

19 2

21

22

23

24

25

Frequency (GHz)

Dual-band PIFA

Quad-band PIFA

Multiband PIFA

082 GHz

minus6 dB

103 GHz

minus6 dB

176 GHzminus6 dB

184 GHz

minus6 dB

197 GHz

minus6 dB

25 GHz

minus5 dB

212 GHz

minus61 dB






0

5

10

15

20

25

30

35

40

45

50

55

820 890 920 960

Frequency (MHz)

Measu

red

to

tal

effi

cie

ncy (

)

Without strip









Main PCB

Upper PCB

x

y

z

(a)

Trap

x

z

y

(b)

















Transmission line





5 Conclusions






Acknowledgments


References


























































































































1 Introduction







1 +07

1 +06

1 +05

1 +04

1 +03

1 +02

1 +01

1 +00

1 +02

1 +01

1 +00

1 minus 01

1 minus 02

1+

2

1+

3

1+

4

1+

5

1+

6

1+

7

1+

8

1+

9

1+

10

1+

11

Permittivity

Conductivity

Frequency




119875119875abs = 10045601004560119881119881

12120590120590|119864119864|2119889119889119881119881119889 (1)


SAR = 10076531007653 1205901205902120588120588100766910076691198641198642119894119894 119889 (2)





3 Simulations





Brain

CSF

Dura

Bone

Fat

Skin


F 5 Dipole antenna



ickness(mm)


Shell 46 0 5







19525 +002

18313 +002

171e+002

15887 +002

14674 +002

13462 +002

12249 +002

11036 +002

98234 +001

86107 +001

73979 +001

61852 +001

49724 +001

37597 +001

2547 +001

13342 +001

1215 +000

E field (Vm)


20

18

16

14

12

10

8

6

4

2

00 10 20 30 40 50 60

Distance (mm)

Local SAR

Average SAR

SAR (wattkg)




4 Results



E field (Vm)

12126 +002

11368 +002

10611 +002

98528 +001

9095 +001

83372 +001

75795 +001

68217 +001

60639 +001

53062 +001

45484 +001

37906 +001

30329 +001

22751 +001

15173 +001

75958 +000

18184 minus 002


E field (Vm)

13783 +002

12922 +002

12062 +002

11201 +002

10341 +002

94802 +001

86196 +001

7759 +001

68984 +001

60379 +001

51773 +001

43167 +001

34561 +001

25956 +001

1735 +001

8744 +00013818 minus 001




Fat 177times 102

Bone 167times 102

Dura 140times 102

Csf 137times 102

Brain 121times 102


E field (Vm)

14096 +002

13216 +002

12336 +002

11457 +002

10577 +002

96977 +001

88181 +001

79385 +001

70589 +001

61793 +001

52997 +001

44201 +001

35405 +001

26609 +001

17813 +001

90171 +00022117 minus 001


E field (Vm)

1672 +002

15676 +002

14631 +002

13586 +002

12541 +002

11496 +002

10451 +002

94058 +001

83609 +001

73159 +001

6271 +001

5226 +001

41811 +001

31361 +001

20912 +001

10462 +00112991 minus 002









Six layers + dipole












E field (Vm)

17717 +002

16612 +002

15507 +002

14401 +002

13296 +002

12191 +002

11086 +002

99804 +001

88752 +001

777e+001

66647 +001

55595 +001

44542 +001

3349e+001

22438 +001

11385 +00133291 minus 001






E field (Vm)

18019 +002

16896 +002

15774 +002

14651 +002

13529 +002

12407 +002

11284 +002

10162 +002

90396 +001

79172 +001

67948 +001

56724 +001

455 +001

34276 +001

23052 +001

11828 +00160453 minus 001


75 mm

10 mm Antenna patch

62 mm

33 mm

Ground plane

50 mm

92 mm








Six layers + PIFA











E field (Vm)

52151 minus 001

48999 minus 001

45847 minus 001

42695 minus 001

39542 minus 001

3639 minus 001

33238 minus 001

30086 minus 001

26934 minus 001

23782 minus 001

2063 minus 001

17478 minus 001

14326 minus 001

11174 minus 001

80214 minus 002

48693 minus 002

17172 minus 002


E field (Vm)

62004 minus 001

58225 minus 001

54445 minus 001

50665 minus 001

46886 minus 001

43106 minus 001

39327 minus 001

35547 minus 001

31768 minus 001

27988 minus 001

24209 minus 001

20429 minus 001

1665 minus 001

1287 minus 001

90907 minus 002

53112 minus 002

15317 minus 002



E field (Vm)

64315 minus 001

60437 minus 001

56559 minus 001

52682 minus 001

48804 minus 001

44926 minus 001

41048 minus 001

3717 minus 001

33292 minus 001

29415 minus 001

25537 minus 001

21659 minus 001

17781 minus 001

13903 minus 001

10025 minus 001

61475 minus 002

22696 minus 002


E field (Vm)

7482 minus 001

70223 minus 001

65625 minus 001

61027 minus 001

56430 minus 001

51832 minus 001

47234 minus 001

42637 minus 001

38039 minus 001

33441 minus 001

28844 minus 001

24246 minus 001

19648 minus 001

15051 minus 001

10453 minus 001

58554 minus 002

12578 minus 002




5 Conclusions


E field (Vm)

78242 minus 001

73529 minus 001

68816 minus 001

64103 minus 001

5939 minus 001

54678 minus 001

49965 minus 001

45252 minus 001

40539 minus 001

35826 minus 001

31114 minus 001

26401 minus 001

21688 minus 001

16975 minus 001

12262 minus 001

75495 minus 002

28367 minus 002



75133 minus 001

70341 minus 001

65549 minus 001

60757 minus 001

55965 minus 001

51173 minus 001

46381 minus 001

41589 minus 001

36798 minus 001

32006 minus 001

27214 minus 001

22422 minus 001

17630 minus 001

12838 minus 001

80463 minus 002

32544 minus 002





References




















Research Article











2is displaced


1 Introduction




Anetnna part



θ ϕ

z

x

y

100 times 60 mm2

main ground


25

B

A Via to a 50Ω SMA

10



ground

(a)

L = 12 nH

65

7

3

2

1

23

15

45

m = 12

t = 225 05

A

B

Gap= 15

xy

z

(b)






2which can be







1 (d

B)

500 1000 1500 2000 2500


0

minus5

minus10

minus15

minus20

minus25

minus30

Frequency (MHz)






824

960

1710

2170

180

170

160

150

140

130

120110

100 90 8070

60

50

40

30

20

10

0

minus170

minus160

minus150

minus140

minus130

minus120


minus70minus60

minus50

minus40

minus30

minus20

minus10

00 02

02

05

05

1

2

1

2 5

5

minus02

minus05

minus1

minus2

minus5






500 1000 1500 2000 2500

Frequency (MHz)

ProposedRef1

Ref2minus6 dB

0

minus5

minus10

minus15

minus20

minus25

minus30S1

1 (d

B)


50000e+001

46500e+001

43000e+001

39500e+001

36000e+001

32500e+001

29000e+001

25500e+001

22000e+001

18500e+001

15000e+001

80000e+001

45000e+001

10000e+001

Jsurf (A per m)

(a) (b)


500 1000 1500 2000 2500 3000

Frequency (MHz)


minus30

minus25

minus20

minus15

minus10

minus5

0

S11

(dB

)



(a) (b)


S11

(dB

)

500 1000 1500 2000 2500

0

minus5

minus10

minus15

minus20

minus25

minus30

Frequency (MHz)

L = 39 nHL = 82 nH

L = 15 nHminus6 dB

(a)

Frequency (MHz)

500 1000 1500 2000 2500

S11

(dB

)

0

minus5

minus10

minus15

minus20

minus25

minus30

m = 7 nH


(b)

Frequency (MHz)

500 1000 1500 2000 2500

S11

(dB

)

0

minus5

minus10

minus15

minus20

minus25

minus30

t = 39 nHt = 82 nH

t = 15 nHminus6 dB

(c)




minus10

0

10

10

0

0

90

180

270


xz-plane

(a)


minus10

0

10

10

0

0

90

180

270


yz-plane

(b)


800 820 840 860 880 900 920 940 960 9800

10

20

30

40

50

60

70

80

90

100

GSM850900

EfficiencyGain

Frequency (MHz)

Rad

iati

on E

ffici

ency

(

)

0

1

2

3

4

5

6

An

tenn

a Gain

(dBi)

minus2

minus1

(a)

1700 1800 1900 2000 2100 22000

10

20

30

40

50

60

70

80

90

100


EfficiencyGain

Frequency (MHz)

Rad

iati

on E

ffici

ency

(

)

0

1

2

3

4

5

6

An

tenn

a Gain

(dBi)

minus2

minus1

(b)





4 Conclusion



References
















Research Article









1 Introduction





2 Antenna Design



Lm

dWm

l

1 2

45XY

Z

Via holes

L2L1

(a)

Lgd2

Lgd1

Wgd1

Lgd3

Wgd2

Ls

Ws

Added ground plane

Removed corners

Wgd3

(b)






07 075 08 085 09 095 1minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)


|S ij|(

dB)

(a)

2 22 24 26 28 3minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)


|S ij|(

dB)

(b)









|S11| without slot


2 22 24 26 28 3minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)






1 bend ending

3 bend endings

05 06 07 08 09 1minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)






Ls

Ws

S

45XY

Z

1 2


|S11| simulation

|S21| simulation

|S11|measurement

|S21|measurement

1 15 2 25minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)



07 075 08 08550

60

70

80

90

100

Frequency (GHz)





25 255 26 265 27

Frequency (GHz)

50

60

70

80

90

100




ρe =∣∣∣

int

Ω

(


lowast2ϕpϕ

)

dΩ∣∣∣

2

int

Ω

(


lowast1ϕpϕ

)

dΩint

Ω

(


lowast2ϕpϕ

)

dΩ (1)











05

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

(a)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 0

5

(b)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn5

φ = 90

(c)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 0

5


(d)







0

5

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

(a)

05

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

Port 1Port 2

(b)




A

B

C

(a)

A

B

C

(b)

A

B

C

(c)



MEG =int

Ω

(XPD

XPD + 1GθPθ +

1XPD + 1

GϕPϕ

)

dΩ (2)









around A

G-In 002 010 10minus3 008

G-Out 020 042 710minus4 039

L-In 007 016 10minus5 022

L-Out 026 049 310minus4 051













Rmin Rmax Rmin Rmax


around A

G-In 070 1 078 1

G-Out 035 1 071 1

L-In 063 1 046 1

L-Out 030 1 058 1

G-In 099 1 099 1



L-Out 096 1 098 1

G-In 094 1 080 1



L-Out 070 1 048 1



5 Conclusion




Acknowledgment


References






















Research Article









1 Introduction




Dc1 Dc

tc1

tc2

tc3

Feed plateh

Feed

d

W

L

X

YZ

t

Ground plane

Wg

Top plate

Lb

Lg

Ws

Wf

CL

x2

x3x3

y2y2y3

x1

x2x2

y1



3 4 5 6 7 8 9 10 11 12minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

SimulatedMeasured




3 4 5 6 7 8 9 10 11 12minus20

minus15

minus10

minus5

0

5

10

Frequency (GHz)

Peak

gai

n (

dB)

SimulatedMeasured





3 4 5 6 7 8 9 10 11 12

050

100150200250

Frequency (GHz)Im

peda

nce

Z


200minus

150minus

100minus

50minus


Gain total


Z

Y

Φ

X

θ


3 Results






015

30

45

60

75

90

105

120

135

150

165plusmn180

minus15

minus30

minus45

minus60

minus75

minus90

minus10 5

minus120

minus135

minus150

minus1650

minus10

minus20



(a)

0

15

30

45

607590105

120

135

150

165

plusmn180

minus15

minus30

minus45

minus60


minus120

minus135

minus150

minus165

5minus5

minus15




(b)



002040608

1

Time (ns)

Mag

nit

ude



4 Parametric Study



3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

x1 = 1 mmx1 = 4 mm

x1 = 7 mmx1 = 10 mm





3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

y2 = 5 mmy2 = 6 mmy2 = 7 mm


3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

y3 = 2 mmy3 = 3 mmy3 = 42 mm

y3 = 5 mmy3 = 6 mm



5 Conclusions


References























































1 Introduction









































































Device under test







Mode stirrers


Turntable


Calibration antenna

Walls of reflective

material

Test object (DUT)


Access panel





or a communication

tester















Anechoic chamber

DUT

Absorbers

Measurement antenna

VNA

(a)


DUT

Mode stirrer

Fixed antenna

VNA

(b)



DUT

Absorbers


Channel


(a)


MIMO link

DUT

Mode stirrer

Channel


Fixed antennas

(b)


















anechoic chamber No












Zoom

Zoom times 3


Feeding port

Shorting port

(a)

FM port

GSM port

Filter Switch

to the PIFArsquos

feeding port

to the PIFArsquos

shorting portL1

(b)







0

1

2

3

4

5

6

7

8

9

10

MonopoleHilbertPIFA

7374 72

Sign

al quality












(i) geometry based

(ii) material based





Microstrip line

Clearance area


0

10

20

30

40

50

60

70

80

90

100

1

15

2

25

3

35

4

45

5

55

6

2 21 22 23 24 25 26 27 28 29 3

To

tal

effi

cie

ncy (

)

SW

R

Frequency (GHz)

VSWR

Total efficiency ()











Driven



















Slot

λ4 band 2

(a)

PIFA

λ4 band 1

(b)

PIFA + slot

(c)

08 09 1 11 12 13 14 15 16 17 18 19 2 21 22 23 24 25

Frequency (GHz)

Refl

ecti

on

co

effi

cie

nt

(d

B)

minus14

minus13

minus12

minus11

minus10

minus9

minus8

minus7

minus6

minus5

minus4

minus3

minus2

minus1

0

DB(|S(11)|)

PIFA band 1

DB(|S(22)|)

Slot band 2

DB(|S(33)|)

PIFA + slot

088 GHz

minus789 dB

096 GHz

minus601 dB 217 GHz

minus639 dB

171 GHz

minus643 dB

(d)


Phase delay

Printed circuit

board

Antenna 1 Antenna 2







(a) (b)


35

3

13

23

46

10

2021

48

19

25 25

0

5

10

15

20

25

30

35

40

45

50


Average a

nte

nn

a

effi

cie

ncy (

)

1 monopole

2 monopoles

3 monopoles





WPCB

LPCB

dgap

L





0

10

20

30

40

50

60

70

80

90

100

Effi

cie

ncy (

)

Frequency (GHz)


16

2

16

6

17

1

17

5

17

9

18

3

18

7

19

1

19

5

19

9

20

4

20

8

21

2

21

6

22

22

4

22

8

23

2

23

7

24

1

24

5

24

9

25

3

25

7

26

1

26

5

27












(a) (b)


(a)

(b) (c)












minus14

minus12

minus10

minus8

minus6

minus4

minus2

0

S1

1(d

B)

07

08

09 1

11

12

13

14

15

16

17

18

19 2

21

22

23

24

25

Frequency (GHz)

Dual-band PIFA

Quad-band PIFA

Multiband PIFA

082 GHz

minus6 dB

103 GHz

minus6 dB

176 GHzminus6 dB

184 GHz

minus6 dB

197 GHz

minus6 dB

25 GHz

minus5 dB

212 GHz

minus61 dB






0

5

10

15

20

25

30

35

40

45

50

55

820 890 920 960

Frequency (MHz)

Measu

red

to

tal

effi

cie

ncy (

)

Without strip









Main PCB

Upper PCB

x

y

z

(a)

Trap

x

z

y

(b)

















Transmission line





5 Conclusions






Acknowledgments


References


























































































































1 Introduction







1 +07

1 +06

1 +05

1 +04

1 +03

1 +02

1 +01

1 +00

1 +02

1 +01

1 +00

1 minus 01

1 minus 02

1+

2

1+

3

1+

4

1+

5

1+

6

1+

7

1+

8

1+

9

1+

10

1+

11

Permittivity

Conductivity

Frequency




119875119875abs = 10045601004560119881119881

12120590120590|119864119864|2119889119889119881119881119889 (1)


SAR = 10076531007653 1205901205902120588120588100766910076691198641198642119894119894 119889 (2)





3 Simulations





Brain

CSF

Dura

Bone

Fat

Skin


F 5 Dipole antenna



ickness(mm)


Shell 46 0 5







19525 +002

18313 +002

171e+002

15887 +002

14674 +002

13462 +002

12249 +002

11036 +002

98234 +001

86107 +001

73979 +001

61852 +001

49724 +001

37597 +001

2547 +001

13342 +001

1215 +000

E field (Vm)


20

18

16

14

12

10

8

6

4

2

00 10 20 30 40 50 60

Distance (mm)

Local SAR

Average SAR

SAR (wattkg)




4 Results



E field (Vm)

12126 +002

11368 +002

10611 +002

98528 +001

9095 +001

83372 +001

75795 +001

68217 +001

60639 +001

53062 +001

45484 +001

37906 +001

30329 +001

22751 +001

15173 +001

75958 +000

18184 minus 002


E field (Vm)

13783 +002

12922 +002

12062 +002

11201 +002

10341 +002

94802 +001

86196 +001

7759 +001

68984 +001

60379 +001

51773 +001

43167 +001

34561 +001

25956 +001

1735 +001

8744 +00013818 minus 001




Fat 177times 102

Bone 167times 102

Dura 140times 102

Csf 137times 102

Brain 121times 102


E field (Vm)

14096 +002

13216 +002

12336 +002

11457 +002

10577 +002

96977 +001

88181 +001

79385 +001

70589 +001

61793 +001

52997 +001

44201 +001

35405 +001

26609 +001

17813 +001

90171 +00022117 minus 001


E field (Vm)

1672 +002

15676 +002

14631 +002

13586 +002

12541 +002

11496 +002

10451 +002

94058 +001

83609 +001

73159 +001

6271 +001

5226 +001

41811 +001

31361 +001

20912 +001

10462 +00112991 minus 002









Six layers + dipole












E field (Vm)

17717 +002

16612 +002

15507 +002

14401 +002

13296 +002

12191 +002

11086 +002

99804 +001

88752 +001

777e+001

66647 +001

55595 +001

44542 +001

3349e+001

22438 +001

11385 +00133291 minus 001






E field (Vm)

18019 +002

16896 +002

15774 +002

14651 +002

13529 +002

12407 +002

11284 +002

10162 +002

90396 +001

79172 +001

67948 +001

56724 +001

455 +001

34276 +001

23052 +001

11828 +00160453 minus 001


75 mm

10 mm Antenna patch

62 mm

33 mm

Ground plane

50 mm

92 mm








Six layers + PIFA











E field (Vm)

52151 minus 001

48999 minus 001

45847 minus 001

42695 minus 001

39542 minus 001

3639 minus 001

33238 minus 001

30086 minus 001

26934 minus 001

23782 minus 001

2063 minus 001

17478 minus 001

14326 minus 001

11174 minus 001

80214 minus 002

48693 minus 002

17172 minus 002


E field (Vm)

62004 minus 001

58225 minus 001

54445 minus 001

50665 minus 001

46886 minus 001

43106 minus 001

39327 minus 001

35547 minus 001

31768 minus 001

27988 minus 001

24209 minus 001

20429 minus 001

1665 minus 001

1287 minus 001

90907 minus 002

53112 minus 002

15317 minus 002



E field (Vm)

64315 minus 001

60437 minus 001

56559 minus 001

52682 minus 001

48804 minus 001

44926 minus 001

41048 minus 001

3717 minus 001

33292 minus 001

29415 minus 001

25537 minus 001

21659 minus 001

17781 minus 001

13903 minus 001

10025 minus 001

61475 minus 002

22696 minus 002


E field (Vm)

7482 minus 001

70223 minus 001

65625 minus 001

61027 minus 001

56430 minus 001

51832 minus 001

47234 minus 001

42637 minus 001

38039 minus 001

33441 minus 001

28844 minus 001

24246 minus 001

19648 minus 001

15051 minus 001

10453 minus 001

58554 minus 002

12578 minus 002




5 Conclusions


E field (Vm)

78242 minus 001

73529 minus 001

68816 minus 001

64103 minus 001

5939 minus 001

54678 minus 001

49965 minus 001

45252 minus 001

40539 minus 001

35826 minus 001

31114 minus 001

26401 minus 001

21688 minus 001

16975 minus 001

12262 minus 001

75495 minus 002

28367 minus 002



75133 minus 001

70341 minus 001

65549 minus 001

60757 minus 001

55965 minus 001

51173 minus 001

46381 minus 001

41589 minus 001

36798 minus 001

32006 minus 001

27214 minus 001

22422 minus 001

17630 minus 001

12838 minus 001

80463 minus 002

32544 minus 002





References




















Research Article











2is displaced


1 Introduction




Anetnna part



θ ϕ

z

x

y

100 times 60 mm2

main ground


25

B

A Via to a 50Ω SMA

10



ground

(a)

L = 12 nH

65

7

3

2

1

23

15

45

m = 12

t = 225 05

A

B

Gap= 15

xy

z

(b)






2which can be







1 (d

B)

500 1000 1500 2000 2500


0

minus5

minus10

minus15

minus20

minus25

minus30

Frequency (MHz)






824

960

1710

2170

180

170

160

150

140

130

120110

100 90 8070

60

50

40

30

20

10

0

minus170

minus160

minus150

minus140

minus130

minus120


minus70minus60

minus50

minus40

minus30

minus20

minus10

00 02

02

05

05

1

2

1

2 5

5

minus02

minus05

minus1

minus2

minus5






500 1000 1500 2000 2500

Frequency (MHz)

ProposedRef1

Ref2minus6 dB

0

minus5

minus10

minus15

minus20

minus25

minus30S1

1 (d

B)


50000e+001

46500e+001

43000e+001

39500e+001

36000e+001

32500e+001

29000e+001

25500e+001

22000e+001

18500e+001

15000e+001

80000e+001

45000e+001

10000e+001

Jsurf (A per m)

(a) (b)


500 1000 1500 2000 2500 3000

Frequency (MHz)


minus30

minus25

minus20

minus15

minus10

minus5

0

S11

(dB

)



(a) (b)


S11

(dB

)

500 1000 1500 2000 2500

0

minus5

minus10

minus15

minus20

minus25

minus30

Frequency (MHz)

L = 39 nHL = 82 nH

L = 15 nHminus6 dB

(a)

Frequency (MHz)

500 1000 1500 2000 2500

S11

(dB

)

0

minus5

minus10

minus15

minus20

minus25

minus30

m = 7 nH


(b)

Frequency (MHz)

500 1000 1500 2000 2500

S11

(dB

)

0

minus5

minus10

minus15

minus20

minus25

minus30

t = 39 nHt = 82 nH

t = 15 nHminus6 dB

(c)




minus10

0

10

10

0

0

90

180

270


xz-plane

(a)


minus10

0

10

10

0

0

90

180

270


yz-plane

(b)


800 820 840 860 880 900 920 940 960 9800

10

20

30

40

50

60

70

80

90

100

GSM850900

EfficiencyGain

Frequency (MHz)

Rad

iati

on E

ffici

ency

(

)

0

1

2

3

4

5

6

An

tenn

a Gain

(dBi)

minus2

minus1

(a)

1700 1800 1900 2000 2100 22000

10

20

30

40

50

60

70

80

90

100


EfficiencyGain

Frequency (MHz)

Rad

iati

on E

ffici

ency

(

)

0

1

2

3

4

5

6

An

tenn

a Gain

(dBi)

minus2

minus1

(b)





4 Conclusion



References
















Research Article









1 Introduction





2 Antenna Design



Lm

dWm

l

1 2

45XY

Z

Via holes

L2L1

(a)

Lgd2

Lgd1

Wgd1

Lgd3

Wgd2

Ls

Ws

Added ground plane

Removed corners

Wgd3

(b)






07 075 08 085 09 095 1minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)


|S ij|(

dB)

(a)

2 22 24 26 28 3minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)


|S ij|(

dB)

(b)









|S11| without slot


2 22 24 26 28 3minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)






1 bend ending

3 bend endings

05 06 07 08 09 1minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)






Ls

Ws

S

45XY

Z

1 2


|S11| simulation

|S21| simulation

|S11|measurement

|S21|measurement

1 15 2 25minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)



07 075 08 08550

60

70

80

90

100

Frequency (GHz)





25 255 26 265 27

Frequency (GHz)

50

60

70

80

90

100




ρe =∣∣∣

int

Ω

(


lowast2ϕpϕ

)

dΩ∣∣∣

2

int

Ω

(


lowast1ϕpϕ

)

dΩint

Ω

(


lowast2ϕpϕ

)

dΩ (1)











05

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

(a)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 0

5

(b)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn5

φ = 90

(c)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 0

5


(d)







0

5

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

(a)

05

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

Port 1Port 2

(b)




A

B

C

(a)

A

B

C

(b)

A

B

C

(c)



MEG =int

Ω

(XPD

XPD + 1GθPθ +

1XPD + 1

GϕPϕ

)

dΩ (2)









around A

G-In 002 010 10minus3 008

G-Out 020 042 710minus4 039

L-In 007 016 10minus5 022

L-Out 026 049 310minus4 051













Rmin Rmax Rmin Rmax


around A

G-In 070 1 078 1

G-Out 035 1 071 1

L-In 063 1 046 1

L-Out 030 1 058 1

G-In 099 1 099 1



L-Out 096 1 098 1

G-In 094 1 080 1



L-Out 070 1 048 1



5 Conclusion




Acknowledgment


References






















Research Article









1 Introduction




Dc1 Dc

tc1

tc2

tc3

Feed plateh

Feed

d

W

L

X

YZ

t

Ground plane

Wg

Top plate

Lb

Lg

Ws

Wf

CL

x2

x3x3

y2y2y3

x1

x2x2

y1



3 4 5 6 7 8 9 10 11 12minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

SimulatedMeasured




3 4 5 6 7 8 9 10 11 12minus20

minus15

minus10

minus5

0

5

10

Frequency (GHz)

Peak

gai

n (

dB)

SimulatedMeasured





3 4 5 6 7 8 9 10 11 12

050

100150200250

Frequency (GHz)Im

peda

nce

Z


200minus

150minus

100minus

50minus


Gain total


Z

Y

Φ

X

θ


3 Results






015

30

45

60

75

90

105

120

135

150

165plusmn180

minus15

minus30

minus45

minus60

minus75

minus90

minus10 5

minus120

minus135

minus150

minus1650

minus10

minus20



(a)

0

15

30

45

607590105

120

135

150

165

plusmn180

minus15

minus30

minus45

minus60


minus120

minus135

minus150

minus165

5minus5

minus15




(b)



002040608

1

Time (ns)

Mag

nit

ude



4 Parametric Study



3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

x1 = 1 mmx1 = 4 mm

x1 = 7 mmx1 = 10 mm





3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

y2 = 5 mmy2 = 6 mmy2 = 7 mm


3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

y3 = 2 mmy3 = 3 mmy3 = 42 mm

y3 = 5 mmy3 = 6 mm



5 Conclusions


References




























































































Device under test







Mode stirrers


Turntable


Calibration antenna

Walls of reflective

material

Test object (DUT)


Access panel





or a communication

tester















Anechoic chamber

DUT

Absorbers

Measurement antenna

VNA

(a)


DUT

Mode stirrer

Fixed antenna

VNA

(b)



DUT

Absorbers


Channel


(a)


MIMO link

DUT

Mode stirrer

Channel


Fixed antennas

(b)


















anechoic chamber No












Zoom

Zoom times 3


Feeding port

Shorting port

(a)

FM port

GSM port

Filter Switch

to the PIFArsquos

feeding port

to the PIFArsquos

shorting portL1

(b)







0

1

2

3

4

5

6

7

8

9

10

MonopoleHilbertPIFA

7374 72

Sign

al quality












(i) geometry based

(ii) material based





Microstrip line

Clearance area


0

10

20

30

40

50

60

70

80

90

100

1

15

2

25

3

35

4

45

5

55

6

2 21 22 23 24 25 26 27 28 29 3

To

tal

effi

cie

ncy (

)

SW

R

Frequency (GHz)

VSWR

Total efficiency ()











Driven



















Slot

λ4 band 2

(a)

PIFA

λ4 band 1

(b)

PIFA + slot

(c)

08 09 1 11 12 13 14 15 16 17 18 19 2 21 22 23 24 25

Frequency (GHz)

Refl

ecti

on

co

effi

cie

nt

(d

B)

minus14

minus13

minus12

minus11

minus10

minus9

minus8

minus7

minus6

minus5

minus4

minus3

minus2

minus1

0

DB(|S(11)|)

PIFA band 1

DB(|S(22)|)

Slot band 2

DB(|S(33)|)

PIFA + slot

088 GHz

minus789 dB

096 GHz

minus601 dB 217 GHz

minus639 dB

171 GHz

minus643 dB

(d)


Phase delay

Printed circuit

board

Antenna 1 Antenna 2







(a) (b)


35

3

13

23

46

10

2021

48

19

25 25

0

5

10

15

20

25

30

35

40

45

50


Average a

nte

nn

a

effi

cie

ncy (

)

1 monopole

2 monopoles

3 monopoles





WPCB

LPCB

dgap

L





0

10

20

30

40

50

60

70

80

90

100

Effi

cie

ncy (

)

Frequency (GHz)


16

2

16

6

17

1

17

5

17

9

18

3

18

7

19

1

19

5

19

9

20

4

20

8

21

2

21

6

22

22

4

22

8

23

2

23

7

24

1

24

5

24

9

25

3

25

7

26

1

26

5

27












(a) (b)


(a)

(b) (c)












minus14

minus12

minus10

minus8

minus6

minus4

minus2

0

S1

1(d

B)

07

08

09 1

11

12

13

14

15

16

17

18

19 2

21

22

23

24

25

Frequency (GHz)

Dual-band PIFA

Quad-band PIFA

Multiband PIFA

082 GHz

minus6 dB

103 GHz

minus6 dB

176 GHzminus6 dB

184 GHz

minus6 dB

197 GHz

minus6 dB

25 GHz

minus5 dB

212 GHz

minus61 dB






0

5

10

15

20

25

30

35

40

45

50

55

820 890 920 960

Frequency (MHz)

Measu

red

to

tal

effi

cie

ncy (

)

Without strip









Main PCB

Upper PCB

x

y

z

(a)

Trap

x

z

y

(b)

















Transmission line





5 Conclusions






Acknowledgments


References


























































































































1 Introduction







1 +07

1 +06

1 +05

1 +04

1 +03

1 +02

1 +01

1 +00

1 +02

1 +01

1 +00

1 minus 01

1 minus 02

1+

2

1+

3

1+

4

1+

5

1+

6

1+

7

1+

8

1+

9

1+

10

1+

11

Permittivity

Conductivity

Frequency




119875119875abs = 10045601004560119881119881

12120590120590|119864119864|2119889119889119881119881119889 (1)


SAR = 10076531007653 1205901205902120588120588100766910076691198641198642119894119894 119889 (2)





3 Simulations





Brain

CSF

Dura

Bone

Fat

Skin


F 5 Dipole antenna



ickness(mm)


Shell 46 0 5







19525 +002

18313 +002

171e+002

15887 +002

14674 +002

13462 +002

12249 +002

11036 +002

98234 +001

86107 +001

73979 +001

61852 +001

49724 +001

37597 +001

2547 +001

13342 +001

1215 +000

E field (Vm)


20

18

16

14

12

10

8

6

4

2

00 10 20 30 40 50 60

Distance (mm)

Local SAR

Average SAR

SAR (wattkg)




4 Results



E field (Vm)

12126 +002

11368 +002

10611 +002

98528 +001

9095 +001

83372 +001

75795 +001

68217 +001

60639 +001

53062 +001

45484 +001

37906 +001

30329 +001

22751 +001

15173 +001

75958 +000

18184 minus 002


E field (Vm)

13783 +002

12922 +002

12062 +002

11201 +002

10341 +002

94802 +001

86196 +001

7759 +001

68984 +001

60379 +001

51773 +001

43167 +001

34561 +001

25956 +001

1735 +001

8744 +00013818 minus 001




Fat 177times 102

Bone 167times 102

Dura 140times 102

Csf 137times 102

Brain 121times 102


E field (Vm)

14096 +002

13216 +002

12336 +002

11457 +002

10577 +002

96977 +001

88181 +001

79385 +001

70589 +001

61793 +001

52997 +001

44201 +001

35405 +001

26609 +001

17813 +001

90171 +00022117 minus 001


E field (Vm)

1672 +002

15676 +002

14631 +002

13586 +002

12541 +002

11496 +002

10451 +002

94058 +001

83609 +001

73159 +001

6271 +001

5226 +001

41811 +001

31361 +001

20912 +001

10462 +00112991 minus 002









Six layers + dipole












E field (Vm)

17717 +002

16612 +002

15507 +002

14401 +002

13296 +002

12191 +002

11086 +002

99804 +001

88752 +001

777e+001

66647 +001

55595 +001

44542 +001

3349e+001

22438 +001

11385 +00133291 minus 001






E field (Vm)

18019 +002

16896 +002

15774 +002

14651 +002

13529 +002

12407 +002

11284 +002

10162 +002

90396 +001

79172 +001

67948 +001

56724 +001

455 +001

34276 +001

23052 +001

11828 +00160453 minus 001


75 mm

10 mm Antenna patch

62 mm

33 mm

Ground plane

50 mm

92 mm








Six layers + PIFA











E field (Vm)

52151 minus 001

48999 minus 001

45847 minus 001

42695 minus 001

39542 minus 001

3639 minus 001

33238 minus 001

30086 minus 001

26934 minus 001

23782 minus 001

2063 minus 001

17478 minus 001

14326 minus 001

11174 minus 001

80214 minus 002

48693 minus 002

17172 minus 002


E field (Vm)

62004 minus 001

58225 minus 001

54445 minus 001

50665 minus 001

46886 minus 001

43106 minus 001

39327 minus 001

35547 minus 001

31768 minus 001

27988 minus 001

24209 minus 001

20429 minus 001

1665 minus 001

1287 minus 001

90907 minus 002

53112 minus 002

15317 minus 002



E field (Vm)

64315 minus 001

60437 minus 001

56559 minus 001

52682 minus 001

48804 minus 001

44926 minus 001

41048 minus 001

3717 minus 001

33292 minus 001

29415 minus 001

25537 minus 001

21659 minus 001

17781 minus 001

13903 minus 001

10025 minus 001

61475 minus 002

22696 minus 002


E field (Vm)

7482 minus 001

70223 minus 001

65625 minus 001

61027 minus 001

56430 minus 001

51832 minus 001

47234 minus 001

42637 minus 001

38039 minus 001

33441 minus 001

28844 minus 001

24246 minus 001

19648 minus 001

15051 minus 001

10453 minus 001

58554 minus 002

12578 minus 002




5 Conclusions


E field (Vm)

78242 minus 001

73529 minus 001

68816 minus 001

64103 minus 001

5939 minus 001

54678 minus 001

49965 minus 001

45252 minus 001

40539 minus 001

35826 minus 001

31114 minus 001

26401 minus 001

21688 minus 001

16975 minus 001

12262 minus 001

75495 minus 002

28367 minus 002



75133 minus 001

70341 minus 001

65549 minus 001

60757 minus 001

55965 minus 001

51173 minus 001

46381 minus 001

41589 minus 001

36798 minus 001

32006 minus 001

27214 minus 001

22422 minus 001

17630 minus 001

12838 minus 001

80463 minus 002

32544 minus 002





References




















Research Article











2is displaced


1 Introduction




Anetnna part



θ ϕ

z

x

y

100 times 60 mm2

main ground


25

B

A Via to a 50Ω SMA

10



ground

(a)

L = 12 nH

65

7

3

2

1

23

15

45

m = 12

t = 225 05

A

B

Gap= 15

xy

z

(b)






2which can be







1 (d

B)

500 1000 1500 2000 2500


0

minus5

minus10

minus15

minus20

minus25

minus30

Frequency (MHz)






824

960

1710

2170

180

170

160

150

140

130

120110

100 90 8070

60

50

40

30

20

10

0

minus170

minus160

minus150

minus140

minus130

minus120


minus70minus60

minus50

minus40

minus30

minus20

minus10

00 02

02

05

05

1

2

1

2 5

5

minus02

minus05

minus1

minus2

minus5






500 1000 1500 2000 2500

Frequency (MHz)

ProposedRef1

Ref2minus6 dB

0

minus5

minus10

minus15

minus20

minus25

minus30S1

1 (d

B)


50000e+001

46500e+001

43000e+001

39500e+001

36000e+001

32500e+001

29000e+001

25500e+001

22000e+001

18500e+001

15000e+001

80000e+001

45000e+001

10000e+001

Jsurf (A per m)

(a) (b)


500 1000 1500 2000 2500 3000

Frequency (MHz)


minus30

minus25

minus20

minus15

minus10

minus5

0

S11

(dB

)



(a) (b)


S11

(dB

)

500 1000 1500 2000 2500

0

minus5

minus10

minus15

minus20

minus25

minus30

Frequency (MHz)

L = 39 nHL = 82 nH

L = 15 nHminus6 dB

(a)

Frequency (MHz)

500 1000 1500 2000 2500

S11

(dB

)

0

minus5

minus10

minus15

minus20

minus25

minus30

m = 7 nH


(b)

Frequency (MHz)

500 1000 1500 2000 2500

S11

(dB

)

0

minus5

minus10

minus15

minus20

minus25

minus30

t = 39 nHt = 82 nH

t = 15 nHminus6 dB

(c)




minus10

0

10

10

0

0

90

180

270


xz-plane

(a)


minus10

0

10

10

0

0

90

180

270


yz-plane

(b)


800 820 840 860 880 900 920 940 960 9800

10

20

30

40

50

60

70

80

90

100

GSM850900

EfficiencyGain

Frequency (MHz)

Rad

iati

on E

ffici

ency

(

)

0

1

2

3

4

5

6

An

tenn

a Gain

(dBi)

minus2

minus1

(a)

1700 1800 1900 2000 2100 22000

10

20

30

40

50

60

70

80

90

100


EfficiencyGain

Frequency (MHz)

Rad

iati

on E

ffici

ency

(

)

0

1

2

3

4

5

6

An

tenn

a Gain

(dBi)

minus2

minus1

(b)





4 Conclusion



References
















Research Article









1 Introduction





2 Antenna Design



Lm

dWm

l

1 2

45XY

Z

Via holes

L2L1

(a)

Lgd2

Lgd1

Wgd1

Lgd3

Wgd2

Ls

Ws

Added ground plane

Removed corners

Wgd3

(b)






07 075 08 085 09 095 1minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)


|S ij|(

dB)

(a)

2 22 24 26 28 3minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)


|S ij|(

dB)

(b)









|S11| without slot


2 22 24 26 28 3minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)






1 bend ending

3 bend endings

05 06 07 08 09 1minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)






Ls

Ws

S

45XY

Z

1 2


|S11| simulation

|S21| simulation

|S11|measurement

|S21|measurement

1 15 2 25minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

|S ij|(

dB)



07 075 08 08550

60

70

80

90

100

Frequency (GHz)





25 255 26 265 27

Frequency (GHz)

50

60

70

80

90

100




ρe =∣∣∣

int

Ω

(


lowast2ϕpϕ

)

dΩ∣∣∣

2

int

Ω

(


lowast1ϕpϕ

)

dΩint

Ω

(


lowast2ϕpϕ

)

dΩ (1)











05

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

(a)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 0

5

(b)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn5

φ = 90

(c)

0

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 0

5


(d)







0

5

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

(a)

05

15

30

45

60

7590105120

135

150

165

180

minus165

minus150

minus135

minus120

minus105 minus75

minus60

minus90

minus45

minus30

minus15


plusmn

φ = 90

Port 1Port 2

(b)




A

B

C

(a)

A

B

C

(b)

A

B

C

(c)



MEG =int

Ω

(XPD

XPD + 1GθPθ +

1XPD + 1

GϕPϕ

)

dΩ (2)









around A

G-In 002 010 10minus3 008

G-Out 020 042 710minus4 039

L-In 007 016 10minus5 022

L-Out 026 049 310minus4 051













Rmin Rmax Rmin Rmax


around A

G-In 070 1 078 1

G-Out 035 1 071 1

L-In 063 1 046 1

L-Out 030 1 058 1

G-In 099 1 099 1



L-Out 096 1 098 1

G-In 094 1 080 1



L-Out 070 1 048 1



5 Conclusion




Acknowledgment


References






















Research Article









1 Introduction




Dc1 Dc

tc1

tc2

tc3

Feed plateh

Feed

d

W

L

X

YZ

t

Ground plane

Wg

Top plate

Lb

Lg

Ws

Wf

CL

x2

x3x3

y2y2y3

x1

x2x2

y1



3 4 5 6 7 8 9 10 11 12minus30

minus25

minus20

minus15

minus10

minus5

0

Frequency (GHz)

SimulatedMeasured




3 4 5 6 7 8 9 10 11 12minus20

minus15

minus10

minus5

0

5

10

Frequency (GHz)

Peak

gai

n (

dB)

SimulatedMeasured





3 4 5 6 7 8 9 10 11 12

050

100150200250

Frequency (GHz)Im

peda

nce

Z


200minus

150minus

100minus

50minus


Gain total


Z

Y

Φ

X

θ


3 Results






015

30

45

60

75

90

105

120

135

150

165plusmn180

minus15

minus30

minus45

minus60

minus75

minus90

minus10 5

minus120

minus135

minus150

minus1650

minus10

minus20



(a)

0

15

30

45

607590105

120

135

150

165

plusmn180

minus15

minus30

minus45

minus60


minus120

minus135

minus150

minus165

5minus5

minus15




(b)



002040608

1

Time (ns)

Mag

nit

ude



4 Parametric Study



3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

x1 = 1 mmx1 = 4 mm

x1 = 7 mmx1 = 10 mm





3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

y2 = 5 mmy2 = 6 mmy2 = 7 mm


3 4 5 6 7 8 9 10 11 12minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus50

Frequency (GHz)

S11

(dB

)

y3 = 2 mmy3 = 3 mmy3 = 42 mm

y3 = 5 mmy3 = 6 mm



5 Conclusions


References

























Advances in Antenna Technology for Wireless Handheld Devices

Documents

Transcript of Advances in Antenna Technology for Wireless Handheld Devices