22090174010000phystatusandremainingtopicsinsept09.docx

22090174010000phystatusandremainingtopicsinsept09.docx

《22090174010000phystatusandremainingtopicsinsept09.docx》由会员分享，可在线阅读，更多相关《22090174010000phystatusandremainingtopicsinsept09.docx（24页珍藏版）》请在冰豆网上搜索。

22090174010000phystatusandremainingtopicsinsept09

IEEEP802.22

WirelessRANs

PHYStatusandRemainingIssuesasinSept2009

Date:

2009-9-18

Author（s）:

Name

Company

Address

Phone

email

ZanderZhongdingLei

InstituteforInfocommResearch,Singapore

1FusionopolisWay#21-01Connexis,Singapore138632

65-6408-2436

leizd@i2r.a-star.edu.sg

Abstract

ThisdocumentsummarizesthecurrentPHYstatusinthecommentsandresolution.TheremainingPHYrelatedsystemandPHYspecificissuesarelisted,togetherwithdetailedupdates.

Notice:

ThisdocumenthasbeenpreparedtoassistIEEE802.22.Itisofferedasabasisfordiscussionandisnotbindingonthecontributingindividual（s）ororganization（s）.Thematerialinthisdocumentissubjecttochangeinformandcontentafterfurtherstudy.Thecontributor（s）reserve（s）therighttoadd,amendorwithdrawmaterialcontainedherein.

Release:

Thecontributorgrantsafree,irrevocablelicensetotheIEEEtoincorporatematerialcontainedinthiscontribution,andanymodificationsthereof,inthecreationofanIEEEStandardspublication;tocopyrightintheIEEE’snameanyIEEEStandardspublicationeventhoughitmayincludeportionsofthiscontribution;andattheIEEE’ssolediscretiontopermitotherstoreproduceinwholeorinparttheresultingIEEEStandardspublication.ThecontributoralsoacknowledgesandacceptsthatthiscontributionmaybemadepublicbyIEEE802.22.

PatentPolicyandProcedures:

ThecontributorisfamiliarwiththeIEEE802PatentPolicyandProcedures

//standards.ieee.org/guides/bylaws/sb-bylaws.pdf>,includingthestatement"IEEEstandardsmayincludetheknownuseofpatent（s）,includingpatentapplications,providedtheIEEEreceivesassurancefromthepatentholderorapplicantwithrespecttopatentsessentialforcompliancewithbothmandatoryandoptionalportionsofthestandard."EarlydisclosuretotheWorkingGroupofpatentinformationthatmightberelevanttothestandardisessentialtoreducethepossibilityfordelaysinthedevelopmentprocessandincreasethelikelihoodthatthedraftpublicationwillbeapprovedforpublication.PleasenotifytheChairasearlyaspossible,inwrittenorelectronicform,ifpatentedtechnology（ortechnologyunderpatentapplication）mightbeincorporatedintoadraftstandardbeingdevelopedwithintheIEEE802.22WorkingGroup.Ifyouhavequestions,contacttheIEEEPatentCommitteeAdministratorat.

1.PHYCommentsSummary

PHYhasreceivedtotal104TR/Tcomments,comprising93insection8and11insection6appointedbyMACteam.UptothebeginningoftheSept.meeting,wehavereviewedoneroundall104comments.Thestatusissummarizedasfollows:

Unresolved

Resolved

-Pending（25）

-Defer

（1）

-Accepted（26）

-Rejected（11）（includes872/931）

-Withdrawn

（2）（830wrongcolumn）

-Counter（33）

-Superseded（6）

Subtotal:

26or25%

Subtotal78or~75%

759,760,771,814,817,819,841,846,857,858,859,861,878,882,883,916,933,936,952,968,969,970,

322,330,338,343

∙872,931->”Defer”to“Reject”or“Counter”（FEC）

∙877->statustoremove“?

”

2.DetailedUpdatesonTR/TComments

▪WorkingGroupSystemIssues

1.General

-Apurva（759/760）

ReplacingPHYwithastrippeddownversionof802.16-2009OFDMA（NoMIMO,NoAAS,NoHARQ）PHY

Jul09:

ithasbeendemonstratedthat802.16PHYdoesnotmeettheFRDfunctionrequirements.Ittook3yearsforthePHYteamtodevelopthecurrentPHYspecstosatisfytheFRDrequirementwhicharedifferentfromthatin802.16.Ifthecommentercandemonstratethat802.16PHYcanmeettheFRD,furtherconsiderationmaytakeplace.Commentwillberejectedasnon-actionableifnodemonstrationiscompletedbyNov.2009.

NoupdateinSept09（Pending）

2.SamplingRates

-Robert（ID771）

Subcarrierspacingis3.35,3.91and4.46KHzfor6,7and8MHzrespectively.Suggesttouse7MHzprofiletounifythesamplingratesfor6MHzand8MHzwhilerelaxthefrontenddesign.

Sept09CC:

Robert:

proposeaunificationofsamplingratesfor6MHz,7MHzand8MHz.Theproposedsystemwillhave:

-EveryMACframeof10msisamultipleofPHYclockcycle

-Samesamplingfrequency,Inter-carrierspacing,FFTperiod,timeunit,symboldurationforall3bands

Itcansimplifytheradiofrontenddesignandrelaxthephasenoise/frequencyoffsetlimitationfor6MHzradio;andsimplifythehardwareandsoftwareimplementationforMACandPHY.

Anumberofconcernswereraisedduringconferencecall:

Ivan:

DoesthenewproposalleadtomoredynamicrangeofADC.

Zander:

Whatandhowimportantisthebenefitofhavingfullcyclesofsamplingrateinevery10msshouldbeelaborated.IncurrentPHY,wehavethesameelementarysamplingclockfrequency8/7MHzforallthree3TVchannels.Thesamplingfrequency,inter-carrierspacing,FFTperiodandTimeUnitscalesnicelywithafactorofBW=6,7,8MHzrespectively.Thisgivesusthesamenumberofguardsub-carriers,usedsub-carriers,datasub-carriers,pilotsub-carriersandwecanusetheexactlysamesubcarrierallocationscheme,interleaverandFECcodecsetc.

Thenewproposalhasdifferentblocksizeandmayhaveimpacttointerleaver,FECcodecsandsubcarrierallocationscheme.

Thenewproposalhasdifferentnumberofsubcarriersfordifferentbandwidths.144guardsubcarriersareusedin8MHzchannelswhereas648subcarriersareusedin6MHzchannels.It’sabigwasteofspectrum.

SungHyun:

Thenewproposalwillleadtodifferentpreamblesequencesfordifferentchannels.

Gerald:

ThepayloadofSCH/FCHmayhavetoberevisitedandimpactneedtobeaddressed.

Ivan:

ItseemsdramaticchangetothePHYisexpected.Itmustbringtothegroupwithheavyadvantagestojustifytheriskofdelayingthestandard.

Sept09（Pending）:

-WiLANandETRItolookintothecriticalcarrierspacingre.LOphasenoiseandtransitionbandsattheedgesofthechannelforfilteringpurposere.55dBrrejectionaspertheRFmaskintheFCCR&O08-260.

-Geraldprepared06-264r9toexploreparameters.

-Agreementwasreachedonhavinganintegernumberofsamplingperiodsper10msframe（thatisintegernumberof10kHz,e.g.,6.78MHz）butitisnotclearthatasinglesamplingfrequencyforthethreebandwidthsisabetterapproach.

-Action:

FurtherworkwillbedonebyRoberttoexplorealternatives.

3.NetworkSynchronization

-Robert（941）

BSsaresynchronizedinWRAN.Thesynctoleranceshouldbeinusecratherthan+/-25%oftheshortestprefix（relevanttosamplingrateissue）

-SteveK（942）

The+/-25%oftheshortestCP（1/32）gives+/-2.333usecfor6MHzchannels,whichdecreasestheamountofchanneldelayspreadthatcanbetoleratedby4.666usectotal.PointoutthattheeffectiveCPfortheuplinkisshorterbye.g.4.666usecduetoimperfectsynchronization.Iflongfiltersareusedtosatisfyspectralmasks,evenmoreofthemargincouldbeconsumed.Theutilityofshort（1/32）CPsisconsequentlyreduced.

Aug09CC:

Robert:

Weneedaspecificnumberinusecratherthan+/-25%oftheshortestprefix.

Gerald:

This+/-25%correspondsto+/-16TUorsamplingclocks.Itworksfor6,7,or8MHzchannelsandthetoleranceisdifferentinusecfordifferentchannels（differentchannelsareusuallyusedindifferentgeographicarea/countries）.Thisiswhyitisexpressedthiswayratherthaninabsoluteusecsincetherangingcontrolloopwillworkintermsofsamplingclockwhatevertheactualsamplingfrequencyisinthegivenbandwidth.

Steve:

+/-25%eatsupintothecyclicprefixrangeandreducesthemultipathcorrectioncapability.ThisrepresentsaneffectivereductionoftheCPof4.66usecforthe6MHxchannelbandwidth.Slightlysmallerreductionappliestothe7and8MHzchannelbandwidths.Wecouldreduceitto+/-8TUoreven+/-4TUiffurtheranalysiscanshowthatwecantimesynchronizetotheBSframewithe.g.95%confidencetoatoleranceof4or8timeunits（insteadofthepresent+/-16TU）atsomeminimumSNRandchannelmodelaveragedoverallrangingcodes.IntheenditcomesdowntowhetherthedelayspreadtoleranceoftheCPisborderlineforagivenenvironmentandthis25%syncrequirestogotothenextlargerCP,resultinginanefficiencyhit.802.16-2009（clause8.4.10.1.2）uses+/-25%anditwaslikelyjustborrowedfromthere.Weretherecontributionsprovidedin802.16thatjustifiedthisvalue?

Maybewecansavesomeworkhere.

Gerald:

Asthereisanotherrelevantcommentonsamplingclocksyettobeaddressed,thecommentissuggestedtobeaddressedtogether/after."

Sept09:

-Robert:

1usec.WiMAXprofileuses+/-1usecasagreedbytheindustry.+/-8TUwouldcorrespondto1.333usecfor6MHz,1.143for7MHzBWand1usecfor8MHzBW.

-Conclusion:

+/-1usectoleranceshouldbeusedwhichcorrespondstoabout+/-8TUor+/-12.5%CP.

-（941-Counter）Thespecificationshouldbe:

"+/-8TU（equivalenttoterms1.333usecfor6MHz,1.143for7MHzBWand1usecfor8MHzBW."

-（942-Counter）Addthefollowingtext:

"indicatethatanyfilterattheoutputoftheOFDMmodulatortohelpmeetingtherejectionrequiredbytheRFmaskwillcreatetemporaldispersionthatwillconsumepartofthecyclicprefixcapabilitytoalleviatechanneltimespreading."

4.Symbolclockfrequencytolerance

-SteveK.（952）

Thesymbolclockfrequencyisthe3348Hzsubcarrierspacingdividedby1+CP（asfarasIunderstandit;symbolclockfrequencyisnotdefinedinthedraft）.ForCP=1/8,thesymbolfrequencyis2976Hz（1/336usec）.2%ofthisis6.7usec.Overa29symbolframe,thetimingcouldslipby29x6.7usec=197usec,overhalfasymbol,clearlyintolera