版權(quán)說明:本文檔由用戶提供并上傳,收益歸屬內(nèi)容提供方,若內(nèi)容存在侵權(quán),請進(jìn)行舉報或認(rèn)領(lǐng)
文檔簡介
TDLTE優(yōu)化案例匯編NPOSystemStreamMainMenu異頻測量PDCCH功控及符號自適應(yīng)TM自適應(yīng)門限優(yōu)化室分高掉線優(yōu)化上行功控優(yōu)化通過權(quán)值修改DL發(fā)射功率PCIMOD3干擾異廠家切換2013.10的ATU測試,反映NSN福州UE發(fā)射功率較高優(yōu)化背景閉環(huán)功控參數(shù)調(diào)整–原理及過程閉環(huán)功控參數(shù)調(diào)整–原理及過程閉環(huán)功控調(diào)整主要依靠2類(SINR,RSSI),各2個(上、下)門限值的設(shè)置,通過當(dāng)前eNB接收到UE的SINR/RSSI對比門限值,在PDSCH/PDCCH里下發(fā)調(diào)整指令,經(jīng)過4個(or6個)子幀后調(diào)整UE發(fā)射功率。城市邊緣RSRP(dBm)覆蓋率RSRP連續(xù)弱覆蓋里程占比脫網(wǎng)里程(km)連續(xù)UE高發(fā)射功率里程占比福州-10993.0%12.7%959.3%UE-PUSCH發(fā)射功率分析備注:UE高發(fā)射功率路段:UE的PUSCH信道發(fā)射功率持續(xù)10秒70%采樣點大于16dBm的連續(xù)路段。UE發(fā)射功率測試結(jié)果RSRP<-85dbm時UE-PUSCH發(fā)射功率會大于16dbm該指標(biāo)為集團(tuán)ATU測試考核項UE發(fā)射功率測試結(jié)果UE-PUSCH發(fā)射功率分析UE-PUSCH發(fā)射功率在不同天線模式下表現(xiàn)相差不大SINR較好時上行發(fā)射功率仍然很高,閉環(huán)功控作用沒有體現(xiàn)UE發(fā)射功率理論分析UE-PUSCH發(fā)射功率分析開環(huán)計算:10log(Mpusch(i))=10log(100)=20,p0NomPusch=-100,α=1,PL=epre-RSRP=12.2-(-85)=97.2Ppusch≈17dbm(RSRP=-85dbm點)UE發(fā)射功率理論分析UE-PUSCH發(fā)射功率分析現(xiàn)網(wǎng)配置:ulpcUpqualSch=20,ulpcLowqualSch=18,ulpcUplevSch=-96,ulpcLowlevSch=-103,由上配置來看,
ulpcUpqualSch=20,ulpcLowqualSch=18,兩個參數(shù)配置偏高會導(dǎo)致UE上行發(fā)射功率上升UE發(fā)射功率驗證UE-PUSCH發(fā)射功率分析參數(shù)配置組alphap0ulpcupqualschulpclowqualsch配置11-1002018配置21-10031配置30.8-1002018配置40.8-752018配置50.8-7531上行功控驗證測試配置UE發(fā)射功率驗證UE-PUSCH發(fā)射功率分析上行功控驗證測試配置參數(shù)配置組下行吞吐量(kbps)RSRPSINRCQIMCSUETxPower_pusch備注配置131677-101169.51920均值(拉遠(yuǎn)測試)配置228447-104158.61716配置331372-103169.51821配置428193-1051491721配置530268-1031591816結(jié)論UE-PUSCH發(fā)射功率分析P0和alpha對整體功率降低作用不明顯閉環(huán)功控參數(shù)可以較好的控制整體功率閉環(huán)功控參數(shù)和開環(huán)功控參數(shù)同時調(diào)整可較好控制上行功率,且對吞吐量影響較小參數(shù)配置組alphap0ulpcupqualschulpclowqualsch配置50.8-7531推薦參數(shù)配置,可在配置5基礎(chǔ)上可適當(dāng)上調(diào)閉環(huán)功控參數(shù)閉環(huán)功控參數(shù)調(diào)整–測試方案現(xiàn)網(wǎng)此4門限的設(shè)置,采用的是goldenconfig里的設(shè)置,其中:SINR的上、下門限分別設(shè)置為20,18。RSSI的上、下門限分別設(shè)置為-96,-103測試輪次p0NomPuschalphaulpcUpqualSchulpcLowqualSchulpcUplevSchulpcLowlevSch1(初始設(shè)置)-10012018-96-1032-950.8不變不變不變不變
3不變不變31不變不變4不變不變97不變不變5不變不變1412不變不變
6不變不變不變不變-100-1037不變不變不變不變-98-103
8不變不變97不變不變9不變不變31不變不變
分析結(jié)果,總結(jié),評估測試結(jié)果測試區(qū)域選定簇15,終端和業(yè)務(wù)模擬ATU,路線采用更細(xì)的簇優(yōu)化路線。閉環(huán)功控參數(shù)調(diào)整–測試結(jié)果下行業(yè)務(wù)終端的統(tǒng)計數(shù)據(jù),可以看到:調(diào)整SINR門限后,連續(xù)UE高發(fā)射功率里程占比明顯下降。閉環(huán)功控參數(shù)調(diào)整–測試結(jié)果分析下行業(yè)務(wù)終端的統(tǒng)計數(shù)據(jù),可以看到:調(diào)整SINR門限后,連續(xù)UE高發(fā)射功率里程占比明顯下降。閉環(huán)功控參數(shù)調(diào)整–測試結(jié)果上行業(yè)務(wù)終端的統(tǒng)計數(shù)據(jù),可以看到:調(diào)整SINR門限后,連續(xù)UE高發(fā)射功率里程占比明顯下降。閉環(huán)功控參數(shù)調(diào)整–測試結(jié)果上行業(yè)務(wù)終端的統(tǒng)計數(shù)據(jù),可以看到:調(diào)整SINR門限后,連續(xù)UE高發(fā)射功率里程占比明顯下降。閉環(huán)功控參數(shù)調(diào)整–結(jié)果分析從測試結(jié)果看到:調(diào)整p0和alpha,對UE發(fā)射功率稍有影響,變化不大。調(diào)整SINR門限,對UE發(fā)射功率影響很大,UE連續(xù)高發(fā)射功率占比下降了一半。調(diào)整SINR的不同門限,從(3,1)、(9,7)、(14,12),對UE發(fā)射功率的影響不大。調(diào)整RSSI的發(fā)射功率,對UE的發(fā)射功率有影響。從測試方案3開始,到測試方案9,測試結(jié)果相差不大,那么采用哪種參數(shù)方案較合適呢?下面結(jié)合UE上報的SINR和RSSI說明。閉環(huán)功控參數(shù)調(diào)整–結(jié)果分析閉環(huán)功控參數(shù)調(diào)整–結(jié)果分析閉環(huán)功控參數(shù)調(diào)整–結(jié)果分析從UE上報的SINR和RSSI看:SINR門限調(diào)整前,門限設(shè)置是(18,20),UE上報的SINR區(qū)間主要是(18,22)。調(diào)整門限后,UE上報的SINR區(qū)間主要是(x,20),X指不同設(shè)置導(dǎo)致不同的主區(qū)間。同理,RSSI也相應(yīng)下降,說明調(diào)整門限確實導(dǎo)致了UE發(fā)射功率的下降。閉環(huán)功控參數(shù)調(diào)整–結(jié)果分析測試步驟6,7、8、9的參數(shù)設(shè)置組合,UE上報的SINR/RSSI集中度差。分析原因,是由于縮小了RSSI的門限區(qū)間,反而導(dǎo)致UE的發(fā)射功率頻繁調(diào)整,反而會導(dǎo)致功控精度變差,這無疑是我們不想看到的。閉環(huán)功控參數(shù)調(diào)整–結(jié)果分析測試步驟3、4、5的參數(shù)設(shè)置組合,組合3、4相較組合5,集中度較好。閉環(huán)功控參數(shù)調(diào)整–結(jié)果分析從初步的測試結(jié)果看,選取參數(shù)組合3或4會大大降低連續(xù)UE高功率發(fā)射占比,同時UE上報的SINR、RSSI較為穩(wěn)定。功控精度好。但是,參數(shù)組合3、4的測試過程中,下行業(yè)務(wù)終端DT指標(biāo)表現(xiàn)良好,上行業(yè)務(wù)終端DT卻都出現(xiàn)了掉線。建議針對3、4,重新測試,驗證掉線情況。閉環(huán)功控參數(shù)調(diào)整–進(jìn)一步的測試為了驗證掉線的情況,選取了有恒定業(yè)務(wù)量的,載有LTE-fi的301公交車沿線的站點,按照下面的參數(shù)修改,驗證掉線次數(shù)。結(jié)果發(fā)現(xiàn)SINR門限調(diào)整后無線掉線的增加非常明顯。
測試輪次p0NomPuschalphaulpcUpqualSchulpcLowqualSch301掉線個數(shù)當(dāng)前設(shè)置-100120182981-1000.820182852-85120181843-850.820184114-85197650+5-8511412800+6-8511715732閉環(huán)功控參數(shù)調(diào)整–進(jìn)一步的測試301公交線的無線掉線率提升的非常明顯,顯然不能采取這套參數(shù)投入日常使用。那么高掉線率的原因是什么呢?我們注意到了2點的區(qū)別:1.簇15的多次測試都是模擬ATU的測試:2個UE,一個只做上行,滿200M后就detach,再attach,重復(fù),一個只做下行,滿500M后就detach,循環(huán)往復(fù)。而301由于車載LTE-fi,造成全天都存在業(yè)務(wù)量,不存在detach/attach的情況。2.簇15的多次測試都是2個UE,一個單做上行,一個單做下行。301公交線由于堵車、雙向道路的情況,存在一個eNB下是多UE的情況,所以懷疑eNB對多UE的功控調(diào)度算法存在問題,造成單UE時指標(biāo)良好,UE發(fā)射功率低,多UE時雖然發(fā)射功率低,但掉線明顯。
閉環(huán)功控參數(shù)調(diào)整–進(jìn)一步的測試為驗證此論點,選取簇15,模擬301的連續(xù)業(yè)務(wù),做了單/多UE對比測試。測試方法是4個UE,同時做上傳/下載的長呼,測試結(jié)果驗證了我們的推斷,無線掉線次數(shù)較多,而且不同SINR門限的設(shè)置下,掉線次數(shù)均較多。作為對比,在簇15,用單個UE,其他測試條件不變的情況下,測試驗證,結(jié)果發(fā)現(xiàn),掉線次數(shù)果然大大減少。測試輪次p0NomPuschalphaulpcUpqualSchulpcLowqualSchulpcUplevSchulpcLowlevSch掉線次數(shù)1-85197-96-103472-8512018不變不變33測試輪次p0NomPuschalphaulpcUpqualSchulpcLowqualSchulpcUplevSchulpcLowlevSch掉線次數(shù)1-85197-96-1031閉環(huán)功控參數(shù)調(diào)整–結(jié)論從測試結(jié)果可以得出結(jié)論:降低UE高功率連續(xù)發(fā)射比,和提升無線掉線率,是一個tradeoff,不能兼顧。但是由于ATU測試時的業(yè)務(wù)模型和我們平時的業(yè)務(wù)模型(301)有很大的不同,所以還是有操作空間。閉環(huán)功控對降低UE高功率連續(xù)發(fā)射比較明顯,在類似ATU的宏站+道路測試的環(huán)境下,SINR門限修改的影響較大。低于正常上報SINR后,幾種SINR的門限組合[14,12],[9,7],[3,1]對高功率連續(xù)發(fā)射比的影響不大。
多UE的調(diào)度或者功率分配上可能存在問題,會造成高掉線,現(xiàn)場沒有做進(jìn)一步的分析??梢圆捎?套功控參數(shù):ATU功控參數(shù)用于保持路測指標(biāo)的同時,提高UE高功率連續(xù)發(fā)射比。網(wǎng)絡(luò)運行的功控參數(shù)用于平時使用,以穩(wěn)定無線掉線率指標(biāo)。MainMenu異頻測量PDCCH功控及符號自適應(yīng)TM自適應(yīng)門限優(yōu)化室分高掉線優(yōu)化上行功控優(yōu)化通過權(quán)值修改DL發(fā)射功率PCIMOD3干擾異廠家切換異頻測量GAP異頻測量的方式取決于終端的實現(xiàn),1)如果UE接受機帶寬能夠同時覆蓋服務(wù)主服務(wù)小區(qū)和待測小區(qū)的頻點(如兩個連續(xù)20M的D頻點),那么就不需要測量間隔GAP(Nongap-assisted類型測量)的輔助而實現(xiàn)異頻測量。
2)但是目前UE的接收機帶寬都是20M的,不足以同時覆蓋服務(wù)小區(qū)頻點與待測小區(qū)所在頻點,因此UE需要測量間隔GAP的輔助(gap-assisted類型測量)才能進(jìn)行異頻測量
協(xié)議考慮是盡量減小終端的處理要求,以簡約化。因此目前UE在20M接收帶寬的情況下,實現(xiàn)的異頻測量,都需要Gap。在Gap測量周期內(nèi),停止所有業(yè)務(wù)和服務(wù)小區(qū)的測量等等,專門用于異頻鄰區(qū)的測量。
3GPP36.508定義了measGAP的2種配置,GAP模式分為40ms周期和80ms周期兩種,GAP測量長度均為6ms。為了驗證異頻測量對終端性能的影響,現(xiàn)場進(jìn)行了對比測試,結(jié)果顯示啟動異頻測量時(40msGAP周期)相比不測量時上下行平均吞吐量均下降25%左右。GAP周期DL平均UL平均不測量55mbps8mbps40ms41mbps6mbpsmappedtotheexistingfeatureLTE1679.80msgapsareonlyplanedfornontimecriticalhandoversasthe80msgapoptionmaydelaytheeventreporting.123453GPP:36.213按照36.313和36.213協(xié)議,1:7(子幀2:2,特殊子幀10:2:2)配比在一個測量周期內(nèi)紅色和藍(lán)色子幀不能被用于下行調(diào)度,共5種場景,異頻測量造成下行調(diào)度率平均損失30%,上行調(diào)度受到的影響類似。異頻測量影響上下行調(diào)度次數(shù),進(jìn)而影響吞吐率性能場景DL調(diào)度次數(shù)影響比例142530%240033%340033%445025%545025%由于LTE為寬頻系統(tǒng),為了減小終端開銷,需要啟動專門的周期性異頻測量過程A2事件觸發(fā)異頻測量和GAP(6ms)建立
A1事件觸發(fā)停止異頻測量
異頻測量對性能的影響ACK/NACK異頻測量對性能的影響---實測數(shù)據(jù)分析測試區(qū)域CRS平均發(fā)射功率(dBm)CDF5%(下行50%加擾)中值(下行50%加擾)RSRP(dBm)SINR(dB)DLTHR(Mbps)ULTHR(Mbps)RSRP(dBm)SINR(dB)DLTHR(Mbps)ULTHR(Mbps)2*20MHz創(chuàng)新頻率方案3,CRS提升0~3dB*,異頻起測門限根據(jù)優(yōu)化需要設(shè)置,主要分布于-86dBm~-94dBm間17.9-51.32-82.520.337.1617.462*20MHz創(chuàng)新頻率方案4,CRS提升0~3dB*,異頻起測門限統(tǒng)一設(shè)置為-80dBm17.9-102.544.988.547.32-82.6120.223.6613.37方案4:異頻測量門限在-80dBm,從紅線上看,70%的區(qū)域低于這個起測電平,導(dǎo)致70%區(qū)域處于異頻測量中;從藍(lán)線上看,90%的切換發(fā)生時的切換RSRP都是低于這個異頻測量門限的,存在異頻起測過早的情況;方案3:異頻起測門限根據(jù)優(yōu)化需要設(shè)置,主要分布于-86dBm~-94dBm間。從紅線上看,平均約40%區(qū)域低于這個測量門限,既約40%區(qū)域處于異頻測量中;理論預(yù)期:方案4由于70%區(qū)域處于異頻測量,中值統(tǒng)計(CDF50%)落于異頻測量采樣點;方案3由于40%區(qū)域處于異頻測量,中值統(tǒng)計(CDF50%)落于非異頻測量采樣點;因此兩者理論差異約30%實測:方案4相對方案3,中值下行速率下降36.3%(37.16M->23.66M),基本符合理論預(yù)期。CDF方案4方案3LTE多頻點組網(wǎng),異頻起測門限不宜過高,建議以RSRPCDF40%對應(yīng)的場強作為起測門限基準(zhǔn),并對各小區(qū)基于覆蓋邊界進(jìn)行門限優(yōu)化異頻切換策略(針對道路優(yōu)化)異頻測量時間盡量?。ㄒ粏訙y量就發(fā)起切換)異頻切換次數(shù)盡量少策略一、測量啟動門限設(shè)置低于DF等RSRP點電平約3dbA3offset設(shè)置略大于或等于3dbTrigtime盡量小策略二、測量啟動門限在保證順利切換的基礎(chǔ)上盡量低(-105dbm)threshold3InterFreq<-102threshold3aInterFreq>-90A3offset=6福僑大廈2小區(qū)Rsrp–晉安花園1小區(qū)Rsrp>3db且持續(xù)相應(yīng)的觸發(fā)時間,UE上報A3報告,觸發(fā)異頻間的切換。MainMenu異頻測量PDCCH功控及符號自適應(yīng)TM自適應(yīng)門限優(yōu)化室分高掉線優(yōu)化上行功控優(yōu)化通過權(quán)值修改DL發(fā)射功率PCIMOD3干擾異廠家切換IntroductionThePDCCHinLTEcarries:UE-specificschedulingassignmentsforDLresourceallocation,ULgrants,PRACH(PhysicalRandomAccessChannel)responses,ULpowercontrolcommands,andcommonschedulingassignmentsforsignalingmessages(suchassysteminformation,paging,etc.)Allocationrules0…3OFDMsymbolspersubframe2…4OFDMsymbolspersubframe(1.4MHz)AggregatedonCCEsQPSKonlybutdifferentaggregationlevels(AGGlevels)TheactualnumberofOFDMsymbolsoccupiedinanygivensubframeisindicatedinthePCFICH(PhysicalControlFormatIndicatorChannel),whichislocatedinthefirstOFDMsymbolofeachsubframeMBBCSNetworkEngineering/Dec2012PDCCHadaptationmechanismsDL-AMC-CCH+DL-PC-CCH
TargetistoadjustAGGlevelsbasedonUE-specificradioconditions,DifferentAGGlevelsresultinvariouscodingratewhichmaybringrobustness(highAGGlevel)orcapacitysavings(lowAGGlevel),Thealgorithmiscellorientedtooandtriestoavoidschedulingblocking,Powerboostingispossible.NumberofPDCCHsymbolspersubframestaticallyconfiguredwith.Available:RL10/RL15TDonwardsUsagebasedPDCCH
LTE616(FDDonly)ThefeatureadjuststhenumberofPDCCHsymbolspersubframe,Additionally,DL/ULbalancewithinPDCCHspacecanbeadaptedbasedonthecurrentloadsituation.Available:RL30onwards
LTE939(TDD)extendstheLTE616featuretoTDDdomainwiththesameconceptAvailable:RL35TDonwardsPresentationscopeWhenthenumberofPDCCHsymbolsisadaptivelyadjustedbased
onthecurrentloadconditionthiswillresultintheuserthroughputimprovementonconditionthatthereisrathersmallPDCCHload(fewUEstobescheduled)ForTDD,notPDCCHsofallsub-frameshavebothDLandULgrants,sodifferentalgorithmsareapplicabletodifferentcases.PurposeandbenefitsThefeatureadjuststhenumberofPDCCHsymbolspersubframeMBBCSNetworkEngineering/Dec2012TechnicaldetailsNote:Adjustmentislimited[1…3]OFDMsymbolsMaximumnumberofPDCCHsymbolsissetwithO&MparametermaxNrSymPdcchRemainingPDCCHutilizationishigherthanupperbound
ofdesiredPDCCHutilizationThereisnopowerforrelocation;noexcesspowerandnoboostingpossibleBlockingduetohashfunctionhappenedBlockingduetolackingschedulercapacityinULand/orDLRemainingPDCCHutilizationissmallerthanlowerboundofdesiredPDCCHutilizationThereisstillpowerforrelocation;excesspowerandpossibleboostingNoblockingduetohashfunctionhappenedNoblockingduetolackingschedulercapacityinULandDLDECREASEnumberofPDCCHsymbolsifINCREASEnumberofPDCCHsymbolsifAlgorithmstep1MBBCSNetworkEngineering/Dec2012IfmaximumamountofsymbolsisreachedandoverallrequiredPDCCHresourcesarestillnotfittingtoavailableamountofresources,thenUL/DLsplitismodifiedBlockingduetohashfunctioninDLschedulerhappenedBlockingduetohashfunctioninULschedulernothappenedPDCCHULusageissmallerthanDLDLsplitisincreasedPDCCHULusageishigherthanDLDLsplitisdecreasedBlockingduetohashfunctioninDLschedulernothappenedBlockingduetohashfunctioninULschedulerhappenedPDCCHULusageishigherthanDLULsplitisincreasedPDCCHULusageissmallerthanDLULsplitisincreasedChangeULPDCCHsplitifChangeDLPDCCHsplitifNote:Adjustmentislimitedforpre-definedR&DsplittresholdsTechnicaldetailsAlgorithmstep2MBBCSNetworkEngineering/Dec2012InTDDnotallPDCCHsubframescontainULandDLschedulinggrants,Therefore,twoinstancesofalgorithmareintroducedforsubframeswith:ULandDLPDCCHtransferOnlyDLPDCCHtransferAdaptationisnotrequiredforULsubframes,sincethere’snoPDCCH.Technicaldetails LTE939TDDspecificsDDSUUDDSUUDUL/DLDLDL/UL--UL/DLDLUL/DL--UL/DL90123456789PDCCHPDSCHPDCCHPDSCHPDCCHPDSCHGPUpPTSPUCCHPUCCHPDCCHPDSCHPDCCHPDSCHPDCCHPDSCHGPUpPTSPUCCHPUCCHPDCCHPDSCHPUSCHPUSCHPUSCHPUSCHPUCCHPUCCHPUCCHPUCCH3OFDMsymbolsDLgrantonlyinPDCCHregion2OFDMsymbolsDL/ULgrantsinPDCCHregionSpecialSubframe–noPDCCHDLgrantsULgrantsUL/DLconfiguration1.dpecialfsubframe7SubframetypeMBBCSNetworkEngineering/Dec2012ConfigurationManagementFeatureactivationAbbreviated
NameDescriptionRange/
StepDefault
ValueRecomendationactLdPdcchActivateordeactivatetheloadadaptivenumberofPDCCHsymbolsinacell
TheactualOFDMsymbolamountusedforPDCCHinaTTIisselectedfromvaluesbetweentheminimumreasonableamountofsymbolsfortheselectedDLbandwidthandmaximumallowednumberofPDCCHsymbols(maxNrSymPdcch).
ForLTETDDnoteverysub-frameneedadaptivePDCCHsymboladjustmentbecausenoteverysub-framehasPDSCH.true/falsefalseactLdPdcchcanbesetto'true'onlyiffollowingconditionsarefulfilled
-phichDur
issetto'Normal'AND
-maxNrSymPdcchisgreaterthan1AND
-DlchBwisconfiguredgreaterthan'5MHz'
IfactLdPdcchissetto'false',configuredULandDLsplitofPDCCHresources(pdcchUlDlBal)isfixedused.
IfactLdPdcchissetto'true',ULandDLsplitofPDCCHresourcesmaybeadjustedruntimebasedonloadandtheconfiguredvalue(pdcchUlDlBal)definestheinitialstartingpointthen.MBBCSNetworkEngineering/Dec2012Thresholdsfordesired(target)PDCCHutilizationarestrictlyR&DinternalparametersstatingthepercentageofCCEspaceusagePDCCH_UTIL_TH_UP; range[0..100]%,step1,default80PDCCH_UTIL_TH_DOWN; range[0..100]%,step1,default60ConfigurationManagementMaximumnumberofsymbolsforPDCCHAbbreviated
NameDescriptionRange/
StepDefault
ValueRecomendationmaxNrSymPdcchDefineshowmanyOFDMsymbolscanbeusedforPDCCHchanneltransmission.eNBselectsbasedonusagetheactualvalueforeachTTI,whichandvalueissignaledtotheUEsinPCFICH.1...3,step13IfmaxNrSymPdcchissetto'1',actLdPdcchcannotbeconfiguredto'true',formaxNrSymPdcchsetgreaterto'1'theallowedvalueofactLdPdcchstilldependsontheconfiguredvaluesfordlChBwandphichDurIncaseULPeakdatathroughput(10MHzand48PRBsforPUSCH)istobeconfigured,usefollowingPUCCHparametervalues:
deltaPucchShift(value1)
n1PucchAn(value10)
nCqiRb(value1)
maxNrSymPdcch(range1-2)(internal)MBBCSNetworkEngineering/Dec2012Featurecanbeactivatedviatheparameter:actLdPdcchWhenactive,maxNrSymPdcchshallbesettothevaluehigherthan1.OtherwisethefeaturedonohaveanyspacetooptimizethenumberofsymbolsOnceactivated,onemayobservethefeatureactivitybytheappropriatecountersDeploymentAspectsFeatureactivationandvalidationIntheBusyHour(12-14):3OFDMsymbolsareneededfor~5%ofthetime2OFDMsymbolsareneededfor~10%ofthetimeIntheremainingtime1OFDMsymbolisusedInthenightime(23-6):1OFDMsymbolisusedinmorethan95%oftime00:00Day1Day2Day3Day4Day5MBBCSNetworkEngineering/Feb2013OnofthecustomernetworkshasbeeninvestigatedwithrelativelylowloadInsuchaconditionsduring91%ofthetimeonly1OFDMsymbolwasneededWhenadditionalPDCCHcapacitywasneededinBusyHour,thenumberofsymbolswasadaptedDeploymentAspectsOFDMsymbolusagedistributionMBBCSNetworkEngineering/Dec20124GMaxconfirmslinklevelgainbymeansoflowerSINRrequirementforPDSCH1..2dBgaindependingonthenumberofallocatedPRBsMatlabsimulationprovesthealgorithmefficiencyUpto18%PDSCHspacegainwhenthereisnottoomanyUEstobescheduledModerategainwhenthenumberofUEsperTTIapproaches10Nogainforhighload
scenariosBenefitsandGainsFeatureperformancePDSCHlinklevelperformanceAlgorithmefficiencyMBBCSNetworkEngineering/Dec2012Thegainsfromthefeatureareprovedalsoindynamicsystem-levelsimulationenvironmentUpto17%morecapacityin10MHzbandwidthand21%in20MHzcanbeachievedwithLTE616withlownumerofUEsinthecellForhighnumberofUEsinthecelltoutilizeschedulinggain–asmuchUEsaspossiblearescheduled,thusmax3PDCCHsymbolsareusedalmostallthetime–thesamecaseasforfixed3symbolsforPDCCH(whenLTE616isswitchedoff).
BenefitsandGainsSystemlevelsimulationresults:FullBufferLTE616:onLTE616:offGainupto21%LTE616:onLTE616:offGainupto17%MBBCSNetworkEngineering/Dec2012ForuserthroughputperspectivegainsfromthefeatureLTE616arethesameasforthecellTP.Upto17%for10MHzand21%for20MHzbandwidth.BenefitsandGainsSystemlevelsimulationresults:FullBufferLTE616:onLTE616:offGainupto17%LTE616:onLTE616:offGainupto21%MBBCSNetworkEngineering/Dec2012There’salsoagaininPRButilizationwhichcomesfrom“l(fā)arger”sizeofthePRBwhenLTE616ison(additionalsymbolssavedfromPDCCHspace),thusthesameamountofdatacanbesentwithlowernumberofPRBs.16%gainforFTPburstytrafficvs.17%gainforFullBuffer(whenallresourcesareoccupiedallthetime).BenefitsandGainsSystemlevelsimulationresults:FTPLTE616:onLTE616:offGainupto16%LTE616:onLTE616:offGainupto16%MBBCSNetworkEngineering/Dec2012NodirectimpactonKPIs,althoughthelong-termaveragethroughputshouldimprove.ThethroughputKPIswillnotbeinfluencedbythisfeature,astheycanalsobeachievedwithfixedsettingsofMaxNrSymPdcchO&Mparameter.InFDDLTE616,thereare3counters:-Numberofsubframeswith1OFDMsymbolallocatedtoPDCCH-Numberofsubframeswith2OFDMsymbolallocatedtoPDCCH-Numberofsubframeswith3OFDMsymbolallocatedtoPDCCH
TDDjustreusesthem.PerformanceAspectsNewcountersintroducedwiththefeatureAssessingthesystemgainsThroughputonPDCPlayer:LTE_5292cE-UTRANaveragePDCPLayerCellThroughputDLLTE_5289cE-UTRANaveragePDCPLayerActiveCellThroughputULLTE_290a MinimumPDCPThroughputDLLTE_287a MinimumPDCPThroughputULLTE_291a MaximumPDCPThroughputDLLTE_288a MaximumPDCPThroughputULThroughputonRLClayer:LTE_5284bE-UTRANaverageRLCLayerCellThroughputDLLTE_5283bE-UTRANaverageRLCLayerCellThroughputUL
Numberofactiveusers(userswithdatainbuffer):LTE_5800bE-UTRANAverageActiveUEswithdatainthebufferDLLTE_5801bE-UTRANAverageActiveUEswithdatainthebufferULLTE_5802aMaximumActiveUEswithdatainthebufferpercellDLLTE_5803aMaximumActiveUEswithdatainthebufferpercellULNumberofconnectedusers(withDRB):LTE_5804bE-UTRANAverageActiveConnectedUesLTE_1082aActiveUEpereNBmaxVerificationthatthefeatureisworkingM8011C59PDCCH_1_OFDM_SYMBOL–numberofsubframeswith1OFDMsymbolallocatedtoPDCCHM8011C60PDCCH_2_OFDM_SYMBOL–numberofsubframeswith2OFDMsymbolallocatedtoPDCCHM8011C61PDCCH_3_OFDM_SYMBOL–numberofsubframeswith3OFDMsymbolallocatedtoPDCCHM8011C38CCE_AVAIL_ACT_TTI-CCEsavailablePDCCHschedulingRecommendationonthechoiceoffeatureactivationclusterFeatureprovidesgainsespeciallyiftheofferedtrafficislow-to-medium,sointheareas/periodswherethereisaroomtoexchangePDCCHwithPDSCH.Iftheno.ofusersperTTIisclosetoitsmaximumvalue,thefeaturegainwillbenegligible.Theexpectedgain(expressedintermsofsymbolsforPDSCH)ispresentedonthefigurebelowMBBCSNetworkEngineering/Dec2012PerformanceAspectsHowtotestthefeature定點測試-cat3NSNTDD現(xiàn)網(wǎng)測試PDCCH自適應(yīng)開關(guān)平均下行吞吐量(mbps)RSRPRSRQRSSISINRCQIMCSPDCCHCFIPDSCH
BLER天線模式UETxPower_pucchUETxPower_pusch關(guān)閉57.80-75-10-49311327(100%)PDCCHCFINum/s#1(0)
PDCCHCFINum/s#2(0)
PDCCHCFINum/s#3(600)1.4TM3/8-179開啟59.40-71-9-44331427(41.5%)/28(58.5%)PDCCHCFINum/s#1(0)
PDCCHCFINum/s#2(500)
PDCCHCFINum/s#3(100)0.3TM3-1811由于3類終端能力限制極好點測試性能無明顯提升拉遠(yuǎn)測試-cat3PDCCH自適應(yīng)開關(guān)平均下行吞吐量(mbps)RSRPRSRQRSSISINRCQIMCSPDCCHCFIPDSCH天線模式UETxPower_puschBLER關(guān)閉33.74-94-12-62201019PDCCHCFINum/s#1(0)5TM3/818PDCCHCFINum/s#2(1)PDCCHCFINum/s#3(599)開啟35.01-94-12-62211020PDCCHCFINum/s#1(0)6TM3/820由于3類終端能力限制拉遠(yuǎn)測試性能無明顯提升NSNTDD現(xiàn)網(wǎng)測試定點測試-cat4PDCCH自適應(yīng)開關(guān)平均下行吞吐量(mbps)RSRPRSRQRSSISINRCQIMCSPDCCHCFIPDSCH天線模式UETxPower_puschBLER關(guān)閉69.46-72-10-42321427(33.5%)/28(66.5%)PDCCHCFINum/s#1(0)0TM312PDCCHCFINum/s#2(0)PDCCHCFINum/s#3(600)開啟79.5-71-10-40301427(8.5%)/28(91.5%)PDCCHCFINum/s#1(0)0TM35PDCCHCFINum/s#2(500)PDCCHCFINum/s#3(100)極好點測試性能提升明顯NSNTDD現(xiàn)網(wǎng)測試PDCCH自適應(yīng)開關(guān)平均下行吞吐量(mbps)RSRPRSRQRSSISINRCQIMCSPDCCHCFIPDSCH天線模式UETxPower_puschBLER關(guān)閉25.76-107-11-7512.58.716.7PDCCHCFINum/s#1(0)7.2
21PDCCHCFINum/s#2(2)PDCCHCFINum/s#3(598)開啟37.99-101-11-69169.819.5PDCCHCFINum/s#1(0)6.6
20PDCCHCFINum/s#2(417)PDCCHCFINum/s#3(183)拉遠(yuǎn)測試-cat4拉遠(yuǎn)測試性能提升明顯NSNTDD現(xiàn)網(wǎng)測試結(jié)論CAT4終端極好點吞吐量和拉遠(yuǎn)吞吐量都有提升明顯CAT3終端測試由于終端能力限制極好點測試性能無明顯提升,拉遠(yuǎn)測試略有提升
PDCCH自適應(yīng)開關(guān)極好點平均下行吞吐量(mbps)拉遠(yuǎn)平均下行吞吐量(mbps)cat4終端關(guān)閉69.4625.76開啟79.537.99cat3終端關(guān)閉57.8033.74開啟59.4035.01NSNTDD現(xiàn)網(wǎng)測試MainMenu異頻測量PDCCH功控及符號自適應(yīng)TM自適應(yīng)門限優(yōu)化室分高掉線優(yōu)化上行功控優(yōu)化通過權(quán)值修改DL發(fā)射功率PCIMOD3干擾異廠家切換SINRTM2(CQI)TM3單流(CQI)TM3雙流(CQI)TM2/3(自適應(yīng))(CQI)0798817978288.758.6939.4999TM3單雙流門限建議-CQI取值2通道小天線SINRTM2(DLMCS)TM3單流(DLMCS)TM3雙流(DLMCS)TM2/3(自適應(yīng))(DLMCS)0.08.711.06.07.01.08.712.05.87.02.013.03.010.3SINRTM2(CQI)TM3單流(CQI)TM3雙流(CQI)TM2/3(自適應(yīng))(CQI)0.03.51.04.36.04.83.05.2SINR取值0~3時對于CQI3~7,考慮單雙流切換時機和減少乒乓切換后,定:TM3雙流到單流切換CQI門限=3;TM3單流到雙流切換CQI門限=7不同天線模式吞吐量對比8通道大天線TM7明顯強區(qū)間TM3雙流強區(qū)間波動區(qū)間(TM3雙流略強)SINR<99<SINR<2020<SINR1、TM8性能在任意時刻幾乎都差于TM3和TM7(TM8當(dāng)前版本性能不夠理想,特別是在中差點)2、TM7在SINR<9時性能表現(xiàn)優(yōu)異3、TM3雙流在SINR>20時性能表現(xiàn)優(yōu)異4、9<SINR<20區(qū)間內(nèi)雙流性能優(yōu)于單流,雖然由于RI限制,存在波動和模式間乒乓切換,但建議最大限度使用雙流模式間轉(zhuǎn)換參數(shù)設(shè)置建議-CQI取值8通道大天線SINR=9時,TM3雙流MCS=11,TM7雙流MCS=20,TM8單流MCS=17。簡單考慮CQI=MCS/2,策略如下TM3雙流到TM7CQI門限定為5,TM7到TM3雙流門限定為11考慮后期TM3/8自適應(yīng)TM3雙流到TM8單流CQI門限定為5,TM8單流到TM3雙流門限定為9MainMenu異頻測量PDCCH功控及符號自適應(yīng)TM自適應(yīng)門限優(yōu)化室分高掉線優(yōu)化上行功控優(yōu)化通過權(quán)值修改DL發(fā)射功率PCIMOD3干擾異廠家切換客戶投訴鴻宇7f經(jīng)常掉線,下載速率不穩(wěn)定該室分話務(wù)量較大問題描述DateRADIO_ACCESS_SUCC_RATIORADIO_DROP_RATIOEPS_BEARER_SETUP_COMPLETIONSENB_INIT_TO_IDLE_RNLENB_INIT_TO_IDLE_OTHERRADIO_BEARER_DROP_RATIO_5004ARB_REL_REQ_NORM_REL2013xxxx99.862.156991050.996992013xxxx99.712.65340902.623282013xxxx99.890.7872195150.6172942013xxxx99.812.93372256531.1337442013xxxx86.218.2567241226114.9862222013xxxx60.9555.3929861650433.9827582013xxxx47.9752.8377624100128.6376242013xxxx84.622.3477751737016.8476122013xxxx77.1431.9710021309910521.9598112013xxxx88.5815.466551025013.2665792013xxxx94.198.368432653527.0683942013xxxx99.111.9414442260.821441對鴻宇7f進(jìn)行了測試,發(fā)現(xiàn)在無線環(huán)境相當(dāng)良好的情況下存在大量的RRCrelease,原因都為other優(yōu)化分析
進(jìn)一步通過BTSlog發(fā)現(xiàn),掉線原因為大部分是由于CQIDTX上行失步導(dǎo)致了RRCrelease優(yōu)化步驟檢查鴻宇SCFC配置文件,發(fā)現(xiàn)Pucch上行功控都處于關(guān)閉狀態(tài),聯(lián)系到鴻宇7f的UE較多,Pucch上行干擾較大導(dǎo)致ENB無法正確解碼UE上報CQI,導(dǎo)致上行失步而釋放RRC鏈路。我們對以下參數(shù)進(jìn)行了優(yōu)化,如下表所示:參數(shù)名原始值修改值參數(shù)注釋ulpcEnableFALSETRUE上行閉環(huán)功控總開關(guān),若該參數(shù)FALSE,則ulpcPucchEn,ulpcPuschEn,ulpcSrsEn無效ulpcPucchEnFALSETRUEPUCCH閉環(huán)功控開關(guān)優(yōu)化效果修改后測試穩(wěn)定,未出現(xiàn)RRCrelease情況MainMenu異頻測量PDCCH功控及符號自適應(yīng)TM自適應(yīng)門限優(yōu)化室分高掉線優(yōu)化上行功控優(yōu)化通過權(quán)值修改DL發(fā)射功率PCIMOD3干擾異廠家切換Themaximalcellpowerreduction(dlCellPwrRed)dependsontheconfiguredmaximaloutputpower(pMax)asfollows
-forpMaxsetto49.0dBm,dlCellPwrRedmaximumvalueis10.0dB
-forpMaxsetto47.0dBm,dlCellPwrRedmaximumvalueis8.7dB
-forpMaxsetto46.0dBm,dlCellPwrRedmaximumvalueis7.7dB
-forpMaxsetto44.8dBm,dlCellPwrRedmaximumvalueis6.5dB
-forpMaxsetto43.0dBm,dlCellPwrRedmaximumvalueis4.7dB
-forpMaxsetto42.0dBm,dlCellPwrRedmaximumvalueis3.7dB
-forpMaxsetto41.8dBm,dlCellPwrRedmaximumvalueis3.5dB
-forpMaxsetto40.8dBm,dlCellPwrRedmaximumvalueis2.5dB
-forpMaxsetto40.0dBm,dlCellPwrRedmaximumvalueis1.7dB
-forpMaxsetto39.0dBm,dlCellPwrRedmaximumvalueis0.7dB
-forpMaxsetto38.8dBm,dlCellPwrRedmaximumvalueis0.5dB
-forpMaxsetto37.0dBm,dlCellPwrRedmaximumvalueis0.0dB
-forpMaxsetto10.0dBm,dlCellPwrRedmaximumvalueis20.0dBForRL25&35,
whilePMaxissetto40dBm,thedlCellPwrRedis1.7dB(currentmaxvalue),whichcannotfulfillfieldoptimizationrequirementinsomescenarios.Workaround:Morepowerreductionviasectorbeamweightsetting.背景通過SITEmanager設(shè)置權(quán)值修改功率,設(shè)置權(quán)值如下圖,在LNBTS創(chuàng)建DOWNLINKSECTORBEAMFORMINGWEIGHTCUSTOMIZEDPROFILE.步驟1根據(jù)天線類型設(shè)置相應(yīng)權(quán)值步驟2廣播波束65°
F頻段:幅度∣Ii∣0.490.9210.470.490.9210.47相位(°)01721701730172170173京信天線權(quán)值設(shè)定:這里需要注意,如果天線不一樣權(quán)值設(shè)定也是不一樣的,所以修改功率之前,一定要先確定天線類型按照附件中的表格。步驟3然后在LNCEL里選擇profile1則將此小區(qū)設(shè)置為以上權(quán)值。步驟4功率降低則將幅度乘以一個系數(shù)后輸入customerizedprofile.比如需要功率降低1DB,那么幅度每一列乘以0.891250938,四舍五入后,然后輸入customerizedprofile.系數(shù)公式:10^(-x/20),X為要降低的DB數(shù)降低DB數(shù)系數(shù)-10.891250938-20.794328235-30.707945784-40.630957344-50.562341325現(xiàn)場驗證測試小區(qū):天際網(wǎng)城3小區(qū)(一、二小區(qū)閉鎖,周圍較遠(yuǎn)的基站未閉鎖)TDL軟件版本:CD2.2天線廠家和型號:京信ODS-090R15NT06修改功率方法:
通過在SiteManager上創(chuàng)建一個可以修改權(quán)值的profile,然后由相關(guān)公式算得一個系數(shù)(需要降低功率的db數(shù)不同,系數(shù)也不同),由天線型號中的幅度乘以上邊的系數(shù)后設(shè)置來對應(yīng)的customerizedprofile;即通過修改天線權(quán)值來改變天線波束的功率測試方法: 1、選取位置相同的一個中點進(jìn)行原始值、功率降3db、功率降5db的定點對比測試 2、對該小區(qū)進(jìn)行原始值、功率降3db、功率降5db的拉網(wǎng)對比測試測試結(jié)果對比定點測試結(jié)果拉網(wǎng)測試結(jié)果定點測試時由原始值降低3db時,RSRP值降低接近3db,再降至5db時可以看到RSRP增大,天線波束有明顯變化;拉網(wǎng)測試RSRP有減弱的趨勢;定點測試(選取一個中點)RSRP(dBm)SINR(dB)T-PUT(Mbps)原始值-8017.544.6功率降3db-82.816.738.96功率降5db-79.219.442.78單小區(qū)拉網(wǎng)統(tǒng)計(平均值)RSRP(dBm)SINR(dB)T-PUT(Mbps)原始值-89.1214.7528.71功率降3db-95.6912.7628.5功率降5db-98.7811.9725.43RSRP覆蓋對比由RSRP拉線圖可以看到覆蓋由上到下有變?nèi)醯内厔萦捎赗SRP減弱,降低3db和5db,和周圍站點切換明顯增多
MainMenu異頻測量PDCCH功控及符號自適應(yīng)TM自適應(yīng)門限優(yōu)化室分高掉線優(yōu)化上行功控優(yōu)化通過權(quán)值修改DL發(fā)射功率PCIMOD3干擾異廠家切換測試環(huán)境:某大樓不同樓層的室內(nèi)覆蓋由兩個同頻的LTE提供。分別將這兩個小區(qū)
的PCI設(shè)置為模3干擾和非模3干擾兩種場景進(jìn)行對比測試。測試結(jié)果:
非模3干擾模3干擾服務(wù)小區(qū)PCI312312服務(wù)小區(qū)RSRP-76dBm
溫馨提示
- 1. 本站所有資源如無特殊說明,都需要本地電腦安裝OFFICE2007和PDF閱讀器。圖紙軟件為CAD,CAXA,PROE,UG,SolidWorks等.壓縮文件請下載最新的WinRAR軟件解壓。
- 2. 本站的文檔不包含任何第三方提供的附件圖紙等,如果需要附件,請聯(lián)系上傳者。文件的所有權(quán)益歸上傳用戶所有。
- 3. 本站RAR壓縮包中若帶圖紙,網(wǎng)頁內(nèi)容里面會有圖紙預(yù)覽,若沒有圖紙預(yù)覽就沒有圖紙。
- 4. 未經(jīng)權(quán)益所有人同意不得將文件中的內(nèi)容挪作商業(yè)或盈利用途。
- 5. 人人文庫網(wǎng)僅提供信息存儲空間,僅對用戶上傳內(nèi)容的表現(xiàn)方式做保護(hù)處理,對用戶上傳分享的文檔內(nèi)容本身不做任何修改或編輯,并不能對任何下載內(nèi)容負(fù)責(zé)。
- 6. 下載文件中如有侵權(quán)或不適當(dāng)內(nèi)容,請與我們聯(lián)系,我們立即糾正。
- 7. 本站不保證下載資源的準(zhǔn)確性、安全性和完整性, 同時也不承擔(dān)用戶因使用這些下載資源對自己和他人造成任何形式的傷害或損失。
最新文檔
- 2024年度餐飲連鎖企業(yè)食材集中配送服務(wù)合同3篇
- 2024年版船舶租賃合同格式3篇
- 2024年版施工單位與監(jiān)理單位合作協(xié)議
- 2024事業(yè)單位合同管理信息系統(tǒng)開發(fā)與維護(hù)合同2篇
- 2025年伊犁下載貨運從業(yè)資格證模擬考試題
- 2024商場餐飲品牌租賃與市場營銷策略合同3篇
- 洛陽師范學(xué)院《大數(shù)據(jù)基礎(chǔ)理論與應(yīng)用》2023-2024學(xué)年第一學(xué)期期末試卷
- 表面處理工程安裝施工承包合同
- 建筑綠化勞務(wù)合同
- 酒店行業(yè)技術(shù)人才招聘合同范本
- 半導(dǎo)體封裝過程wirebond中wireloop的研究及其優(yōu)化
- 15m鋼棧橋施工方案
- FZ∕T 97040-2021 分絲整經(jīng)機
- 應(yīng)聘人員面試登記表(應(yīng)聘者填寫)
- T∕CAAA 005-2018 青貯飼料 全株玉米
- s鐵路預(yù)應(yīng)力混凝土連續(xù)梁(鋼構(gòu))懸臂澆筑施工技術(shù)指南
- 撥叉831006設(shè)計說明書
- 10KV高壓線防護(hù)施工方案——杉木桿
- 石油鉆井八大系統(tǒng)ppt課件
- 對標(biāo)管理辦法(共7頁)
- R語言入門教程(超經(jīng)典)
評論
0/150
提交評論