Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Ericsson Technology Review - Issue 1, 2019

292 views

Published on

Our participation at MWC in Barcelona this year revealed that a steadily growing number of mobile network operators and representatives from various industries are keen to explore the myriad of new opportunities that 5G represents for their businesses. In particular, we found that many are curious to learn more about the role of 5G in Industry 4.0 and other industry transformations, where it enables manufacturing companies leverage automation and data exchange technologies that require seamless communication across industrial processes.

Fittingly, the feature article in this issue of the magazine explains how 5G can be used most effectively in the fully-connected factories of the future. We also have excellent articles about the role of distributed cloud in supporting emerging industrial use cases, the necessity of business support systems that can handle IoT use cases, and important technology choices to consider in the design of massive IoT devices. Last but not least, we have included two articles that provide expert guidance regarding two key aspects of 5G deployment.

Feel free to share links to the magazine and/or individual articles with your colleagues and other contacts via e-mail or social media. Happy reading!

Published in: Technology
  • Be the first to comment

Ericsson Technology Review - Issue 1, 2019

  1. 1. ERICSSON TECHNOLOGY C H A R T I N G T H E F U T U R E O F I N N O V A T I O N | V O L U M E 9 8 I 2 0 1 9 – 0 1 BOOSTINGSMART MANUFACTURING WITH5G DISTRIBUTEDCLOUD INAUTOMOTIVE& INDUSTRY4.0 USECASES TECHNOLOGY CHOICESFOR MASSIVEIoT DEVICES
  2. 2. CONTENTS ✱ 08 THE ADVANTAGES OF COMBINING 5G NR WITH LTE 5G at mid and high bands is well suited for deployment at existing site grids, especially when combined with low-band LTE. Adding new frequency bands to existing deployments is a future-proof and cost- efficient way to improve performance, meet the growing needs of mobile broadband subscribers and deliver new 5G-based services. 18 SIMPLIFYING THE 5G ECOSYSTEM BY REDUCING ARCHITECTURE OPTIONS Previous mobile generations have taught us that industry efforts to reduce fragmentation yield massive benefits. In the case of 5G, an industry effort to focus deployment on a limited set of key connectivity options will be critical to bringing 5G to market in a timely and cost-efficient way. 28 DISTRIBUTED CLOUD: A KEY ENABLER OF AUTOMOTIVE AND INDUSTRY 4.0 USE CASES Emerging use cases in the automotive industry – as well as in manufacturing industries where the first phases of the fourth industrial revolution are taking place – have created a variety of new requirements for networks and clouds. At Ericsson, we believe that distributed cloud is a key technology to support such use cases. 48 KEY TECHNOLOGY CHOICES FOR OPTIMAL IoT DEVICES The latest cellular communication technologies LTE-M and NB-IoT enable the introduction of a new generation of IoT devices that deliver on the promise of scalable, cost-effective massive IoT applications using LPWAN technology. However, a few key technology choices are necessary to create IoT devices that can support the multitude of existing and emerging massive IoT use cases. 60 BSS AND ARTIFICIAL INTELLIGENCE – TIME TO GO NATIVE The growing need to support disruptive services emerging from the IoT and 5G requires a fundamental transformation of business support systems (BSS). At Ericsson, we believe that the best way to achieve this is by forging BSS and artificial intelligence (AI) together to create truly AI-native BSS. FEATURE ARTICLE Boosting smart manufacturing with 5G wireless connectivity Industry 4.0 – the fourth industrial revolution – is already transforming the manufacturing industry, with the vision of highly efficient, connected and flexible factories of the future quickly becoming a reality in many sectors. Fully connected factories will rely on cloud technologies, as well as connectivity based on Ethernet Time-Sensitive Networking and wireless 5G radio. 38 08 Standalone NR and 5GC with appropriate features and coverage for addressed use cases. EPC eNBeNB eNB gNB eNBeNB eNB gNB NR NRNR-low LTE LTE LTELTELTE Option 1 Options 1, 3 Current industry focus Options 1, 2, 3 1 3 2 EPC+ 5GCEPC+     18 Local Regional Regional DCLocal DC MTSO MTSO Local and regional sites Service exposure HD maps HD maps Data exposure for automotive services Access sites Video stream ECU sensors HD maps Video stream ECU sensors HD maps Intelligent driving Intelligent driving Advanced driver assistance Advanced driver assistance Huge amount of data 28 Enterprise Strategic level Tactical level Operational level Business model #1 Business model #2 Common enterprise resources Resources common to all business models, for example: ∙ Party model ∙ Channels ∙ Credit check service Credit rules Channel serviceCustomer Contract Sales Onboarding Billing Settlement Rating Customer support Inten Intent Intent In 60 48 38
  3. 3. ERICSSON TECHNOLOGY REVIEW ✱ #01 2019 Ericsson Technology Review brings you insights into some of the key emerging innovations that are shaping the future of ICT. Our aim is to encourage an open discussion about the potential, practicalities, and benefits of a wide range of technical developments, and provide insight into what the future has to offer. a d d r e s s Ericsson SE-164 83 Stockholm, Sweden Phone: +46 8 719 00 00 p u b l i s h i n g All material and articles are published on the Ericsson Technology Review website: www.ericsson.com/ericsson-technology-review p u b l i s h e r Erik Ekudden e d i t o r Tanis Bestland (Nordic Morning) tanis.bestland@nordicmorning.com e d i t o r i a l b o a r d Håkan Andersson, Anders Rosengren, Mats Norin, Erik Westerberg, Magnus Buhrgard, Gunnar Thrysin, Håkan Olofsson, Dan Fahrman, Robert Skog, Patrik Roseen, Jonas Högberg, John Fornehed and Sara Kullman f e at u r e a r t i c l e Boosting smart manufacturing with 5G wireless connectivity by Kenneth Wallstedt, Fredrik Alriksson and Göran Eneroth a r t d i r e c t o r Liselotte Eriksson (Nordic Morning) p r o d u c t i o n l e a d e r Susanna O’Grady (Nordic Morning) l ay o u t Liselotte Eriksson (Nordic Morning) i l l u s t r at i o n s Jenny Andersen (Nordic Morning) c h i e f s u b e d i t o r Ian Nicholson (Nordic Morning) s u b e d i t o r s Paul Eade (Nordic Morning) i s s n : 0 0 1 4 - 0 17 1 Volume: 98, 2019 ■ i find it deeply gratifying to witness the growing enthusiasm among mobile network operators (MNOs) around the globe about the massive growth opportunities that 5G represents for their businesses. In particular, 5G is now widely recognized as a prime enabling technology of the fourth industrial revolution, helping manufacturing companies leverage automation and data exchange technologies that require seamless communication between all the participants and components in industrial processes. Using 5G effectively in the fully-connected factories of the future is the theme of the feature article in this issue of the magazine. Among other aspects, itexplainshow5Gcanprovidedeterministic ultra- reliable low-latency communication to bring wireless connectivity to demanding industrial equipment, like industrial controllers and actuators. Emerging industrial use cases in the automotive and manufacturing sectors, among others, are creating a variety of new requirements for networks and clouds. Our distributed cloud article explains how distributed cloud technology exploits key features in both 4G and 5G networks to enable an execution environment that ensures performance, short latency, high reliability and data locality. Like 5G, the Internet of Things (IoT) is also playing a pivotal role in Industry 4.0, as well as in transforming business and society in a myriad of other ways. In light of this, MNOs need business support systems (BSS) that can handle IoT use cases, which often involve complex business situations, and optimize CAPITALIZING ON THE POWER OF 5G ✱ EDITORIAL #01 2019 ✱ ERICSSON TECHNOLOGY REVIEW 7 outcomes with minimal manual intervention. In this magazine we argue in favor of architectural changes to traditional BSS to fully integrate artificial intelligence. As IoT use cases continue to grow and spread, it is critically important to take action to ensure that the devices are secure, both in terms of communication and data integrity end-to-end, from device to data usage. It is our opinion that certain key technology choices are necessary to achieve the desired device characteristics and create IoT devices that support the multitude of existing and emerging massive IoT use cases. While 5G is highly relevant for many industrial (and other) applications that reach far beyond traditional telco, it is also designed to address a myriad of challenges within the traditional telco sphere. One example of this is the way it enables MNOs to overcome the challenge of capacity exhaustion caused by the rapidly increasing data consumption of their subscribers. Rather than densifying 4G networks with new sites, we recommend that operators use 5G technology to add new frequency bands at existing 4G sites. Of course, one of the most critical aspects of a successful 5G deployment is the operator’s ability to support user equipment, radio network, core network and management products that are manufactured by a multitude of device and network equipment vendors. Achieving this can be more difficult than it sounds, however. We propose a smart approach to 5G deployment in this issue that reduces network upgrade cost and time, simplifies interoperability between networks and devices, and enables a faster scaling of the 5G ecosystem. I hope you will find the articles in this magazine valuable. Please feel free to share them with your colleagues and business associates. You can find all of the articles, along with those published in previous issues, at: www.ericsson.com/ericsson- technology-review 5GISNOWWIDELYRECOGNIZED ASAPRIMEENABLINGTECHNOLOGYOF THEFOURTHINDUSTRIALREVOLUTION ERIK EKUDDEN SENIOR VICE PRESIDENT, GROUP CTO AND HEAD OF TECHNOLOGY & ARCHITECTURE EDITORIAL ✱
  4. 4. ✱ COMBINING 5G NR WITH LTE COMBINING 5G NR WITH LTE ✱ 8 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 9 ThemainnewNRfrequencybandswilltypically beallocatedasTDDinthemid(3-6GHz)andhigh (24-40GHz)bands.Thesebandspresentseveral interestingchallengesandopportunities.Bymeans ofmeasurementsandradionetworksimulationsof coverageandcapacity,wehavedemonstrated thatitisfeasibletodeploybothmidandhigh (alsoknownasmillimeterWaveormmWave) bandsonexistingsites. Thankstobeamforming,afundamental techniqueinNR,theneedforsitedensificationis muchsmallerthananticipated–particularlywhen interworkingwithLTEisapplied.Beamforming andmassivemultiple-input,multiple-output (MIMO)techniquesalsoprovidehighercapacity fromexisting4Gsites,whichcreatesroomfornew 5G-basedservicesandusecasesinadditionto MBB. AT EXISTING SITES Combining5GNR 5G at mid and high bands is well suited for deployment at existing site grids, especially when combined with low-band LTE. Adding new frequency bands to existing deployments is a future-proof and cost-efficient way to improve performance, meet the growing needs of mobile broadband subscribers and deliver new 5G-based services. FREDRIC KRONESTEDT, HENRIK ASPLUND, ANDERS FURUSKÄR, DU HO KANG, MAGNUS LUNDEVALL, KENNETH WALLSTEDT High-frequencychallengesandopportunities Theuseofmidandhighbandsfor5Gmakesit possibletoutilizemuchhigherbandwidths. However,theincreasedcarrierfrequencycanalso makeitmorechallengingtoprovidecoveragethatis similartoexistinglow-banddeployments.Thereare threeprimaryreasonsforthis:(1)physicallimitson thepowerreceptioncapabilitiesofantennas;(2) radiofrequencyoutputpowerlimitations;and(3) increasedpropagationlosses,asshowninFigure1. The speed expectations and data consumption of mobile broadband (MBB) subscribers continue to grow rapidly. Already today, there are 4G networks in urban areas that are being densified with new sites (macro sites, small cells and indoor solutions, for example) as a result of spectrum exhaustion. Further, in regions such as western Europe and North America, the data demand per smartphone is projected to grow by 30-40 percent yearly [1], resulting in a four- to fivefold increase in five years. Adding new frequency bands at existing sites is a cost-efficient way to meet this demand and improve performance. The ability to achieve indoor coverage is particularly important, because the majority of the traffic is generated indoors [2]. ■ Manypeopleinthetelecomindustrytendto associatethedeploymentofhigh-frequencybands withpoorcoverage,whichresultsintheneedfornew sites,whichleadstohighdeploymentcosts.Thisis, however,notatallthecasefor5GNewRadio(NR) [3].5GNRisdesignedtomakeuseoffrequency bandsabove3GHzandoffersthepossibilityto introducenewfrequencybands–typicallyabove 3GHz–intoexisting4Gnetworks.Takingadvantage ofthispossibilitymakesiteasiertomeetthe increasingdemandsfromMBB-basedservices, whilesimultaneouslyensuringthatsiteandbackhaul infrastructureinvestmentscanbereused.5GNR isalsoavailableforuseinnewbandsbelow1GHzand existing3G/4Gbands.Smoothmigrationfrom4G to5GinexistingspectruminaRANcanbedone bymeansofspectrumsharing,whereNRis introducedinparallelwithLTE. THANKSTOBEAMFORMING... THENEEDFORSITEDENSIFICATION ISMUCHSMALLERTHAN ANTICIPATED Figure 1 Schematic indication of antenna and propagation factors affecting downlink coverage positively (blue) or negatively (red) compared to coverage at a reference frequency of 1.8GHz. The numbers are indicative and may vary. Output power and regulatory requirements 3.5GHz: 0dB (typical) 28GHz: -10 to -15dB (typical) Antenna gain and beamforming 3.5GHz: +9dB (typical) 28GHz: +15dB (typical) Outdoor to indoor propagation loss 3.5GHz: -0 to -2dB (-0 to -3dB for IRR glass) 28GHz: -1 to -10dB (-2 to -20dB for IRR glass) Outdoor propagation loss 3.5GHz: 0 to -2dB 28GHz: 0 to -7dB Antenna gain and beamforming 3.5GHz: +6dB 28GHz: +9dB Rx effective antenna area 3.5GHz: -6dB 28GHz: -24dB withLTE THE ADVANTAGES OF
  5. 5. ✱ COMBINING 5G NR WITH LTE COMBINING 5G NR WITH LTE ✱ 10 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 11 the1.8GHzreference.Evenso,usersintheworst positionsrequirethesupportofalowerfrequency band,especiallyintheuplink(UL)direction. Forhighbands(around30GHz),thesituation differssubstantiallyfromthereference.Verygood outdoorcoverageisachievedonexistinggrids. Outdoor-to-indoorcoveragecanbeachievedby targeteddeploymentswithline-of-sighttothe buildingsintendedtobecovered. Measuredbeamformingperformance andoutdoor-incoverage Earlyproofpointsofthe5Gconceptandits performancecanbeobtainedfrommeasurements inaradionetworkprototype.Ericssonhas developed5Gprototypesforseveral5Gfrequencies, including3.5GHzand28GHz.Initialtrial deploymentsaretypicallysetupwithafewradio sitesandoneorafewmobileterminals,allowingfor acontrolledmeasurementenvironment.Testresults onbeamformingperformancearereportedin references[5],[6]and[7].Theresultsdemonstrate thathighantennagainscanindeedberealized throughbeamforming,andthatthebeamforming isabletotrackfast-movinguserswithsustained communicationquality.Moreover,goodindoor coveragecanbeachievedwith5Gat3.5GHz, provingthefeasibilityofdeploying5Gatexisting 4Gsites.Oneexamplefromourmeasurementsis showninFigure2,whereindoorthroughputina buildingatthecelledgereaches200-400Mbps onan80MHzcarrierusingconservativerank-2 MIMOtransmission. Butthehigherfrequenciesalsoallowhigherantenna gainstobegeneratedwithoutincreasingphysical antennasize.5Gcanutilizetheseincreasedantenna gainsthroughbeamformingbothatthetransmitter andatthereceiver,whichhelpsmitigatetheimpact oncoverageathigherfrequencies. Additionally,increasingthefrequencywillallow theantennastobecomesmallerwhilemaintaining thesameantennagain.Itisimportanttonotethat anyfixed-gainantennainreceivingmodeactually captures20dBlessenergyforeachtenfoldincrease ofthefrequency.Thisisoftenmisunderstoodasa propagationloss,wheninrealityitisaresultofa decreasingeffectiveantennaarea.Ifthephysical antennaareaoftheantennaismaintained,itspower capturecapabilitiesbecomeindependentof frequency,whileitsantennagain,forbothreception andtransmission,growswiththefrequencyatthe sametimeasthebeamwidthbecomessmaller. Thus,athigherfrequencies,thereisatrade-off betweenreducingtheantennasizeandincreasing theantennagain.Coverageandimplementation aspectsdeterminethesweetspot. Theachievableoutputpowerathigherfrequency bandssuchasmmWavefrequenciescanalsobe limitedbypoweramplifiertechnologyandby regulatoryrequirements[4].Theoretically,the antennagainofafixed-sizetransmittingantenna wouldgrowby20dBperdecadeinfrequency (dB/decade),butinpracticetheincreaseinEIRP (effectiveisotropicradiatedpower)maybesmaller duetosuchconstraints. Electromagneticwavepropagationincellular networksinvolvessomeprocessesthatarestrongly frequency-dependent,suchasdiffractionor transmissionthrough,forexample,wallsorfoliage, butalsootherssuchasfreespacepropagationand reflectionorscatteringthatshowlittletono differenceoverfrequency.Effectively,theoutdoor propagationlossissimilarorincreasesslightlywith increasedfrequency,asindicatedinFigure1. Outdoor-to-indoorpropagationlossescanbe challengingtoovercome,especiallyforbuildings equippedwiththermally-efficientwindowglass, whichcanaddupto20-40dBofadditionallossat agivenfrequency.Whenincreasingthefrequency, theoutdoortoindoorlossesalsotendtoincrease, particularlyfordeepindoorlocations.Thisincrease issmalltomoderateforregularbuildingsbutcan bestrongforthermally-efficientbuildings, asshowninFigure1. Thehigherpropagationlossescanbemitigatedby usinghigh-gainantennasonbothtransmittersand receivers.Theseantennasbecomedirective,forming beamswithstronggainincertaindirections,andlow gaininotherdirections.Thebeamsneedtobesetup andmaintainedtopointintherightdirectionsin ordertosupportmobility.InNR,thisissupported bybeammanagement.Besidesthebenefitof amplifyingthesignalinthedesireddirection, beamformingalsoattenuatesthesignalinother directions,leadingtolessinterferenceandbetter channelquality.Thiscanbedonetotheextentthat multipleusers,usingdifferentbeams,can communicatewithabasestationonthesame frequencyandtimeresource.Thisisknownas multi-userMIMO(MU-MIMO),anditenables asignificantcapacityimprovement. Evenwithbeamforming,usingexistingsitegrids, itcanbedifficulttoreachfullcoverageonhigher frequencies.Butsincealowerfrequencybandtends tobeavailable,thisisnotaproblem.Usersoutof coverageonthehigherfrequenciessimplyfallback tothelowerfrequencybands.Thiscanbe accomplishedbyinterworkingtechniquessuchas dualconnectivityorcarrieraggregation.Theresult isa‘forgiving’situation,whereamid-orhigh-band deploymentdoesnotneedtobedimensionedfor 100percentcoverage.Instead,itsimplytakes careofthetrafficthatitcovers. Tosummarize,thenumbersinFigure1illustrate thattheuseoftoday’stechnologies,powerlevelsand beamforminggainsonthemidband(3-6GHz) providesbetterdownlink(DL)coveragethan THEMIDBAND(3-6GHZ) PROVIDESBETTERDOWNLINK COVERAGETHAN THE1.8GHZREFERENCE Figure 2 5G outdoor-in throughput measurement results from an NR 3.5GHz radio prototype NR prototype base station         128Tx 3.5GHz, 80MHz bandwidth 5W output power NR prototype UE 8Rx Downlink throughput at 3.5GHz > 400Mbps 200-400Mbps 100-200Mbps 50-100Mbps 10-50Mbps 1-10Mbps Out of coverage THEHIGHERPROPAGATION LOSSESCANBEMITIGATEDBY USINGHIGH-GAINANTENNAS
  6. 6. ✱ COMBINING 5G NR WITH LTE COMBINING 5G NR WITH LTE ✱ 12 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 13 NRsystem,usingawiderbandwidth,beamforming andMU-MIMO.Theabilitytoserveusersinpoor coverageareasonalowerbandavoidsthe consumptionofextensiveresourcesonthe 3.5GHzband,makingitmoreefficient.Toquantify thebenefitofintroducingNR,wemeasuredthe maximumtrafficloadforwhich(95percentof) theusersstillachieveauserthroughputexceeding 20Mbps.WhenaddingNRintheDLdirection, thismaximumtrafficloador‘capacity’increases byafactorofeightfrom1Gbps/km2 to8Gbps/km2 (correspondingto135GB/subscriber/month, assuming10,000subscribersperkm2 andabusy hourtrafficof8percentofthedailytraffic).IntheUL direction,thecapacitygainissmallerthantheDL duetoTDDasymmetry(25percentfortheUL)and alowertransmitpower.Thecapacitygainsobserved herearetypicalforalow-riseurbanscenariowith decentcoverage.Thegainsarescenario-dependent andtypicallyincreasewithimprovedcoverageand increasedverticalspreadofusers,anddecreasewith worsecoverageandasmallerverticalspread. Predictedurbanmid-bandcoverage andcapacity Topredict5Gcoverageandcapacityonalarger scale,wehaveperformedradionetworksimulations. WechoseapartofcentralLondonwithaninter-site distanceofapproximately400m,whichis representativeofmanyEuropeanurbanareas. Similarstudiesofmajorcitiesinotherpartsofthe world,includingAsiaandtheUS,indicatethatthe findingsfromthisstudyarealsoapplicableinthose scenarios.Radiobasestationcharacteristicssuch asbeamformingcapabilities,powerandsensitivity reflecttheimplementationsofthefirstproduct generations,andterminalsaremodeledwithexpec- tedtypicalsmartphonecharacteristicsformidand highbands.Fourand32receiveantennasareassumed forterminalsinmidandhighbands,respectively. Formaximalfidelity,adigital3Dmapisused togetherwithanaccurate3Dsite-specificpropagation model,explicitlycapturingrelevantpropagation phenomenaalongthepropagationpaths[8]. WehavemodeledLTEsystemsoperatingat 800MHz,1.8GHzand2.6GHz,aswellasanNR systemoperatingat3.5GHz.Thisconfigurationis representativeofthenon-standaloneversionofNR thatwasdevelopedin3GPPRel-15.TheLTEsystem usesFDD,2x10MHzat800MHzand2x20MHz ateachof1.8GHzand2.6GHzaddingupto100MHz pairedspectrum,andregularsectorantennas. TheNRsystemusesTDD,100MHzofunpaired spectrum,anda64T64Rantennaarrayof8x8cross- polarizedantennas.Weapplieduser-specificdigital beamforming,andMU-MIMOwithmultiplexingof uptofourusersissupportedbothintheDLandUL. WhenLTEandNRsystemsareevaluatedtogether, carrieraggregationbetweenLTEsystemsanddual connectivitybetweenLTEandNRcarriers areappliedfortheinterworking.Althoughnot consideredinthisevaluation,thereareseveral interestingpossibilitiestoevolvetheLTEsystems –withmoreadvancedantennas,forexample. Figure3showsDLandULcoverageinterms ofachievabledataratesinanunloadednetwork withoutinterference.Eightypercentofusersare indoors,andtheyareshownonlyfrommiddlefloors. WhenexistingLTErooftopsitesarereusedwith 3.5GHz,bothindoorandoutdoorusershavevery goodcoverageintheDL.Theblacklineinthe colorbarindicatesthat95percentoftheindoor subscribershavecoveragefor200MbpsintheDL comparedwith50Mbpswhenaggregatingall LTEsystems(notshowninthefigure).Inaddition, 95percentofoutdooruserscanexceed500Mbps intheDLwithNR3.5GHzalone.TheULismuch morelimitedwith3.5GHzalone.NR-LTE interworkingimproves,andmanyoftheblankspots inthe3.5GHzbandarecovered.Theremaining areaswithpoorcoverageareconcentratedtoinside largebuildingswithhigh-lossouterwalls.These buildingsaresuitablecandidatesforindoor deployments.Comparingthegainsfromadding 3.5GHzintheDLandUL,itisclearthatthegains arelargerintheDL.ThisisduetoaDL-heavy TDDasymmetry(75percent)at3.5GHz, andthefactthattheUL,becauseofthelower transmitpower,ismorepower-limitedandthus gainslessfromadditionalbandwidth. Whentrafficloadincreases,moreusersareactive simultaneously,sharingthebasestationcapacity, causingincreasedinterferencelevels,andleadingto areductioninuserthroughputcomparedwiththe unloadedcase.Theseeffectsaremitigatedbythe Figure 3 DL and UL coverage maps. The black line in the color legends represents the fifth percentile of an indoor user data rate, and the purple areas indicate antenna positions. The white circles mark indoor areas with limited coverage, improved by interworking. > 200Mbps 50-200Mbps 20-50Mbps 5-10Mbps 2.5-5Mbps 1-2.5Mbps Out of coverage > 200Mbps 50-200Mbps 10-50Mbps 5-10Mbps 2.5-5Mbps 1-2.5Mbps Out of coverage 600 400 200 0 -200 -400 -600 600 400 200 0 -200 -400 -600 Uplink NR 3.5GHz alone Uplink NR LTE interworking -600 -400 -200 0 200 400 600 -600 -400 -200 0 200 400 600 > 400Mbps 200-400Mbps 100-200Mbps 50-100Mbps 10-50Mbps 1-10Mbps Out of coverage 600 400 200 0 -200 -400 -600 -600 -400 -200 0 200 400 600 Downlink NR 3.5GHz alone 〉〉 Better user data speeds – 95 percent of indoor subscribers have more than 200Mbps with today’s typical site grids. 〉〉 Higher capacity – adding NR 100MHz TDD (75 percent DL) on top of LTE with 2x50MHz paired spectrum provides an eight times higher DL capacity than using only LTE. Normalized with the 1.5 times higher spectrum usage, NR is thus five times more efficient. BENEFITSOFOVERLAYING5G NR3.5GHZATEXISTINGSITES
  7. 7. ✱ COMBINING 5G NR WITH LTE COMBINING 5G NR WITH LTE ✱ 14 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 15 wouldbeagoodexampleofacapacity-drivenone. Passivedistributedantennasystems(DASs)are currentlythemostcommonsolutionusedfor indoordeployments. ThehardwarecomponentsofapassiveDASoften haveanoperatingfrequencyrangethatislimitedto bandsbelow3GHz,whichmeansthataddingthe newNRmidorhighbandsrequiresanew5Gindoor solution.Aradiodot[9]solutionat3.5GHzprovides goodcoverageandmuchhigherspeedsthancurrent LTEbandsatthesameradionodedensity,aswellas consuminglesspowerthanaDAS.Forextreme demandsintermsofuserspeedsorcapacity,an indoorsolutionbasedonmmWavesmallcellsmight bethebestchoice.Inthiscase,itisimportantto deployamid-bandcoveragecomplement. Conclusion Thekeybenefitsofdeploying5GNewRadiowith midbands(3-6GHz)atexisting4Gsitesarethat doingsoresultsinasignificantperformanceboost andallowsformaximalreuseofsiteinfrastructure investments.ByaddingNRwith100MHzunpaired spectrum,itispossibletoachieveeighttimeshigher downlinkcapacityrelativetoLTE(2x50MHzpaired spectrum)alongwithimproveddownlinkdatarates –bothoutdoorsandindoors–bymeansofmassive MIMOtechniquessuchasbeamformingand multi-userMIMO.Uplinkcoveragedeepindoors ismaintainedthroughinterworkingwithLTE and/orNRonlowbandsusingdualconnectivity orcarrieraggregation(new,refarmedorbyusing LTE/NRspectrumsharing).Asaresultofthese possibilitiesin5GNR,growingdatademands canbemetwithlimitedsitedensification. Furtherspeedandcapacityincreasescanbe attainedbydeploying5GNRathighbands (26-40GHz),alsoknownasmmWaves.Thehigh bandsareparticularlyeffectiveoutdoorsandinside buildingswithline-of-sightfromthedeployedradio nodeandwithlowwalllossproperties.Buildings thathaveorneeddedicatedindoorsolutionsdueto highpenetrationlossandinteriorlossescanbe successfullyupgradedwithupcomingNRbands forhigherspeedsandcapacityatsimilarradionode densitytothoseusedforLTEin-building deploymentstoday. Predictedurbanhigh-bandcoverage andcapacity ThewidebandwidthavailableonmmWave spectrumcanprovidefurtherincreaseddatarates andadditionalcapacityontopofthecombined 3.5GHzmid-bandNRandLTEsystem.Higher frequenciesallowahighergainofantennaarrayat thesamephysicalarea–bothinabasestationand atauserterminalside–soastoincreasethe maximumantennagain. SimulationstudiesinthecentralLondonscenario showthatanNR200MHzTDDsystemat26GHz withthe256T256Rantennaarrayof16x16cross- polarizedantennascanprovideverygoodDL coveragetooutdoorusers–forexample,50-60 percentapproaching1Gbps.Withlargerspectrum allocationssuchas400MHz,itispossibletoreach multi-Gbpsspeeds.Whenthereisline-of-sightfrom thebasestationtoabuildingandthebuildingisa low-losstype,thereisalsoagoodchancethat indooruserswillbewellcovered. Ourresultsshowthatdeployingthe3.5GHzand 26GHzbandonexistingmacrositescanprovide acapacityimprovementofapproximately10times comparedwiththeLTEsystemsinlowandmid bands.Thisadditionalgainisbecause26GHz offloadsthelowerfrequencybandsbylettinggood- coverageusersutilizeanadditional200MHz, whichtherebyimprovesoverallperformance. ApplyingmmWavespectrumatstreet-levelsites canalsobeagoodalternative.Byplacingantennas onlampposts,outerwallsandthelike,itispossible toavoidtypicaldiffractionlossesfromrooftopsand achieveshorterdistancestousersonoutdoor hotspotsorintargetedbuildings.Oursimulation studiesintheLondonscenarioindicatethatthe street-levelradiodeploymentofanNRsystemwith 64T64Rantennaarrayprovidesgoodcoverageboth innearbyoutdoorareasandforindoorusersinlow- lossbuildingswithline-of-sighttothebasestation. Suburbanandruraldeploymentconsiderations Despitethetypicallylargercellsinsuburbanand ruralscenarios,itispossibletoachievesimilarresults tothosethatwehaveseeninurbanscenariosdueto differencesintheradiopropagation.Whiletheurban environmentischaracterizedbyrelativelylow antennas,frequentlargeobstaclesandlarge, highly attenuatingbuildings,thesuburbanandrural environmentshavetallerantennas,fewerobstacles andsmallersizedbuildingswithwalltypesthat areeasiertopenetrate.Thiscompensatesforthe differencesincellrange,andasaresultitistypical toachieveverygoodperformanceinsuburban andruralscenariosaswell. Indoordeployments In-buildingdeploymentsplayacentralrolein providinggoodindoorperformanceinmanypartsof theworldtoday.Largebuildingswithhighbuilding entrylossesareanexampleofacoverage-driven in-buildingdeployment,whereasacrowded publicvenuelikeatrainstationorastadium APPLYINGMMWAVE SPECTRUMATSTREET-LEVEL SITESCANALSOBEAGOOD ALTERNATIVE Terms and abbreviations DAS – Distributed Antenna System | DL – Downlink | IRR – Infrared Reflective | MBB – Mobile Broadband | MIMO – Multiple-input, Multiple-output | mmWave – Millimeter Wave | MU-MIMO – Multi-User Multiple-input, Multiple-output | NR – New Radio | RAN – Radio Access Network | Rx – Radio Receiver | Tx – Radio Transmitter | UL – Uplink ITISPOSSIBLETOACHIEVE EIGHTTIMESHIGHERDOWNLINK CAPACITYRELATIVETOLTE
  8. 8. ✱ COMBINING 5G NR WITH LTE COMBINING 5G NR WITH LTE ✱ 16 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 17 theauthors Fredric Kronestedt ◆ joined Ericsson in 1993 to work on RAN research. Since then he has taken on manydifferentroles, includingsystem design and system management. He currently serves as Expert, Radio Network Deployment Strategies, at Development Unit Networks, where he focuses on radio network deployment and evolution aspects for 4G and 5G. Kronestedt holds an M.Sc. in electrical engineering from KTH Royal Institute of Technology, Stockholm, Sweden. Henrik Asplund ◆ received his M.Sc. in engineering physics from Uppsala University, Sweden, in 1996, and joined Ericsson the same year. His current positionisMasterResearcher, Antennas and Propagation, at Ericsson Research, with responsibility for propagation measurements and modeling within the company and in cooperation with external organizations such as 3GPP and ITU-R. He has been involved in propagation research supporting predevelopment and standardization of all major wireless technologies from 2G to 5G. Kenneth Wallstedt ◆ is Director, Technology Strategy, in Ericsson’s CTO office, where he focuses on the company’s radio and spectrum management strategy. He joined Ericsson in 1990 and since then he has held various leading positions in Ericsson’s research, development and market units in Canada, Sweden and the US. He holds an M.Sc. in electrical engineering from KTH Royal Institute of Technology in Stockholm, Sweden. Du Ho Kang ◆ joined Ericsson Research in 2014 and currently serves as a Senior Researcher. He holds a Ph.D. in radio communication systems from KTH Royal Institute of Technology, Sweden, and an M.Sc. in electrical and electronics engineering from Seoul National University, South Korea. His expertise is concept developments of 4G/5G radio networks and performance evaluation toward diverse international standardizationandspectrum regulation bodies including 3GPP RAN, CBRS alliance, Multifire alliance (MFA), ETSI BRAN and ITU-R. Kang’s particular interest at present is developingsolution conceptsforinternetworking and massive MIMO for 5G base station products. Magnus Lundevall ◆ is Expert, Radio Network Performance, in Ericsson’s RD organization, where he currently focuses on 5G radio network deployment and evolution strategies. He joined Ericsson in 1998 and has 20 years of experience in radio network modeling, simulation and performance analysis. He holds an M.Sc. in electrical engineering from KTH Royal Institute of Technology in Stockholm, Sweden. Anders Furuskär ◆ joined Ericsson Research in 1997 and is currently a senior expert focusing on radio resource management and performance evaluation of wireless networks. He holds an M.Sc. in electrical engineering and a Ph.D. in radio communications systems, both from KTH Royal Institute of Technology in Stockholm, Sweden. Saknar bild på Du Ho Kang Further reading ❭❭ 5G deployment considerations, available at: https://www.ericsson.com/en/networks/trending/insights-and- reports/5g-deployment-considerations ❭❭ Massive MIMO increasing capacity and spectral efficiency, available at: https://www.ericsson.com/en/ networks/trending/hot-topics/5g-radio-access/massive-mimo ❭❭ Going massive with MIMO, available at: https://www.ericsson.com/en/news/2018/1/massive-mimo-highlights ❭❭ Superior indoor coverage with 5G Radio Dot, available at: https://www.ericsson.com/en/networks/ offerings/5g/5g-supreme-indoor-coverage References 1. EricssonMobilityReport,June2018,availableat: https://www.ericsson.com/en/mobility-report/reports/june-2018 2. Ericsson ConsumerLab report, Liberation from Location, October 2014, available at: https://www.ericsson. com/res/docs/2014/consumerlab/liberation-from-location-ericsson-consumerlab.pdf 3. 5G NR: The Next Generation Wireless Access Technology, 1st Edition, August 2018, Dahlman, E; Parkvall, S; Sköld, J, available at: https://www.elsevier.com/books/5g-nr-the-next-generation-wireless-access-technology/ dahlman/978-0-12-814323-0 4. GSMA, 5G, the Internet of Things (IoT) and Wearable Devices: What do the new uses of wireless technologies mean for radio frequency exposure?, September 2017, available at: https://www.gsma.com/ publicpolicy/wp-content/uploads/2017/10/5g_iot_web_FINAL.pdf 5. IEEE, Beamforming Gain Measured on a 5G Test-Bed, June 2017, Furuskog, J; Halvarsson, B; Harada, A; Itoh, S; Kishiyama, Y; Kurita, D; Murai, H; Simonsson, A; Tateishi, K; Thurfjell, M; Wallin, S, available at: https://ieeexplore.ieee.org/document/8108648/ 6. IEEE,High-SpeedBeamTrackingDemonstratedUsinga28GHz5GTrialSystem,September2017,Chana,R; Choi, C; Halvarsson, B; Jo, S; Larsson, K; Manssour, J; Na, M; Singh, D, available at: http://ieeexplore.ieee.org/ document/8288043/ 7. IEEE,5GNRTestbed3.5GHzCoverageResults,June2018,Asplund,H;Chana,R;Elgcrona,A;Halvarsson,B; Machado, P; Simonsson, A, available at: https://ieeexplore.ieee.org/document/8417704/ 8. Proceedings of the 12th European Conference on Antennas and Propagation (EuCAP 2018), A set of propagationmodelsforsite-specificpredictions,April2018,Asplund,H;Johansson,M;Lundevall,M;Jaldén,N, 9. Ericsson Radio Dot System, available at: https://www.ericsson.com/ourportfolio/radio-system/radio-dot-system
  9. 9. ✱ SIMPLIFYING THE 5G ECOSYSTEM SIMPLIFYING THE 5G ECOSYSTEM ✱ 18 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 19 BY REDUCING ARCHITECTURE OPTIONS Simplifyingthe Previous mobile generations have taught us that industry efforts to reduce fragmentation yield massive benefits. In the case of 5G, an industry effort to focus deployment on a limited set of key connectivity options will be critical to bringing it to market in a timely and cost-efficient way. TORBJÖRN CAGENIUS, ANDERS RYDE, JARI VIKBERG, PER WILLARS The multiple connectivity options in the 3GPP architecture for 5G have created several possible deployment alternatives. Initial deploymentsfocusonoptions3(non-standalone New Radio) and 2 (standalone New Radio). However, the deployment of several additional options would create a level of complexity that impacts the whole 5G ecosystem – across operator network operations, equipment vendors and user equipment (UE) chipset vendors as well as spectrum assets. To avoid ecosystem fragmentation, we believe that the best approach is to limit the number of options that are deployed. ■ Thereismuchmoretointroducing5Gthan simplydeployingNewRadio(NR)technology.Fora successful5Glaunch,theoperatorneedstosecurea networkthatincludesend-to-end(E2E)capabilities alignedacrossdevices,RAN,coreandmanagement systems.5Gisalsoatechnologytransformationfor operatorsstrivingformoreflexibilityandspeedin networkdeployment–andwithanexpectationof beingabletoaddressnewbusinessopportunities withusecasesbeyondmobilebroadband(MBB). Oneofthekeystrategictopicsthatoperatorsneedto decideoniswhichconnectivityoptionstosupportin thenetworktoaddressthetargetedusecases. 5Gconnectivityoptions InRelease15,the3GPP[1]hasdefinedmultiple architecturaloptionsforaUEtoconnecttothe network,usingLTE/eLTEand/orNRaccessto connecttoEvolvedPacketCore(EPC)or5GCore (5GC)networks.Anewuseofdualconnectivityhas alsobeenappliedtouseLTE/eLTEandNRasthe masterorsecondaryradioaccesstechnology(RAT) 5Gecosystem Figure 1 UE connectivity options Connectivity option Core network Master RAT Secondary RAT 3GPP term 3GPP release Option 1 EPC LTE - LTE Rel. 8 Option 3 EPC LTE NR EN-DC Rel. 15, Dec 2017 Option 2 5GC NR - NR Rel. 15, June 2018 Option 4 5GC NR eLTE NE-DC Rel. 15, March 2019 Option 5 5GC eLTE - eLTE Rel. 15, June 2018 Option 7 5GC eLTE NR NGEN-DC Rel. 15, March 2019 indifferentcombinations.Thishasresultedinsix connectivityoptionsforaUE,asshowninFigure1. Notethatwhiletheoptionterminologyisnot explicitlyusedinthe3GPPstandardsspecifications, itoriginatesfromthe5Gstudyphaseof3GPP Release15andiswidelyusedintheindustry. ThesixconnectivityoptionsshowninFigure1 definehowanysingleUEisconnectedtothe networkatagiventime.Inmostcases,anetworkwill supportasetofsuchoptionssimultaneously.One basestationmayhavedifferentUEsconnectedvia differentconnectivityoptions,aswellasmovinga UEconnectionbetweentheoptionsdependingon factorssuchasradioconditions.LegacyLTE/EPC (option1)isthebaseline,andtheindustryhasan alignedviewthattheinitial5Gdeploymentsare basedonoptions3and2.Thenextstep,therefore, istoestablishindustryalignmentonthepotential useofoptions4,5and7. Theneedforindustryalignment Mobilenetworkoperatorsthatdeploy5Gmustbe abletosupportUE,radionetwork,corenetworkand managementproductsthataremanufacturedbya multitudeofdeviceandnetworkequipmentvendors. Withmultipleconnectivityoptions,andevenmore possiblecombinationsofoptions,thereisahighrisk thatdifferentoperatorswilldeploydifferentoptions, inadifferentorder.Ifthathappens,chipset,device andnetworkequipmentvendorsarelikelytoget contradictoryrequirementsfromdifferentoperators ormarkets.Thiswouldcausesignificantproductand integrationcomplexity,aswellascreating interoperabilityissuesthatprolongthetimeittakes toestablishacompleteecosystemthatsupportsthe deployedoptions. Thecomplexitycausedbyamultitudeofdeployed connectivityoptionswouldalsohaveanimpacton theE2Etestingofservicesintheoperatornetwork,
  10. 10. ✱ SIMPLIFYING THE 5G ECOSYSTEM SIMPLIFYING THE 5G ECOSYSTEM ✱ 20 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 21 Figure2illustratestheevolutionofspectrum usageinanetwork,startingwithLTEdeployedon sub-1GHzand1–3GHzbands.First,NRisdeployed on3.5GHzand/ormmWandwithLTEbandsusing option3.Thenextstepistodeployoption2for specificusecasesinlocalareas–suchasforFWA andindustrialdeployments. ExpandingstandaloneNRcoverage andcapacity Whendeployingoption2forwide-areausecases likeMBB,itisimportanttoensurecontinuousNR coveragewithinthetargetedarea(initiallyurbanfor example).SpottyNRcoveragewouldresultinfrequent mobilityeventsbetweenNRandLTEforwide-area usecases,eventhoughintersystemmobilitybetween option2andLTE/EPCwillbewellsupported.For theseusecases,option2requiresasufficientlylow NRbandinrelationtothesitegrid.Inmanycases, thesitegridfora3.5GHzdeploymentwillgivegood DLcoveragebothoutdoorsandindoors,butnot enoughULcoverage.NRon3.5GHzshould thereforetypicallybecombinedwithNRonlow bandtoprovidecontinuouscoverageinboththeUL andDL[3].ThelowNRbandcanbenew,refarmed oranexistingLTEbandthatissharedbetweenNR andLTE.Withrefarmingorsharing,akeyenableris thatthespectrumlicenseallowsNRdeployment (seefactboxonpage4,spectrumregulation). Tosupportoption2forMBBinanarea,itisalso advisabletodeployNRinoneormorelegacyLTE bandsusingLTE-NRspectrumsharing(seefactbox onpage4,LTE-NRspectrumsharing).Together withNRonlowandmid/highbands,thismaximizes thethroughputviaNRcarrieraggregation(CA). ThisisessentialtoprovidegoodMBBperformance, especiallyinareaswithoutDLcoveragefromnew NRbands.WhileNRdeploymentislimited,mobility tooption2shouldonlybetriggeredwhentheUE includingbothexistingserviceslikevoiceaswellas newones.Further,thehigherthenumberofoptions deployed,themorecomplexandtimeconsumingit willbefortheoperatorcommunitytoestablish5G roamingintheindustry. Networkdeploymentsbasedonoptions3and2 Option3isthebestshort-termalternativefor5G deployment,asitreliesonexistingLTE/EPC(option 1).Option3willprovidegoodperformancein severalaspects,allowingoptimizedtransmissionon NRwhenNRcoverageisgood,extendingNR downlink(DL)usageonahigherbandbycombining withalower-bandLTEforuplink(UL)data,and,if needed,aggregatingthroughputoverbothNRand LTEspectrum.Italsoprovidesreliableandsmooth mobilitybasedonanchoringinLTE/EPC,evenif theNRcoverageisspotty.Theuseofdual connectivityhas,however,introducedsomechallenges ontheUEsidewithdualtransmitters,which,insome cases,willlimitperformanceandcoverage. Oneofthemaindriversforgoingbeyond option3istoprovide5GC-enabledcapabilities likeenhancednetworkslicing,edgecomputing supportandoperationalbenefits,eventhough EPCcanalsosupporttheseservicestosome extent(slicingbasedonDECOR,forexample). Anothermaindriverforgoingbeyondoption3 istobeabletodeploystandaloneNRandgetthe radioperformancebenefitsofanNR-onlybased radiointerface. Option2(standaloneNR)isthefirst 5GC-basedoptionavailableinUEsandnetworks. EvenifgeneralNRcoverageislimited,option2 caninitiallybedeployedforspecificusecasesin localareas,wheredevicesstaywithingoodNR coverageonamidorhighband.Examplesinclude industrialdeploymentswithultra-reliablelow latencycommunicationrequirements,andfixed wirelessaccess(FWA),evenifthelatterisalso wellservedviaoption3. Figure 2 Spectrum migration steps for the 5G network Add option 2 5GC NR SA for local use cases in mid/high bands Option 2 on wide area NR in new or existing low bands with spectrum sharing/ refarming NR-NR CA Extend wide NR on additional bands Baseline: option 1 LTE-LTE CA LTE NR LTE+NR Add option 3 New NR spectrum on mid or high bands LTE-NR DC (EN-DC) High bands (24GHz–40GHz) Mid bands (3.5GHz–8GHz) Mid bands (1GHz–2.6GHz) Low bands (sub–1GHz) WITHREFARMINGOR SHARING,AKEYENABLERIS THATTHESPECTRUMLICENSE ALLOWSNRDEPLOYMENT Key enablers ❭❭ LTE-NR spectrum sharing 3GPP specifications allow efficient sharing of operator spectrum, so that one carrier appears as an NR carrier to NR UEs, and an LTE carrier to LTE UEs. Resources are pooled and distributed dynamically between the two RATs, according to instant needs. There is no impact on legacy LTE UEs, and the impact on LTE capacity is very small. Compared with classic refarming, this provides a smooth migration of spectrum from LTE to NR as NR-capable UE penetration increases, enabling NR to be rolled out on new and legacy bands. ❭❭ Spectrum regulation Spectrum is becoming technology neutral in most of the world except for a few markets and frequency bands where the spectrum license is currently tied to a specific RAT, prohibiting NR to operate in existing frequency bands.ItisimportantthatregulatorsacknowledgetheneedforNRdeploymentinallbands.Thisisakeyenabler formigrationtowideareacoverageofserviceslikeMBB/voiceandcMTCover5G,dependingonthepossibility to deploy NR in lower frequency bands. ❭❭ Dual-mode core network 4G devices will be the major device type and traffic consumer for a long time [2].Inaddition,operatorsare introducingnew5GdevicesdependingonbothEPC(option3)and5GC(option2). A “dual-mode” core network with both EPC and 5GC functionality will support the evolving device fleet in the networkandenableasmooth networktransformation.Toensureservicecoverageduringthemigrationperiod,thedual-modecorenetwork willprovidetightinterworkingbetweenEPCand5GCforseamless4G-5Gmobility.
  11. 11. ✱ SIMPLIFYING THE 5G ECOSYSTEM SIMPLIFYING THE 5G ECOSYSTEM ✱ 22 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 23 areas,NR(inorange)isdeployedon3.5GHzor mmWtoaddcapacitytothenetwork.NRisalso deployed,onasub-1GHzband(tocomplementUL coverage),andinlegacyLTEbandswithLTE-NR spectrumsharing. ThehorizontalblacklinesinFigure3represent thecoverageofoptions1,2and3.Option1isused inalargepartofthenetworktosupportMBBand actasthemainsolutionforMassiveMachineType Communication(mMTC)–specifically,Narrowband InternetofThings(NB-IoT)andLTE-MTC standard(LTE-M).Option3canbeusedanywhere thereisNRcoverage.Earlyoption2deploymentsin localareasincludeFWAandindustrialdeployments. Option2forgeneralMBBissupportedwhere thereislow-bandNRandsufficientNRbandwidth (mid/highbandand/oron1-3GHz). TheorangearrowsinFigure3indicatethatareas ofgoodNRcoverageareexpandedgeographically, coveringmoreurbanareas,andintimealso extendingintosuburbanareasandbeyond. Withthesupportofoptions1,2and3,keyusecases suchasmMTC,MBBandindustrialcritical-MTC (cMTC)willbesupportedinthenear-and mid-termwithgoodperformance. Targetarchitecture Theindustryhasspecifiedanewradioaccess technology–NR–andanewcorenetwork– 5GC–asthefoundationfortheevolutionof3GPP networks,which,inourview,makesoption2the long-termtargetarchitecturefortheindustry.Inthe long-termtargetnetwork,option2isdeployedwith widecoverage,usedbroadlyinmostdevices,and shouldbethebasisforfutureinvestmentsand featuregrowth. Figure4illustratesthemigrationstepsto the5Gtargetarchitectureforthemobileindustry, recognizingoption2asthelong-termtarget.The firststepistoaddoption3,followedbyoption2in selectedareas.Bygraduallyexpandingtheareas whereoption2isdeployed,theoperatorandthe hasenoughcoverageofsufficientNRspectrum, whichcanbehandledwiththresholdsandoffsets. ThepossibilityofaggregatingbandsusingCA,with asingleULtransmitterintheUE,isanimportant benefitofoption2,comparedwiththedual connectivityusedinoptions3,4and7.Thethird stepofFigure2showstheuseofNRonmultiple legacyLTEbandsusingLTE-NRspectrumsharing. Options1and3providegoodsupportfor smartphonesandMBB.MovingMBBtrafficto option2requiressupportforvoicetelephony. ThismeansthatNRmustbeabletosupportvoice natively,aswellassupportingseamlessmobilityvia handovertoLTE/EPCwhenleavingtheoption2 coveragearea.AsanintermediatestepbeforeNR supports(andisdimensionedfor)voice,thevoice servicecanrelyonEPSfallbacktoLTE/EPC. Tight5GC-EPCinterworkingisneededforboth voicesolutions,andthiswillalsoprovidegood intersystemmobilityforotherservices(seefactbox onpage4,dual-modecorenetwork). WhenaUEleavesanareawheretheNRcoverage isnotgoodenoughforoption2,thenetworkcan triggerintersystemmobilitytoEPC,eithertooption 3or1.Thesupportofoption2canthusbeextended graduallyinever-largerareasinanoperator’snetwork, startingwithdenseurbanareas.Bydeploying option2inhigh-trafficareasfirst,asignificant amountoftrafficcanbemigratedfromEPCto 5GC,evenifthegeographiccoverageisinitially morelimited. ManyLTEsiteswillbemodernizedwithmore advancedradiosforimprovedperformance(suchas 4T4R)orbyaddingmodernbasebandhardware, andwillthentypicallybepreparedtosupportNR ontheLTEbands.Thedeploymentofoption2 inaRANcapableofoption3isthendonewitha softwareupgrade.ThesamegNBwillservesome UEsinthesameNRcellwithoption3andothers withoption2. Figure3illustratesnetworkdeploymentduring themigrationfromLTEtoNR.Inselectedurban Figure 3 Network deployment during migration, including supported use cases 3.5GHz/mmW Denseurban Urban Suburban Rural 2 2 2 2 1 3 3 3 2 1-3GHz 1GHz mIoT 1 2 SpectrumUsecases 3 2 3 1 3 2 LTE NR NRLTEshared MBB mMTC FWA Local-areacMTC mIoT Figure 4 Migration steps toward target architecture Standalone NR and 5GC with appropriate features and coverage for addressed use cases. Full 5GC and NR coverage NR+5GC mainstream LTE/EPC for legacy devices EPC eNBeNB eNB gNB eNB gNB eNB gNBeNBeNB eNB gNB NR NRNRNR-low NRNR LTE LTE LTE LTE/NR LTE/NRLTELTE Option 1 Options 1, 3 Current industry focus Target architecture Options 1, 2, 3 Option 2 (1,3) 1 3 2 231 EPC+ 5GCEPC+ EPC+ 5GC    
  12. 12. ✱ SIMPLIFYING THE 5G ECOSYSTEM SIMPLIFYING THE 5G ECOSYSTEM ✱ 24 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 25 Insummary,option5isunlikelytoprovideafaster routeto5GCwide-areacoveragethanthewide deploymentofoption2.Widedeploymentofoption2 isthebetteralternative,particularlysinceoption5 wouldmeaninvestingintechnologythatdoesnot capitalizeonthebenefitsofthelatestradio technology(NR). Option 7 Option7buildsonoption5andcannotexistwithout it.Ifoption5weretobeused,itisverylikelythat option7wouldalsobesupportedinareaswithNR. Thedriverforoption7isthesameasforoption3; thatis,tousedualconnectivitytoaggregateNRand LTEbandstoenhancecapacity,butinthiscasefora UEconnectedviaeLTEto5GC.Accordingtothe samelogicexplainedintheOption5sectionabove, werecommendusingoption2instead. Option4 Option4isanadditiontooption2,usingdual connectivitytoaddeLTEtoanNRanchor.Itis primarilyrelevantwhenservingMBBtrafficvia 5GC.Thedriverforoption4istomaximize throughputwhentheamountofNRspectrumis limited.Anexampleofthistypeofsituationwould beifNRisdeployedon700MHz,3.5GHzandmmW, buttheUEisoutsidecoverageofthetwohigherbands. Intermsofdrawbacks,option4wouldrequire newsoftwaresupportineNB,gNBandUE,with relatedinteroperabilitytesting.Further,thefuture evolutionoffeatureswouldneedtoconsideroption4, anditsusewouldrequirecontinuedinvestmentsin eLTEdeploymentsforalongtime. Option4isnotnecessary,andperformsworse thanoption2withNR-NRCAandenoughNR spectrum, inareasservingMBBvia5GC.Using option2insteadofoption4alsofocusesinvestments ontherolloutofthelong-termtargetarchitecture. Conclusion Ouranalysisshowsthatthemobileindustryhasan opportunitytosimplifythe5Gecosystemby focusingnetworkdeploymentsonconnectivity options3and2,whicharecapableofdelivering allthe5Gbenefitswithoutaddingunnecessary complexityandcost(asinoptions5,7and4). Theflexibledesignofradioandcorenetworks supportsasmoothmigrationwithLTE-NR spectrumsharinganddual-modecoretechnologies. Theregulationoffrequencybandsshouldallow NRdeploymentinexistingLTEbandsthatare insyncwiththerequiredspectrummigration. Operatorshavetheopportunitytoavoid connectivityoptions5,7and4byimplementinga proactivespectrummigrationstrategythat considersNRfornewlowbands,andbyrefarming orintroducingLTE-NRspectrumsharingin existinglow/midbands.Thisapproachwill reducenetworkupgradecostandtime,simplify interoperabilitybetweennetworksanddevices, andenableafasterscalingofthe5Gecosystem. A5Gdeploymentapproachbasedexclusivelyon options3and2ensuresthatinvestmentisfocused onthelong-termtargetarchitecture,leveragingfull 5GScapabilities.Earlykeyusecasesforwide-area, likeMBBincludingvoiceservices,arefullysupported duringthemigrationperiod,alongwithservices toexistingdevices. industrywillalwaysinvestinstepsleadingtothe long-termtargetarchitecture.Eventually,the option2coveragewillbesufficienttoalsosupport wide-areacMTCusecasesthatwillbenefitfrom bothNRand5GC. Atsomepointinthefuturetherewillalsobe mMTCsolutionsbasedonNR/5GC.However,many mMTCservicesarealreadyadequatelyservedbythe existingmMTCsolutionsNB-IoTandLTE-M.The mMTCservicesinthelow-endLowPowerWide Area(LPWA)segmentarejustoneexample.Toavoid fragmentation,thebestalternativefortheseusecases iscontinueduseofNB-IoTandLTE-Mforalongtime. Thetimingtoreachthislong-termtargetmayvary betweenmarkets.Itshouldbenotedthatevenwhen thetargetisreached,networkswillneedtocontinue tosupportasetoflegacydevices(LTE/EPC-based), inparticularintheareaofmMTC.WhentheUE penetrationforNRsupportishighenough,selected bandscanbefullyrefarmedtoNR-only,asshownin thelaststepofFigure2. Analysisofoptions5,7and4 Theindustryhasdecidedtobasetheinitial deploymentof5Gonoptions3and2.Whileoptions 5,7and4mayinitiallyseembeneficialforspecific operators’deploymentcases,itisimportantto recognizethatnoneofthemaredirectstepsleading towardthelong-termtargetarchitecture.Further, theuseofoptions5,7and4wouldaddunnecessary complexityinthetargetarchitecture,inthe interactionwithothernetworkfunctions,andinthe evolutionofnewfeatures,whichneedstotakethe combinationofallexistingoptionsintoaccount. Afterathoroughanalysisofoptions5,7and4 thatencompassedthemaindrivers,potential benefitsanddrawbacks,wehavecometothe conclusionthatallthreecanandshouldbeavoided. Wehavealsoidentifiedpreferredalternative solutionsforeachoption. Option5 Themaindriverfordeployingoption5istoallow devicesthatmoveoutsidetheareacoveredbyoption 2toremainconnectedto5GC,whichwouldalso increasethe5GCcoveragetoeLTEareas. Akeyquestiontoconsideris:whichusecases requirenationwide5GCcoverage?Traditional MBB/voiceobviouslyrequireswide-areasupport, butthisiswellsupportedwithintersystemmobility duringthebuild-outofNRcoverage,asitwasin previousgenerationshifts.5GCprovidesarange ofnewvaluesbuttheneedforotherwide-area 5GC-basedservicesintheneartermisundefined. Inthelongerterm,weexpectwide-areaoption2 toenablethenewusecasesthatemerge. Option5couldbeusedtoincreasewide-area 5GCcoverage,butreachingfullwide-area5GC coveragewouldtaketimeandinvestment,asitwould requirenewUEs,newRANfunctionalityand retestingthesystem.Option5wouldhaveamajor impactontheUEsintermsofsupportingthe5GC non-accessstratumandthenewpartsoftheeLTE radiointerface,aslegacyLTEdevicesarenot supported.Inaddition,substantialinteroperability retestingbetweennetworksandUEswouldbe requiredtoensuretheoperationoflegacyfeatures andservices,includingVoLTE.Further,option5 requiressubstantialupgradesoftheeNBsoftware and,inmanycases,theeNBbasebandhardware aswell. THEINDUSTRYHAS DECIDEDTOBASETHEINITIAL DEPLOYMENTOF5GON OPTIONS3AND2 THEMOBILEINDUSTRY HASANOPPORTUNITY TOSIMPLIFYTHE5G ECOSYSTEM
  13. 13. ✱ SIMPLIFYING THE 5G ECOSYSTEM SIMPLIFYING THE 5G ECOSYSTEM ✱ 26 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 27 Further reading ❭❭ Ericsson, 5G deployment options, 2018, available at: https://www.ericsson.com/assets/local/narratives/ networks/documents/5g-deployment-considerations.pdf ❭❭ Ericsson, Core evolution from EPC to 5G Core, download available from: https://pages.digitalservices. ericsson.com/core-evolution-to-5g References 1. 3GPP Release 15 specifications, e.g. TS 23.501, TS 38.401, available at: http://www.3gpp.org/release-15 2. Ericsson Mobility Report, available at: https://www.ericsson.com/en/mobility-report 3. Ericsson Technology Review, November 2018, The advantages of combining 5G NR with LTE, available at:https://www.ericsson.com/en/ericsson-technology-review/archive/2018/the-advantages-of-combining-5g-nr- with-lte theauthors Torbjörn Cagenius ◆ is a senior expert in network architecture at Business Area Digital Services. He joined Ericsson in 1990 and has worked in a variety of technology areas such as fiber-to-the- home, main-remote RBS, fixed-mobile convergence, IPTV, network architecture evolution, software-defined networking and Network Functions Virtualization. In his current role, he focuses on 5G and associated network architecture evolution. He holds an M.Sc. in electrical engineering from KTH Royal Institute of Technology in Stockholm, Sweden. Anders Ryde ◆ is a senior expert in network and service architecture at Business Area Digital Services, based in Sweden. He joined Ericsson in 1982 and has worked in a variety of technology areas in network and service architecture developmentformultimedia- enabled telecommunication, targeting both enterprise and residential users. This includes the evolution of mobile telephony to IMS and VoLTE. In his current role, he focuses on bringing voice and other communication services into 5G, general 5G evolution and associated network architecture evolution. He holds an M.Sc. in electrical engineering from KTH Royal Institute of Technology in Stockholm, Sweden. Jari Vikberg ◆ is a senior expert in network architecture and the chief network architect at CTO office. He joined Ericsson in 1993 and has both wide and deep technology competence covering network architectures for all generations of RANs and CNs. He is also skilled in the application layer and other domains, and the impact and relation these have to mobile networks. He holds an M.Sc. in computer science from the University of Helsinki, Finland. Per Willars ◆ is an expert in network architecture and radio network functionality at Business Area Networks. He joined Ericsson in 1991 and has worked intensively with RAN issues ever since. This includes leading the definition of 3G RAN, before and within the 3GPP, and more lately indoor solutions. He has also worked with service layer research and explored new business models. In his current role, he analyzes the requirements for 5G RAN (architecture and functionality) with the aim of simplifying 5G. He holds an M.Sc. in electrical engineering from KTH Royal Institute of Technology. Terms and abbreviations 4T4R – 4-Branch Transmit/Receive Antenna and Radio Arrangement | 5GC – 5G Core | 5GS – 5G System | CA – Carrier Aggregation | cMTC – Critical Machine Type Communication | CN – Core Network | DC – Dual Connectivity | DECOR – Dedicated Core Network | DL – Downlink | E2E – End-to-end | eLTE – Evolved LTE | eNB – Evolved Node B | EN-DC – E-UTRA – NR Dual Connectivity | EPC – Evolved Packet Core | FWA – Fixed Wireless Access | gNB – Next Generation Node B | IoT – Internet of Things | LPWA – Low Power Wide Area | LTE-M – LTE-MTC Standard | MBB – Mobile Broadband | mMTC – Massive Machine Type Communication | mmW – Millimeter Wave | NB-IoT – Narrowband Internet of Things | NR – New Radio | RAT – Radio Access Technology | UE – User Equipment | UL – Uplink
  14. 14. ✱ DISTRIBUTED CLOUD DISTRIBUTED CLOUD ✱ 28 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 29 ofalargeamountofdatabetweenvehiclesandthe cloud,oftenwithreal-timecharacteristicswithin alimitedtimeframewhilethevehicleisinactive operation. Highdatavolume Lookingattheautomotiveindustry,weoftenfocus onthereal-timeusecasesforsafety,asdefinedby V2X/C-ITS(vehicletoeverything/cooperative intelligenttransportsystem),wherereal-time aspectssuchasshortlatencyarethemostsignificant requirements.However,theautomotiveindustry’s newmobilityservicesalsoplacehighdemandson networkcapacityduetotheextremeamountofdata thatmustbetransportedtoandfromhighlymobile devices,oftenwithnear-real-timecharacteristics. Dataneedstobetransportedwithinalimitedtime window(~30min/day),withavaryinggeographical concentrationofvehiclesusingamultitudeof differentnetworktechnologiesandconditions. Themarketforecaststhataregenerallyreferred toindicatethattheglobalnumberofconnected vehicleswillgrowtoapproximately700millionby 2025andthatthedatavolumetransmittedbetween Emerging use cases in the automotive industry – as well as in manufacturing industries where the first phases of the fourth industrial revolution are taking place – have created a variety of new requirements for networks and clouds. At Ericsson, we believe that distributed cloud is a key technology to support such use cases. CHRISTER BOBERG, MALGORZATA SVENSSON, BENEDEK KOVÁCS vehiclesandthecloudwillbearound100petabytes permonth.AtEricsson,however,weanticipatethat theautomotiveservicesofthenearfuturewillbe muchmoredemanding.Weestimatethatthedata trafficcouldreach10exabytesormorepermonthby 2025,whichisapproximately10,000timeslargerthan thepresentvolume.Gartnerrecentlyraisedthe expectationsfurtherinitslatestreport(June2018), estimatingthevolumetobeashighasoneterabyte permonthpervehicle[1]. Suchmassiveamountsofdatawillplacenew demandsontheradionetwork,asthemainpartis ULdata.Newbusinessmodelswillberequired,asa resultofthehighcostofhandlingmassiveamounts ofdata.AsexplainedintheAECC(AutomotiveEdge ComputingConsortium)whitepaper[2],thecurrent mobilecommunicationnetworkarchitecturesand conventionalcloudcomputingsystemsarenotfully optimizedtohandleallofthisdataeffectivelyona globalscale.Thewhitepapersuggestsmanypossible optimizationstoconsider–basedontheassumption thatmuchofthedatacouldbeanalyzedandfiltered atanearlystagetolimittheamountofdata transferred. Both 4G and 5G mobile networks are designed to enable the fourth industrial revolution by providing high bandwidth and low-latency communication on the radio interface for both downlink (DL) and uplink (UL) data. Distributed cloud exploits these features, enabling a distributed execution environment for applications to ensure performance, short latency, high reliability and data locality. ■ Distributedcloudmaintainstheflexibilityof cloudcomputingwhileatthesametimehidingthe complexityoftheinfrastructure,withapplication componentsplacedinanoptimallocationthat utilizesthekeycharacteristicsofdistributedcloud. Theautomotivesectorandmanymanufacturing industriesalreadyhaveusecasesthatmakethem verylikelytobeearlyadoptersofdistributed cloudtechnology. Next-generationautomotiveservices andtheirrequirements Mobilecommunicationinvehiclesisincreasing inimportanceastheautomotiveindustryworks tomakedrivingsafer,smooththeflowoftraffic, consumeenergymoreefficientlyandlower emissions.Automatedandintelligentdriving, thecreationanddistributionofadvancedmaps withreal-timedata,andadvanceddrivingassistance usingcloud-basedanalyticsofULvideostreams areallexamplesofemergingservicesthatrequire vehiclestobeconnectedtothecloud.Theseservices alsorequirenetworksthatcanfacilitatethetransfer A KEY ENABLER OF AUTOMOTIVE AND INDUSTRY 4.0 USE CASES Distributed cloud Definition of key terms ❭❭ Distributed cloud is a cloud execution environment for applications that is distributed across multiple sites, including the required connectivity between them, which is managed as one solution and perceived as such by the applications. ❭❭ Edge computing refers to the possibility of providing execution resources (compute and storage) with the adequate connectivity (networking) at close proximity to the data sources. ❭❭ The fourth industrial revolution is considered to be the fourth big step in industry modernization, enabled by cyber-physical systems, digitalization and ubiquitous connectivity provided by 5G and Internet of Things (IoT) technologies. It is also referred to as Industry 4.0.
  15. 15. ✱ DISTRIBUTED CLOUD DISTRIBUTED CLOUD ✱ 30 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 31 datalocally.Thisreducesthetotalamountofdata exchangedbetweenvehiclesandcloudswhile enablingtheconnectedvehiclestoobtainfaster responses.Theconceptischaracterizedbythree keyaspects:alocalizednetwork,edgecomputing anddataexposure. Alocalizednetworkisalocalnetworkthatcovers alimitednumberofconnectedvehiclesinacertain area.Thissplitsthehugeamountofdatatrafficinto reasonablevolumesperareaofdatatrafficbetween vehiclesandtheclouds. Edgecomputingreferstothegeographical distributionofcomputationresourceswithinthe vicinityoftheterminationofthelocalizednetworks. Thisreducestheconcentrationofcomputationand shortenstheprocessingtimeneededtoconclude atransactionwithaconnectedvehicle. Dataexposuresecuresintegrationofthedata producedlocallybyutilizingthecombinationofthe localizednetworkandthedistributedcomputation. Bynarrowingrelevantinformationdowntoa specificarea,datacanberapidlyprocessedto integrateinformationandnotifyconnectedvehicles inrealtime.Theamountofdatathatneedstobe exchangediskepttoaminimum. Privateandlocalconnectivity Aspartofthefourthindustrialrevolution,industry verticalsandcommunicationserviceproviders (CSPs)aredefiningasetofnewusecasesfor5G[3]. Privatedeploymentsand5Gnetworksprovidedby CSPstomanufacturingcompanies,smartcitiesand otherdigitalindustriesareonthehorizonaswell. However,therearetwomainchallengestomobile networkoperators’abilitytodeliver.Thefirstisthe toughlatency,reliabilityandsecurityrequirements ofthesenewusecases.Thesecondisfiguringout howtoshieldtheindustriesfromthecomplexity oftheinfrastructure,toenableeaseofusewhen programmingandoperatingnetworks. Secureprivatenetworkswith centralizedoperations Securityanddataprivacyarekeyrequirements forindustrialnetworks.Insomecases,regulations orcompanypoliciesstipulatethatthedatamust notleavetheenterprisepremises.Inothercases, someorallofthedatamustbeavailableatremote locationsforpurposessuchasproductionanalytics oremergencyprocedures.Atypicalindustrial environmenthasmultipleapplicationsdeployedand operatedbydifferentthirdparties.Whatthismeans inpracticeisthatthesameon-premises,cloud-edge instancethatafactoryalreadyusesforbusiness supportandITsystemswouldalsoneedtosupport theconnectivityforitsrobotstointeractwitheach other.Asaresult,thereisarequirementofmulti- tenancyforboththedevicesandtheinfrastructure. Tactileinternetandaugmentedreality Augmentedreality(AR)andmachinelearning(ML) technologiesarewidelyrecognizedasthemain pillarsofthedigitalizationofindustries[4],and researchsuggeststhatwidedeploymentof interactivemediaapplicationswillhappenon5G networks.Manyobserversenvisiontheworker oftomorrowassomeonewhoisequippedwith eye-trackingsmartglasses[5]andtactilegloves ratherthanscrewdriversets[6].Human-to-machine applicationsrequirelowlatencywhiledemanding highnetworkbandwidthandheavycompute resources.Runningthemonthedeviceitself wouldresultinhighbatteryconsumptionandheat dissipation.Atthesametime,latencyrequirements donotallowtherunningofthecompleteapplication inlargecentraldatabasesduetothephysicallimits oflightspeedinopticalfibers. Topology-awarecloudcomputingandstorageis anexampleofonesuchsolutionthatprovideswhat wecallaglobalautomotivedistributededgecloud. Thelimitationontheamountofdatathatcanbe effectivelytransportedoverthecellularnetwork mustnotbeallowedtoaffecttheserviceexperience negatively,asthatwouldhindertheevolutionofnew automotiveservices.Itisthereforenecessaryto increasecapacity,availabilityandcoverageaswellas findingappropriatemechanismstolimittheamount ofdatatransferred.Orchestratingapplicationsand theirdifferentcomponentsrunninginamultitudeof differentcloudsfromdifferentvendorsisoneofthe challenges.Vehiclesconnectingtonetworkswithout anexistingapplicationedgeinfrastructureis another. Theplacementofapplicationcomponentsat edgesdependsonthebehavioroftheapplication andtheavailableinfrastructureresources. Whendealingwithhighlymobiledevicesthat connecttoamultitudeofnetworks,itmustbe possibletomoveexecutionoftheedgeapplication automaticallywhenamoreappropriatelocation forthevehicleisdiscovered.Someapplications requiretransferofpreviouslyanalyzeddataand findingstothenewlocation,whereanewapplication componentinstancewillseamlesslytakeovertoserve themovingvehicle. Distributedcomputingonalocalizednetwork Wehavedevelopedtheconceptofdistributed computingonalocalizednetworktosolvethe problemsofdataprocessingandtrafficinexisting mobileandcloudsystems.Inthisconcept,several localizednetworksaccommodatetheconnectivity ofvehiclesintheirrespectiveareasofcoverage. AsshowninFigure1,computationpowerisadded totheselocalizednetworks,sothattheycanprocess Figure 1 High-volume data automotive services and their characteristics Local Regional Regional DCLocal DC MTSO MTSO MTSO H National DC National sitesLocal and regional sites Service exposure HD maps HD maps Data exposure for automotive services Access sites Hub sites Video stream ECU sensors HD maps Video stream ECU sensors HD maps Mobile telephone switching office Intelligent driving Intelligent driving Advanced driver assistance Advanced driver assistance Huge amount of data INDUSTRYVERTICALS ANDCOMMUNICATION SERVICEPROVIDERSARE DEFININGASETOFNEW USECASESFOR5G
  16. 16. ✱ DISTRIBUTED CLOUD DISTRIBUTED CLOUD ✱ 32 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 33 models.Oneexampleofapossiblescenarioisfora CSPtoofferconnectivityandacloudexecution environmenttoenterprisesasaservice.Inthiscase, aCSPmanagesthecomputationandconnectivity resources,butthesearelocatedattheenterprise premises.Theapplicationcharacteristicsdetermine theplacementofapplicationsatvariousgeolocations. InthecaseofAR/VRandimagerecognition applicationsusedbytechnicianstofixabroken powerstation,forexample,itwouldbemosteffective toplacethemclosetothebrokenpowerstation. Edgecomputing Ourdistributedcloudsolutionenablesedge computing,whichmanyapplicationsrequire. Wedefineedgecomputingastheabilitytoprovide executionresources(specificallycomputeand storage)withadequateconnectivityatclose proximitytothedatasources. Intheautomotiveusecase,thenetworkis designedtosplitdatatrafficintoseverallocations thatcoverreasonablenumbersofconnected vehicles.Thecomputationresourcesare hierarchicallydistributedandlayeredinatopology- awarefashiontoaccommodatelocalizeddataandto allowlargevolumesofdatatobeprocessedina timelymanner.Inthisinfrastructureframework, localizeddatacollectedvialocalandwidearea networksisstoredinthecentralcloudandintegrated AsimpleARapplicationanditsmaincomponents areshowninFigure2.Thecomponentsofthe applicationcouldbeexecutedeitheronthedevice itself,theedgeserverorinthecentralcloud. Deployingapplicationcomponentsatthenetwork edgemaymakeitpossibletooffloadthedevicewhile maintainingshortlatency.Edgecomputeisalso optimizingtheflowwhencoordinationisrequired– forexample,whenusingmultiplereal-timecamera feedstodeterminethe3Dpositionofobjects,also asshowninFigure2.Furthermore,advancedcloud softwareasaservice–ML,analyticsandDBsasa service,forexample–mayalsobeprovidedonthe edgesite. Ourdistributedcloudsolution Ericssonhasdevelopedadistributedcloudsolution thatprovidestherequiredcapabilitiestosupport theusecasesofthefourthindustrialrevolution, includingprivateandlocalizednetworks.Our solutionsatisfiesthespecificsecurityrequirements neededtodigitalizeindustrialoperations,with automotivebeingoneofthekeyusecases.Ericsson’s distributedcloudsolutionprovidesedgecomputing andmeetsend-to-endnetworkrequirementsaswell asofferingmanagement,orchestrationandexposure forthenetworkandcloudresourcestogether. AsshowninFigure3,wedefinethedistributed cloudasacloudexecutionenvironmentthatis geographicallydistributedacrossmultiplesites, includingtherequiredconnectivityinbetween, managedasoneentityandperceivedassuchby applications.Thekeycharacteristicofour distributedcloudisabstractionofcloud infrastructureresources,wherethecomplexityof resourceallocationishiddentoauserorapplication. Ourdistributedcloudsolutionisbasedonsoftware- definednetworking,NetworkFunctions Virtualization(NFV)and3GPPedgecomputing technologiestoenablemulti-accessandmulti-cloud capabilitiesandunlocknetworkstoprovideanopen platformforapplicationinnovations.Inthe managementdimension,distributedcloudoffers automateddeploymentinheterogeneousclouds. ThiscouldbeprovidedbymultipleCSPs,where workloadplacementispolicydrivenandbased onvariousexternalizedcriteria. Toenablemonetizationandapplicationinnovation, distributedcloudcapabilitiesareexposedon marketplacesprovidedbyEricsson,thirdparties andCSPs.Thedistributedcloudcapabilitiescanbe offeredaccordingtovariousbusinessandoperational Figure 3 Distributed cloud architecture Service and resource orchestration Any workload Access sites Local and regional DC sites National sites Anywhere in the network End-to-end orchestration Marketplace Service exposure Global clouds Public safety Automotive FWA Factory Video streaming Metering APP APP VNF VNF APP APP APP VNF VNF VNF VNF VNFVNF Figure 2 An AR application and its modules optimized for edge computing Capturing Preprocessing Object detection feature extraction Recognition database match DB Display Tracking and annotation Position estimation Template matching IoT device/user equipment -20ms BW reduction -20ms/frame Computation heavy -20ms Computation heavy Multiple device data aggregation -100ms Requires access to central storage Edge site National site OURDISTRIBUTED CLOUDSOLUTIONENABLES EDGECOMPUTING,WHICH MANYAPPLICATIONS REQUIRE
  17. 17. ✱ DISTRIBUTED CLOUD DISTRIBUTED CLOUD ✱ 34 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 35 Thesecondstepwillbetogainindustryacceptance forthemechanisms,beforefinallybeingableto implementthesolutionsandestablishthebusiness models. OnewaytoevolvethecloudedgesthatCSPs currentlysupplyistoprovideanenvironmentabove thecurrentinfrastructurethatishomogeneousfrom aconsumptionperspectivebutdiscoverablethrough APIsandorchestratedinthesamewayastheCSPs’ infrastructure.Thiswouldprovideanintermediate step,whereCSPswithoutanedgecloud infrastructurecouldbecomeapartoftheglobal scaledistributedcloud.Followingthisapproach, anindustryactorcouldconnecttoanyCSPaccess networkasopposedtobeinglimitedtocertainCSPs. Whilethesenetworkswillhavethesamefunctional scope,theywillnotbeabletoprovidefulledge characteristics.Thiswillalsoserveasacatalystfor otherCSPstojointheglobalscaledistributedcloud. Otherwise,theywillnotbecomepreferredsuppliers. Embracingindustryinitiatives andstandardizations Webelievethattheevolutiontowardtheglobal multi-operatordistributedcloudisdependentona fewkeyactions.First,wemusttakeactiontoaddress thefactthatthecurrentmobilecommunication networkarchitecturesandconventionalcloud computingsystemsarenotdesigned,orchestrated orexposedinawaythatcanhandletheindustries’ requirementseffectively.Wemustscrutinizethe systemarchitecturesandinvestigatenetwork deploymentsandpreferredprofilingtobetter accommodatetheoutlinedrequirements.The architectureevolutionwillbedrivenbytherelevant standardizationssuchas3GPPandETSI(European TelecommunicationsStandardsInstitute)NFV. Secondly,webelievethatitiscriticaltodrive industryalignmentbygettingreference implementationsofedgecloudsoftware.Thisiswhy Ericssonhasjoinedtheindustrycollaboration projectOPNFV(OpenPlatformforNFV)and ONAP(OpenNetworkAutomationPlatform)[7], whichprovidesthemanagementcapabilitiesof distributedcloud. Finally,webelievethatparticipatingin ecosystemsthatprovidetheopportunityfor interactionsbetweentheindustriesandvendorsis criticaltotheevolution.Thisisparticularlytruefor ecosystemsthatformulaterequirementsandways ofworking,defineusecases,agreeonacommon, easy-to-usereferenceimplementation,anddrive alignmentinstandardizationbodiesbasedonthose implementations.Examplesofsuchecosystemsare theAECCand5GAA(the5GAutomotive Association)forautomotiveand5G-ACIA(the5G AllianceforConnectedIndustriesandAutomation) [8],Industry4.0andtheIIoT(IndustrialInternetof Things)forthefourthindustrialrevolution. Conclusion Distributedcloudisacornerstoneoftheintelligent networksthatwillplayakeyenablingroleinthe fourthindustrialrevolution.Arobustdistributed cloudsolutionrequiresefficientandintelligent managementandorchestrationcapabilitiesthat spanheterogeneouscloudssuppliedbymultiple actors.Serviceexposurewillenablemonetization andapplicationinnovationthroughintegration withthemarketplacesand/orintegrationwiththe industries’ITsystems. Theevolutiontowardgloballydistributedcloud requiresactiontoaligntheindustryboththrough traditionalstandardizationsaswellasactive participationinopen-sourceprojectsaimedat providingreferenceimplementations.Ecosystems suchastheAECCplayanimportantroleby examiningthehigh-volumedatausecasesforthe automotiveindustry. ontheedgecomputingarchitecturetoprovide real-timeinformationnecessaryforservicesof connectedvehicles. Theexactlocationsofthemicroandsmalldata centersmaybedependentontheCSP’snetwork topologyandtherequirementsoftheusecases– thisappliestocentralofficesites,basestationsand newDCsbuiltonindustrialsites.Thisinfrastructure shouldbeflexible,sothatitispossibletostartwitha fewsitesandgrowbyaddingnewsitesasrequired. Managementandorchestration Thedistributedcloudreliesonefficientmanagement andorchestrationcapabilitiesthatenableautomated applicationdeploymentinheterogeneousclouds suppliedbymultipleactors.Figure3illustrateshow theserviceandresourceorchestrationspansacross distributedandtechnologicallyheterogeneous clouds.Itenablesservicecreationandinstantiation incloudenvironmentsprovidedbymultiplepartners andsuppliers.Discovery,onboardingandauto- enrollmentofedgesareotherimportantcapabilities ofdistributedcloudmanagement. Whendeployinganapplicationoravirtual networkfunction(VNF),theplacementdecisions canbebasedonmultiplecriteria,wherelatency, geolocation,throughputandcostareafewexamples. Thesecriteriacanbedefinedeitherbyan applicationdeveloperand/oradistributedcloud infrastructureprovider,servingasinputtothe placementalgorithm.Onceatargetcloudhasbeen selected,theworkloadplacementcontinuesinany ofthesubordinatedclouds. Intheautomotiveapplicationsexample,the placementdecisioncouldbemadebasedonthe geolocationofthemovingcar,availabilityofthe computationresourcesandabilitytomeetregulatory requirementsattheedgesservingthemovingcar. TactileinternetandARapplicationsthatarevery sensitivetonetworklatencywhiledemanding highbandwidthandhighcomputingpower willbedeployedattheedgesthatcanfulfillthe requirements. Theserviceorchestrationmanagesthe distributedcloudresourcesaswellastheefficient distributionandreplicationoftheapplicationsthat utilizethedistributedcloudcomputationand connectivityresources.Theserviceandresource managementcapabilitiesarealsodeployedina distributedfashiontoenableefficientmanagement. Forexample,thescalingordatafunctionswillbe deployedclosetotheapplicationtheysupervise. Serviceexposure Theapplicationsdeployedinthedistributedcloud willpresenttheircapabilitiesthroughtheservice exposure.Withmulti-dimensionalexposure,eachof thelayersinthedistributedcloudstackwillexpose itscapabilities.Thecloudinfrastructurelayerand theconnectivitylayerwillexposetheirrespective capabilitiesthroughtheapplicationprogramming interface(s)(API(s)),whichwillthenbeusedby applicationdevelopersoftheindustriesmakinguse ofthemobileconnectivity.Bysettingdeveloper needsinfocus,theexposedAPI(s)willbeabstracted sothattheyareeasytouse. Evolutiontowardtheglobalmulti-operator distributedcloud Globalindustriessuchasautomotiverequire solutionsthatworkseamlesslyfromlocaltoglobal scale.Inlightofthis,theevolutiontowardtheglobal multi-operatordistributedcloudisnotrivialmatter. Tobepartofthegloballydistributedcloud,the edgecloudsthatCSPsprovideataccessandlocal sitesmustsupportastringentsetoffunctionsand APIs.ThisimpliesthatCSPsmustjoinforcesto createafederatedmodel.Doingsowillrequire significanteffort,withthefirststepbeingtoreachan agreementonthestandardmechanismstouse. GLOBALINDUSTRIES SUCHASAUTOMOTIVE REQUIRESOLUTIONS THATWORKSEAMLESSLY FROMLOCALTOGLOBAL SCALE ITISCRITICALTODRIVE INDUSTRYALIGNMENTBY GETTINGREFERENCE IMPLEMENTATIONSOFEDGE CLOUDSOFTWARE
  18. 18. ✱ DISTRIBUTED CLOUD DISTRIBUTED CLOUD ✱ 36 #01 2019 ✱ ERICSSON TECHNOLOGY REVIEWERICSSON TECHNOLOGY REVIEW ✱ #01 2019 37 Further reading ❭❭ Ericsson Consumer IndustryLab, 5G business value: A case study on real-time control in manufacturing, April 2018, available at: https://www.ericsson.com/assets/local/reports/5g_for_industries_ report_blisk_27062018.pdf ❭❭ Ericsson, Turn on 5G: Ericsson completes 5G Platform for operators, February 8, 2018, available at: https://www.ericsson.com/en/press-releases/2018/2/turn-on-5g-ericsson-completes-5g-platform-for-operators ❭❭ Ericsson, Going beyond edge computing with distributed cloud, available at: https://www.ericsson.com/ digital-services/trending/distributed-cloud ❭❭ Ericsson/KTH Royal Institute of Technology, Resource monitoring in a Network Embedded Cloud: An extension to OSPF-TE, available at: https://www.ericsson.com/assets/local/publications/conference- papers/03-cloud-ospf-camera-ready.pdf ❭❭ M2 Optics Inc., Calculating Optical Fiber Latency, January 9, 2012, Miller, K, available at: http://www.m2optics.com/blog/bid/70587/Calculating-Optical-Fiber-Latency ❭❭ Application function placement optimization in a mobile distributed cloud environment, Anna Peale, Péter Kiss, Charles Ferrari, Benedek Kovács, László Szilágyi, Melinda Tóth, in Studia Informatica - Issue no. 2/2018, pp37-52, available at: http://www.studia.ubbcluj.ro/arhiva/abstract_en. php?editie=INFORMATICAnr=2an=2018id_art=15974 References 1. TelecomTV, Gartner says 5G networks have a paramount role in autonomous vehicle connectivity, June 21, 2018, available at: https://www.telecomtv.com/content/tracker/gartner-says-5g-networks-have-a- paramount-role-in-autonomous-vehicle-connectivity-31356/ 2. AECC White Paper, available at: https://www.ericsson.com/res/docs/2014/consumerlab/liberation-from- location-ericsson-consumerlab.pdf 3. Government Technology, Making 5G a Reality Means Building Partnerships — Not Just Networks, June 5, 2018, Descant, S, available at: http://www.govtech.com/network/Making-5G-a-Reality-Means- Building-Partnerships--Not-Just-Networks.html 4. Wired, Eye tracking is coming to VR sooner than you think. What now?, March 23, 2018, Rubin, P, available at: https://www.wired.com/story/eye-tracking-vr/ 5. Think Act, Digital factories – The renaissance of the U.S. automotive industry, Berger, R, available at: https://www.rolandberger.com/en/Publications/pub_digital_factories.html 6. Ericsson, Technology Trends 2018, Five technology trends augmenting the connected society, 2018, Ekudden, E, available at: https://www.ericsson.com/en/ericsson-technology-review/archive/2018/technology- trends-2018 7. Ericsson Technology Review, Open, intelligent and model-driven: evolving OSS, February 7, 2018, Agarwal, M; Svensson, M; Terrill, S; Wallin, J, available at: https://www.ericsson.com/en/ericsson-technology- review/archive/2018/open-intelligent-and-model-driven-evolving-oss 8. 5G-ACIA, 5G for Connected Industries and Automation, April 11, 2018, available at: https://www.5g-acia. org/publications/5g-for-connected-industries-and-automation-white-paper/ theauthors Malgorzata Svensson ◆ is an expert in operations support systems. She joined Ericsson in 1996 and has worked in various areas within research and development. For the past 10 years, her work has focused on architecture evolution. Svensson has broad experience in business process, function and information modeling, information and cloud technologies, analytics, DevOps processes and tool chains. She holds an M.Sc. in technology from the Silesian University of Technology in Gliwice, Poland. Christer Boberg ◆ serves as a director at Ericsson’s CTO office, responsible for IoT technology strategies aimed at solving networking challenges for the industry on a global scale. He initially joined Ericsson in 1983 and has in his career within and outside Ericsson focused on software and system design as a developer, architect and technical expert. In recent years, Boberg’s work has centered on IoT and cloud technologies with a special focus on the automotive industry. As part of this work, he drives the AECC consortium together with industry leading companies. Benedek Kovács ◆ joined Ericsson in 2005 as a software developer and tester, and later worked as a system engineer. He was the innovation manager of the Budapest RD site 2011-13, where his primary role was to establish an innovative organizational culture and launch internal startups based on worthy ideas. Kovács went on to serve as the characteristics, performance management and reliability specialist in the development of the 4G VoLTE solution. Today he is working on 5G networks and distributed cloud, as well as coordinating global engineering projects. He holds an M.Sc. in information engineering and Ph.D. in mathematics from the Budapest University of Technology and Economics in Hungary. Theauthorswould liketothank thefollowing peoplefortheir contributions tothisarticle: CarlosBravo, AlaNazari, StefanRuneson, OlaHubertsson, ThorstenLohmar andTomas Nylander. Terms and abbreviations AECC – Automotive Edge Computing Consortium | API – Application Programming Interface | APP – Application | AR – Augmented Reality | BW – Bandwidth | CSP – Communication Service Provider | DB – Database | DC – Data Center | ECU – Engine Control Unit |ETSI – European Telecommunications Standards Institute | FWA – Fixed Wireless Access | IoT – Internet of Things | ML – Machine Learning | MS – Millisecond | MTSO – Mobile Telephone Switching Office | NFV– Network Functions Virtualization | UL – Uplink | VNF – Virtual Network Function | VR – Virtual Reality | V2X/C-ITS – Vehicle-to-everything/ Cooperative Intelligent Transport System
  19. 19. #01 2019 ✱ ERICSSON TECHNOLOGY REVIEW 3938 ERICSSON TECHNOLOGY REVIEW ✱ #01 2019 Industry 4.0 – the fourth industrial revolution – is already transforming the manufacturing industry, with the vision of highly efficient, connected and flexible factories of the future quickly becoming a reality in many sectors. Fully connected factories will rely on cloud technologies, as well as connectivity based on Ethernet Time-Sensitive Networking (TSN) and wireless 5G radio. JOACHIM SACHS, KENNETH WALLSTEDT, FREDRIK ALRIKSSON, GÖRAN ENEROTH Automation(5G-ACIA)[2]showthatindustries recognizethisneedfor5Gtechnology. ThelowersectionofFigure1isoftenreferred toastheoperationaltechnology(OT)partofthe manufacturingplant,comprisingboththefield level(industrialdevicesandcontrollers)andthe manufacturingexecutionsystem.Thetopsection istheinformationtechnology(IT)part,madeup ofgeneralenterpriseresourceplanning.For connectivityatfieldlevel,avarietyoffieldbusand industrialEthernettechnologiesaretypicallyused. EthernetandIParewellestablishedcommunication protocolsathigherlevels(ITandthetoppartofOT). TheOTnetworkdomainiscurrentlydominated (90percent)bywiredtechnologies[3]andisa heavilyfragmentedmarketwithtechnologiessuchas PROFIBUS,PROFINET,EtherCAT,Sercosand Modbus.Currentlydeployedwirelesssolutions (whicharetypicallywirelessLANbasedusing unlicensedspectrum)constituteonlyasmallfraction The goal of Industry 4.0 is to maximize efficiency by creating full transparency across all processes and assets at all times. Achieving this requires communication between goods, production systems, logistics chains, people and processes throughout a product’s complete life cycle, spanning everything from design, ordering, manufacturing, delivery and field maintenance to recycling and reuse. The integration of 5G ultra-reliable low-latency communication (URLLC) in the manufacturing process has great potential to accelerate the transformation of the manufacturing industry and make smart factories more efficient and productive. ■ Today’sstate-of-the-artfactoriesare predominantlybuiltonahierarchicalnetworkdesign thatfollowstheindustrialautomationpyramid,as showninFigure1.Thefourthindustrialrevolution willrequireatransitionfromthissegmentedand hierarchicalnetworkdesigntowardafullyconnected one.Thistransition,incombinationwiththe introductionof5Gwirelesscommunication technology,willprovideveryhighflexibilityin buildingandconfiguringproductionsystemson demand.Theabilitytoextractmoreinformationfrom themanufacturingprocessandfeeditintoadigital representationknownasthe“digitaltwin”[1]enables moreadvancedplanningprocesses,includingplant simulationandvirtualcommissioning.Initiativeslike the5GAllianceforConnectedIndustriesand Figure 1 Hierarchical network design based on the industrial automation pyramid IT domain OT domain Field level Enterprise resource planning Manufacturing execution system GW GWIndustrial controllers Industrial devices Definition of key terms ❭❭ Ultra-reliable low-latency communication (URLLC) refers to a 5G service category that provides the ability to successfully deliver a message within a specified latency bound with a specified reliability, such as delivering a message within 1ms with a probability of 99.9999 percent. ❭❭ The fourth industrial revolution is considered to be the fourth big step in industry modernization, enabled by cyber-physical systems, digitalization and ubiquitous connectivity provided by 5G and Internet of Things (IoT) technologies. It is also referred to as Industry 4.0. smart manufacturingWITH 5G WIRELESS CONNECTIVITY BOOSTING FEATURE ARTICLE – 5G AND SMART MANUFACTURING ✱✱ FEATURE ARTICLE – 5G AND SMART MANUFACTURING
  20. 20. PHY(physicallayer)featuresaswellasnewmulti- connectivityarchitectureoptionshavebeenadded tothe5GNRspecificationsin3GPPrelease15,and additionalenhancementsarebeingstudiedin release16.Thegoalinrelease16istoenable0.5-1ms one-waylatencywithreliabilityofupto99.9999 percent.Newcapabilitiesincludefasterscheduling, smallerandmorerobusttransmissions,repetitions, fasterretransmissions,preemptionandpacket duplication[5].Allinall,theyensureNRisequipped withapowerfultoolboxthatcanbeusedtotailorthe performancetothedemandsofeachspecificdevice andtrafficflowonafactoryshopfloor. Theachievableround-triptime(RTT)depends bothonwhichfeaturesandspectrumareused.For example,theRANRTTforamid-banddeployment optimizedforMBBcanbeintheorderof5ms(FDD 15kHzSCSorTDD30kHzwithDL-DL-DL-UL TDDconfiguration).ThecorrespondingRTTfora URLLC-optimizedmillimeterwave(mmWave) deployment(TDD120kHzSCS,DL-ULTDD configuration)canbebelow2ms,thusmatchingthe 3GPPone-waylatencygoal. Thereisatrade-offbetweenlatency,reliability andcapacity,anddifferentschedulingstrategiescan beusedtoachieveacertainlevelofreliabilityand latency.Apacketcanbeencodedwithaverylowand robustcoderate,andjustbetransmittedonce,butif theRTTisshorterthantheapplicationlatency constraint,itcanbemoreefficienttouseahigher, lessrobustinitialcoderateandperform retransmissionsbasedonfeedbackincasetheinitial transmissionfails.Thus,theshortertheRANRTT iscomparedwiththeapplicationlatencyconstraint, thehigherspectralefficiency(capacity)maybe achieved. Licensedspectrumforinterferencecontrol Theavailabilityofspectrumresourcesiskeyto meetingrequirementsoncapacity,bitratesand latency.Toprovidepredictableandreliableservice levelsonthefactoryshopfloor,thespectrum resourcesneedtobemanagedcarefully.The achievableperformancedependsonseveralfactors: ❭❭ the amount of spectrum available ❭❭ which spectrum is used – low band (below 2GHz), mid-band (2-5GHz) or high band/mmWave (26GHz and above) ❭❭ which licensing regime applies ❭❭ whether the spectrum is FDD or TDD ❭❭ which radio access technology is used ❭❭ the coexistence scenarios that apply for the spectrum. Estimatesofspectrumneedsareintherangeoftens tohundredsofmegahertz.Mostnewmid-band spectrumthatiscurrentlybeingallocatedusesTDD, whilelargepartsofthespectrumalreadyallocated tomobileoperatorsareFDD.LatencyforanFDD systemisinherentlylowerthanthatofa correspondingTDDsystem. Mid-bandspectrumiswellsuitedforindoor deploymentssinceitspropagationcharacteristics makeiteasytoprovidegoodcoveragewithalimited setoftransmissionpoints.CoverageatmmWaveis generallyspottier,requiringdenserradiodeployment, butmmWaveisstillagoodcomplementtomid-band forin-factorydeploymentssinceitenables: ❭❭ higher system capacity, as larger bandwidths are available and as advanced antenna systems and beamforming can be implemented in a small form factor suitable for indoor deployment ❭❭ significantly shorter latencies (even though the spectrum is TDD), as a higher numerology with shorter transmission time intervals is used ❭❭ easier management of the coexistence between indoor shop floor networks and outdoor mobile networks, as mmWave radio signals are easier to confine within buildings. oftheinstalledbase;theymainlyplayarolefor wirelesslyconnectingsensorswherecommunication requirementsarenon-critical. Today,thefieldlevelconsistsofconnectivity islandsthatareseparatedbygateways(GWs),which helpstoprovidetherequiredperformancewithin eachconnectivityisland.TheGWsarealsoneeded forprotocoltranslationbetweenthedifferent industrialnetworkingtechnologies.However,this segmenteddesignputslimitationsonthe digitalizationoffactories,asinformationwithin onepartofthefactorycannotbeeasilyextracted andusedelsewhere. Onenear-termbenefitofleveragingwireless connectivityinfactoriesisthesignificantreduction intheamountofcablesused,whichreducescost, sincecablesaretypicallyveryexpensivetoinstall, rearrangeorreplace.Inaddition,wireless connectivityenablesnewusecasesthatcannotbe implementedwithwiredconnectivity,suchas movingrobots,automatedguidedvehiclesandthe trackingofproductsastheymovethroughthe productionprocess.Wirelessconnectivityalso makesitpossibletoachievegreaterfloorplanlayout flexibilityanddeployfactoryequipmentmoreeasily. Keymanufacturingindustryrequirements Themanufacturingindustryhasspecific5G requirementsthatdiffersignificantlyfrompublic mobilebroadband(MBB)services.Theseinclude URLLCwithultra-highavailabilityandresilience, whichcanonlybesatisfiedwithadedicatedlocal networkdeploymentusinglicensedspectrum. Theabilitytointegratewiththeexistingindustrial EthernetLANandexistingindustrialnodesand functionsisanotherfundamentalrequirement. Dataintegrityandprivacyarealsocritical,aswellas real-timeperformancemonitoring.Inaddition, 5Gcapabilitiesintermsofpositioning,time synchronizationbetweendevices,securityand networkslicingwillalsobeessentialformany manufacturingusecases. Ultra-reliablelow-latencycommunication Oneofthetwoservicecategoriesofmachine-type communication(MTC)in5G–criticalMTC(cMTC) –isdesignedtomeetcommunicationdemandswith stringentrequirementsonlatency,reliabilityand availability.IntensestandardizationandRDwork isongoingtoensure5GNewRadio(NR)technology isabletofullyaddresstheneedforURLLC. WithNRwewillseelarge-scaledeploymentsof advancedantennasystemsenablingstate-of-the-art beamformingandMIMO(multiple-input,multiple- output)techniques,whicharepowerfultoolsfor improvingthroughput,capacityandcoverage[4]. Multi-antennatechniqueswillalsobeimportantfor URLLC,astheycanbeusedtoimprovereliability. ThescalablenumerologyofNRprovidesgood meanstoachievelowlatency,aslargersubcarrier spacing(SCS)reducesthetransmissiontime interval. Tofurtherreducelatencyandincreasereliability, severalnewMAC(mediumaccesscontrol)and MMWAVEISAGOOD COMPLEMENTTOMID- BANDFORIN-FACTORY DEPLOYMENTS Terms and abbreviations cMTC – Critical Machine-type Communication | CN – Core Network | DL – Downlink | GHz – Gigahertz | GW – Gateway | IoT – Internet of Things | kHz – Kilohertz | LTE-M – LTE Machine-type Communication | MBB – Mobile Broadband | mMTC – Massive Machine-type Communication | mmWave – Millimeter Wave | ms – Millisecond | MTC – Machine-type Communication | NB-IoT – Narrowband IoT | NR – New Radio | OT – Operational Technology | RTT – Round-trip Time | SCS – Subcarrier Spacing | TSN – Time-sensitive Networking | UE - User Equipment | UL – Uplink | URLLC – Ultra-reliable Low-latency Communication FEATURE ARTICLE – 5G AND SMART MANUFACTURING ✱✱ FEATURE ARTICLE – 5G AND SMART MANUFACTURING #01 2019 ✱ ERICSSON TECHNOLOGY REVIEW 4140 ERICSSON TECHNOLOGY REVIEW ✱ #01 2019
  21. 21. Ethernettransporthasbeenspecifiedwithinthe release15standardofthe5Gsystem. Aspartoftheongoingindustrialtransformation, thewiredcommunicationsegmentsofindustrial networksareexpectedtoevolvetowardacommon openstandard:EthernetwithTSNsupport[6]. Therefore,a5Gsystemneedstobeabletointegrate withaTSN-basedindustrialEthernet,forwhich 3GPPhasdefineddifferentstudyandworkitemsin release16ofthe5Gstandards. TSNisanextensionoftheIEEE802.3Ethernet andisstandardizedwithintheTSNtaskgroupin IEEE802.1.AprofileforTSNinindustrial automationisbeingdevelopedbytheIEC/IEEE 60802jointproject[7].TSNincludesthemeansto providedeterministicboundedlatencywithout congestionlossesforprioritizedtrafficonan Ethernetnetworkthatalsotransportstrafficoflower priority.TSNfeaturesincludepriorityqueuingwith resourceallocationmechanisms,time synchronizationbetweennetworknodesand reliabilitymechanismsviaredundanttrafficflows. 5Genhancementsincludesupportofredundant transmissionpaths,whichcanbecombinedwiththe TSNfeature‘Framereplicationandeliminationfor reliability’(FRER)thatisstandardizedinIEEE 802.1CB.Oneoftheresourceallocationfeaturesof TSNforboundingthelatencyforperiodiccontrol trafficis‘Time-awarescheduling’(standardizedin IEEE802.1Qbv),forwhichtransmissionqueuesare time-gatedineveryswitchonthedatapathtocreate aprotectedconnection.ThisrequiresallEthernet switchestobetime-synchronizedaccordingtoIEEE P802.1AS-Rev.Featuresthatarebeingdevelopedin 5Gstandardizationtosupporttime-aware transmissionacrossamixedTSN-5Gnetworkareto time-alignthe5GsystemwiththeTSNnetworkand provide5Gtransmissionwithdeterministiclatency. Keepingthingslocal OntopofURLLCperformanceandintegration withindustrialEthernetnetworks,many manufacturersalsorequirefullcontrol(thatis, independentofexternalparties)oftheircriticalOT domainconnectivityinordertofulfillsystem availabilitytargets.Fullcontrolcanbeexpressed asrequirementsonkeepingthingslocal: ❭❭ local data – the ability to keep production- related data locally within the factory premises for security and trust reasons ❭❭ local management – the ability to monitor and manage the connectivity solution locally ❭❭ local survivability – the ability to guarantee the availability of the connectivity solution independently of external factors (for example, shop-floor connectivity must continue uninterrupted even when connectivity to the manufacturing plant is down). Additionalrequirementsandfeaturesofinterest One5Gfeaturethatcouldhavesignificantimportance formanufacturingusecasesispositioning.For3GPP release16,theobjectiveistoachieveindoor positioningaccuraciesbelow3m,butNRdeployed inafactoryenvironmenthasthetechnologypotential tosupportmuchmoreprecisepositioning.Thereare severalaspectswhichallcontributetobetter positioningaccuracy: ❭❭ the wide bandwidths of mid- and high-band spectrum enable better measurement accuracy ❭❭ beam-based systems enable better ranging and angle-of-arrival/departure estimation ❭❭ the higher numerology of NR implies shorter sampling intervals and hence improved positioning resolution ❭❭ dense and tailored deployments with small cells and large overlaps improve accuracy and, together with beam-based transmissions, provide more spatial variations that can be exploited for radio frequency fingerprinting. In5Grelease16,anewrequirementisbeing introduced,wherebythe5Gsystemwillbeableto synchronizedevicestoamasterclockofoneormore timedomains[8].Onereasonforthisisthatseveral Forcriticalapplications,theremustbeguarantees againstuncontrolledinterference,whichimpliesthat licensedspectrumisnecessary.Asillustratedin Figure2,unlicensedtechnologiessuchasWi-Fiand MulteFirecannotguaranteeboundedlowlatency withhighreliabilityastheloadincreases.Thisisdue totheuseoflisten-before-talkback-off,whichdoes notperformwellduringuncontrolledinterference. Unlicensedspectrummaynonethelessberelevant forlesscriticalapplications. Licensedspectrumcanbeprovidedbyoperators aspartofalocalconnectivitysolution,including networkequipment.Operatorsmayalsochooseto leasepartsoftheirspectrumassetslocallyto industrieswithoutprovidingtheconnectivity solution.Anotheremergingoptionisforregulators tosetasidededicatedspectrumforlocallicensingto industries,asisunderconsiderationinsome EuropeancountriessuchasGermanyandSweden on3.7-3.8GHz. IntegrationwithindustrialEthernetandTSN Theintroductionof5Gonthefactoryshopfloorwill happeninsteps.When5Gisaddedtoexisting productionsystems,thevariouspartsofthesystem willbemovedto5Gconnectivityatdifferentstages, dependingontheevolutionplanoftheproduction systemandwherethehighestbenefitsofwireless5G communicationcanbeobtained.Overtime,more partsoftheshopfloorcanbemigratedto5G,inpart duetotheintroductionofnewcapabilitiesinfuture 5Greleases.Eveningreenfieldindustrial deployments,notallcommunicationwillbebased on5G.Theneedforwirelessconnectivitymaynotbe prominentforsomesubsystems,whileothersmay requireperformancelevels(isochronoussub- millisecondlatency,forexample)thatarenot currentlyaddressedby5G.Consequently,alocal industrial5Gdeploymentwillcoexistandrequire integrationwithwiredindustrialLANs.Tothisend, thetransportofEthernettrafficisrequired,and Figure 2 Latency and reliability aspects of spectrum and technology choice Asset monitoring Wireless sensors Non-real-time Soft real-time Mobile robots Automated guided vehicles Hard real-time Time-critical closed-loop control Wi-Fi Low (milliseconds) Low High High (seconds) End-to-end latency Reliability (with load) Wi-Fi MulteFire LTE NR Unlicensed spectrum Licensed spectrum MulteFire LTE NR FEATURE ARTICLE – 5G AND SMART MANUFACTURING ✱✱ FEATURE ARTICLE – 5G AND SMART MANUFACTURING #01 2019 ✱ ERICSSON TECHNOLOGY REVIEW 4342 ERICSSON TECHNOLOGY REVIEW ✱ #01 2019

×