scieee Open visual document viewer

D1.1 Refined scenarios and requirements, consolidated use cases, and qualitative techno-economic feasibility assessment

Monserrat del Río, José Francisco,Martín-Sacristán Gandía, David

Abstract

This document describes scenarios, consolidated use cases and associated requirements for wireless access networks in the 2020-2030 timeframe. These are based on METIS project and also taking into account work done in other 5G projects and forums such as ITU-R and NGMN. The document introduces spectrum authorization modes and describes spectrum usages scenarios, spectrum bands and spectrum demand for 5G services. Finally, this document provides qualitative techno-economic feasibility assessment by analyzing main players involved in service delivery, from the radio access network point of view, and describing their mutual positions and relationships.

Full text

Mobile and wi eless communica ions Enable s o he Twen y- wen y In o ma ion Socie y-II Deli e able D1.1 Re ined scena ios and equi emen s, consolida ed use cases, and quali a i e echno-economic easibili y assessmen Ve sion: 1.0 2016-01-31 2 Deli e able D1.1 Re ined scena ios and equi emen s, consolida ed use cases, and quali a i e echno-economic easibili y assessmen G an Ag eemen Numbe : 671680 P ojec Name: Mobile and wi eless communica ions Enable s o he Twen y- wen y In o ma ion Socie y-II P ojec Ac onym: METIS-II Documen Numbe : METIS-II/D1.1 Documen Ti le: Re ined scena ios and equi emen s, consolida ed use cases, and quali a i e echno-economic easibili y assessmen Ve sion: 1.0 Deli e y Da e: 2016-01-31 Edi o s: Salah Eddine El Ayoubi (O ange), F édé ic Pujol (IDATE), Mikael Fallg en (E icsson) Au ho s: Salah Eddine El Ayoubi, Sébas ien Jeux, F édé ic Ma ache (O ange), F édé ic Pujol (IDATE), Mikael Fallg en (E icsson), Panagio is Spapis (HWDU), Changqing Yang (HWHQ), Ash a Widaa, Jan Ma kendahl, Ami hossein Ghanba i (KTH), Rauno Ruismäki, Mikko A. Uusi alo (Nokia), Wol gang Koenig (ALU), Alessand o T ogolo (TI), Isabelle Ko hals, Thomas Rosowski, Ge d Zimme mann (DT), Luis Campoy (TID), Jose F. Monse a , Da id Ma ín- Sac is án (UPV), Hans Scho en (UKL), Yinan Qi (Samsung), Shubh anshu Singh (ITRI), Pa ick Agyapong (DCM) Keywo ds: 5G use cases, spec um scena ios, 5G RAN business models S a us: Final 3 Dissemina ion le el: Public Abs ac This documen desc ibes scena ios, consolida ed use cases and associa ed equi emen s o wi eless access ne wo ks in he 2020-2030 ime ame. These a e based on METIS p ojec and also aking in o accoun wo k done in o he 5G p ojec s and o ums such as ITU-R and NGMN. The documen in oduces spec um au ho iza ion modes and desc ibes spec um usages scena ios, spec um bands and spec um demand o 5G se ices. Finally, his documen p o ides quali a i e echno-economic easibili y assessmen by analyzing main playe s in ol ed in se ice deli e y, om he adio access ne wo k poin o iew, and desc ibing hei mu ual posi ions and ela ionships. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 4 Execu i e summa y One objec i e o he METIS-II is o acili a e he discussion on scena ios, use cases, KPIs and equi emen s o 5G, building upon he comp ehensi e wo k conduc ed in he METIS p ojec [MET13-D11] and aking he wo k o o he Eu opean p ojec s and o he bodies such as ITU-R, NGMN e c. in o accoun . This documen iden i ies i e 5G use cases ha co e he h ee main se ices (xMBB, mMTC and uMTC), ha e s ingen equi emen s and whose echnical solu ions a e expec ed o se e also o o he simila use cases, as ollows: - he “Dense u ban in o ma ion socie y” use case is conce ned wi h he connec i i y equi ed a any place and a any ime by humans in dense u ban en i onmen s, - he “Vi ual eali y o ice” use case is ela ed o he e olu ion o oday‟s ele-p esence se ices in o high- esolu ion 3D e sions, which will allow people o ha e he amazing expe ience “as i you we e he e”, - he “B oadband access e e ywhe e” use case is ela ed o he cons an inc ease o he demand o e y high da a a e In e ne access a any ime and a any place and o he challenge o sa is ying he ubiqui ous capaci y demands o u u e use s in a eas wi h spa se ne wo k in as uc u e, such as sca cely popula ed a eas, u al and e en subu ban a eas, - he “Massi e dis ibu ion o senso s and ac ua o s” use case co e s he massi e deploymen o low cos and low ene gy consump ion de ices ha need o communica e wi h o he de ices and wi h he ne wo k, - he “Connec ed ca s” use case add esses in o ma ion exchange among ehicles and wi h he in as uc u e o enable he p o ision o sa e y hin s o he d i e o wa nings abou he oad s a us. I also add esses xMBB se ices on-boa d o ca s. Ha ing de ined hese use cases, his documen add esses he analysis o he use cases and echnical solu ions de eloped in METIS-II om a quali a i e echno-economic easibili y poin o iew. This quali a i e assessmen consis s in explo ing he mobile access ne wo k ecosys em e olu ions, based on he spec um scena ios and he i s 5G RAN design conside a ions de eloped in METIS-II. We iden i ied he main cu en playe s in he Radio Access Ne wo k and p esen ed he possible e olu ions o hei posi ioning in he alue chain in he 5G e a. We also iden i ied new playe s ha may play impo an oles in 5G and iden i ied hei ela ionships wi h he classical playe s. These ecosys em e olu ions a e also highligh ed by a alue ne analysis, whe e mobile ope a o s occupy he cen e o he scene and he cu en and new playe s su ound hem wi hin ou ca ego ies: cus ome s, supplie s, compe i o s and complemen o s. In o de o be mo e speci ic, we end he documen wi h a de ailed analysis o he dense u ban in o ma ion socie y use case, whe e he oles o he key playe s a e p esen ed o di e en possible 5G ne wo k deploymen s a egies. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 5 Con en s Glossa y....................................................................................................................................10 1 In oduc ion ........................................................................................................................12 1.1 Objec i e o he documen ..........................................................................................12 1.2 S uc u e o he documen ...........................................................................................12 2 METIS-II 5G use cases ......................................................................................................13 2.1 Consolida ed use cases ..............................................................................................15 2.1.1 UC1: Dense u ban in o ma ion socie y .................................................................15 2.1.2 UC2: Vi ual eali y o ice .....................................................................................15 2.1.3 UC3: B oadband access e e ywhe e ...................................................................16 2.1.4 UC4: Massi e dis ibu ion o senso s and ac ua o s .............................................16 2.1.5 UC5: Connec ed ca s ...........................................................................................17 2.2 Re ined equi emen s ..................................................................................................17 3 5G spec um scena ios ......................................................................................................19 3.1 Spec um au ho iza ion modes and usage scena ios ..................................................19 3.2 Spec um bands o 5G gene ic se ices ....................................................................22 3.3 Spec um demand.......................................................................................................22 3.3.1 Spec um below 6 GHz ........................................................................................23 3.3.2 Resul o WRC-15 on bands abo e 6 GHz ...........................................................24 4 Ecosys em e olu ions ........................................................................................................25 4.1 Cu en alue chains in he mobile sec o ...................................................................25 4.1.1 Iden i ica ion o he main playe s and hei espec i e oles .................................25 4.1.2 P esen a ion o he cu en alue chains ..............................................................26 4.2 Gene al ends o he e olu ion o he alue chains ....................................................30 4.2.1 Assump ions on 5G a chi ec u e ..........................................................................30 4.2.2 E olu ions o he posi ioning o cu en playe s wi h 5G .......................................36 4.2.3 Iden i ica ion o new playe s in he 5G ield ..........................................................41 4.2.4 New alue chains .................................................................................................44 4.2.5 MTC Ac o s in Sma Ci y and he co esponding Value Ne wo k .........................48 4.2.6 E olu ions o he MNO-cen ic alue Ne ..............................................................50 4.3 Focus on he dense u ban in o ma ion socie y use case .............................................55 Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 6 4.3.1 E olu ion in RAN a chi ec u e and deploymen op ions ........................................55 4.3.2 Mapping ac o s o he oles: xMBB alue ne wo k ................................................57 5 Summa y ...........................................................................................................................61 6 Re e ences ........................................................................................................................62 A Key Pe o mance Indica o s (KPIs) ....................................................................................66 A.1 T a ic olume densi y .................................................................................................66 A.2 Expe ienced use h oughpu ......................................................................................68 A.3 La ency .......................................................................................................................69 A.4 Reliabili y ....................................................................................................................70 A.5 A ailabili y and e ainabili y .........................................................................................71 A.6 Ene gy consump ion/e iciency....................................................................................71 A.7 Cos ............................................................................................................................73 B Analysis o SOTA 5G use cases ........................................................................................75 B.1 Analysis o METIS-I use cases....................................................................................75 B.2 Analysis o EU p ojec s use cases ..............................................................................78 B.3 Analysis o NGMN use cases ......................................................................................81 B.4 Analysis o ITU-R usage scena ios .............................................................................83 B.5 3GPP S udy on New Se ices and Ma ke s Technology Enable s ..............................87 C METIS-II 5G use cases ......................................................................................................90 C.1 UC1: Dense u ban in o ma ion socie y ........................................................................90 C.1.1 Backg ound and mo i a ion ..................................................................................90 C.1.2 KPIs and equi emen s ........................................................................................91 C.2 UC2: Vi ual eali y o ice.............................................................................................93 C.2.1 Backg ound and mo i a ion ..................................................................................93 C.2.2 KPIs and equi emen s ........................................................................................94 C.3 UC3: B oadband access e e ywhe e ..........................................................................96 C.3.1 Backg ound and mo i a ion ..................................................................................96 C.3.2 KPIs and equi emen s ........................................................................................96 C.4 UC4: Massi e dis ibu ion o senso s and ac ua o s ....................................................99 C.4.1 Backg ound and mo i a ion ..................................................................................99 C.4.2 KPIs and equi emen s ...................................................................................... 100 Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 7 C.5 UC5: Connec ed ca s ................................................................................................ 102 C.5.1 Backg ound and mo i a ion ................................................................................ 102 C.5.2 KPIs and equi emen s ...................................................................................... 102 D De ailed analysis o he cu en alue chains in he mobile sec o .................................... 108 D.1.1 Iden i ica ion o he main playe s & desc ip ion o hei espec i e oles ............. 108 D.1.2 De ailed alue chains in he mobile sec o ......................................................... 121 Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 8 Lis o abb e ia ions and ac onyms 2G Second Gene a ion 3D Th ee Dimensions 3GPP Thi d Gene a ion Pa ne ship P ojec 4K e e s o a esolu ion o 4000 pixels 5G Fi h Gene a ion 5G-PPP 5G P i a e Public Pa ne ship ADSL Asymme ic Digi al Subsc ibe Line AEP Applica ion Enablemen Pla o m API Applica ion P og amming In e ace ARPU A e age Re enue Pe Use ATG Ai - o-G ound B2B Business- o-Business B2B2C Business- o-Business o-Cus ome B2C Business- o-Cus ome BB Base Band BBU Base band Uni C-RAN Cen alized/Cloud-Radio Access Ne wo k CAPEX CAPi al EXpendi u es CBRS Ci izens B oadband Radio Se ice CDN Con en Deli e y Ne wo k CDP Connec ed De ice Pla o m CEPT Eu opean Con e ence o Pos al and Telecommunica ions Adminis a ions CN Co e Ne wo k CP Con ol Plane CRM Cus ome Rela ionship Managemen D2D De ice- o-De ice DAS Dis ibu ed An enna Sys em dB decibel DL DownLink E2E End- o-End ECU Elec onic Con ol Uni FCC Fede al Communica ions Commission FM Facili y Manage GAA Gene alized Au ho ized Access GEO Geos a iona y Ea h O bi GPRS Gene al Packe Radio Se ice HARQ Hyb id Au oma ic Repea eQues HTHP High Towe High Powe HW Ha dwa e IaaS In as uc u e as a Se ice ICT In o ma ion and Communica ion Technology IEEE Ins i u e o Elec ical and Elec onics Enginee s IMT In e na ional Mobile Telecommunica ions IMT-2000 e m used by ITU o a globally ha monized s anda ds o 3G IoO In e ne o Objec s IoT In e ne o Things ISP In e ne Se ice P o ide ITU In e na ional Telecommunica ion Union ITU-R ITU Radiocommunica ion Sec o ITS In elligen T anspo Sys ems KPI Key Pe o mance Indica o LAA Licensed-Assis ed Access LEO Low Ea h O bi LTLP Low Towe Low Powe LPWA Low Powe Wide A ea LSA Licensed Sha ed Access M2M Machine- o-Machine MAC Medium Access Con ol MBB Mobile B oadBand METIS Mobile and wi eless communica ions Enable s o he Twen y- wen y In o ma ion Socie y Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 9 MFCN Mobile/Fixed Communica ions Ne wo k MMC Massi e Machine Communica ion MNO Mobile Ne wo k Ope a o mMTC Massi e Machine-Type Communica ions MSP Managed Se ice P o ide MTC Machine Type Communica ion MVNO Mobile Vi ual Ne wo k Ope a o NaaS Ne wo k as a Se ice NEP Ne wo k Equipmen P o ide NF Ne wo k Func ion NFV Ne wo k Func ion Vi ualiza ion NGMN Nex Gene a ion Mobile Ne wo k NRA Na ional Regula o y Au ho i y NW Ne Wo k OPEX Ope a ional EXpendi u es OSS Ope a ional Suppo Sys em OTT O e The Top One T ip Time PaaS Pla o m as a Se ice PAL P io i y Access License PDCP Packe Da a Con e gence P o ocol PPDR Public P o ec ion and Disas e Relie PVNO P i a e Vi ual Ne wo k Ope a o QoE Quali y o Expe ience QoS Quali y o Se ice RAN Radio Access Ne wo k RAT Radio Access Technology RF Radio F equency RLC Radio Link Con ol RRC Radio Resou ce Con ol RTT Round T ip Time SCaaS Small Cell as a Se ice SCells Seconda y Cells SDN So wa e De ined Ne wo k SINR Signal- o-In e e ence-plus-Noise Ra io SMS Sho Message Se ice SON Sel -O ganising Ne wo k SW So wa e TC Tes Case Towe Company TOo5G Towe O e lay o e 5G UAV Unmanned Ae ial Vehicle UC Use Case UE Use Equipmen UHD Ul a-High De ini ion UL UpLink uMTC Ul a- eliable Machine-Type Communica ions UP Use Plane V2D Vehicle- o-De ice V2V Vehicle- o-Vehicle V2X Vehicle- o-Any hing Wi-Fi Wi eless local access ne wo k echnologies acco ding IEEE 802.11 speci ica ions and ce i ied by he Wi-Fi Alliance WRC-15 Wo ld Radiocommunica ion Con e ence 2015 WRC-19 Wo ld Radiocommunica ion Con e ence 2019 XaaS Any hing as a Se ice xMBB Ex eme Mobile B oadBand Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 16 2.1.3 UC3: B oadband access e e ywhe e The demand o e y high expe ienced use h oughpu In e ne access a any ime and a any place is cons an ly inc easing. The ubiqui ous capaci y demands o u u e use s will be challenging o sa is y in a eas wi h spa se ne wo k in as uc u e, such as sca cely popula ed a eas, u al and e en subu ban a eas. A consis en use expe ience wi h espec o h oughpu , needs a minimum expe ienced use h oughpu gua an eed e e ywhe e. The a ge alue o 50 Mbps e e ywhe e should be unde s ood as he minimum expe ienced use h oughpu and no a single use ‟s heo e ical peak a e. Fu he mo e, i is emphasized ha his use a e has o be deli e ed consis en ly ac oss he co e age a ea (i.e. e en a cell edges). Fu he mo e, he ba e y consump ion o sma phones and able e minals in a eas wi h low co e age inc eases signi ican ly due o highe p opaga ion losses. A a ge o 50% ene gy consump ion educ ion compa ed wi h legacy ne wo k should be achie ed bo h a ne wo k in as uc u e and a end use le el. While cell densi ica ion is p omising o boos ing capaci y in u u e u ban en i onmen , wide co e age solu ions as well as lexible, ene gy and cos e icien solu ions mus be de eloped in u u e wi eless communica ion sys ems o p o ide ubiqui ous co e age in subu ban and mo e o less emo e u al a eas. This is a new METIS use case buil on elemen s o a NGMN use case [NGM15] combined wi h some aspec s om he METIS-I es case se en (Blind spo s) in [MET13-D11]. This use case mainly add esses he use case amily xMBB. 2.1.4 UC4: Massi e dis ibu ion o senso s and ac ua o s The impo ance o his use case will g ow oge he wi h he massi e deploymen o low cos and o low ene gy consump ion de ices. In o de o ge he maximum o in o ma ion om hese de ices, o inc ease en i onmen al awa eness and be e use expe ience, he e is a need o hese de ices o be able o communica e wi h o he de ices, he ne wo k, o wi h o he mobile phones. People communica ing and exchanging con en can ake place in combina ion wi h ecei ing wind, i e and humidi y upda es om senso s. This use case is based on a es case ele en in [MET13- D11]. I mainly add esses he use case amily mMTC. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 17 2.1.5 UC5: Connec ed ca s The use o emo e se ices is also applicable a highe use mobili y, e.g. while d i ing ca s o using public anspo a ion, and no me ely aking place in slow mobili y o s a iona y se ings. Wi h highe use mobili y enabled on- he-way wo ke s as well as a leisu ed people can enjoy he bene i s o eal- ime emo e compu ing o mobile e minals. A he same ime he connec ed ca also p o ides a sa e and e icien jou ney ia he communica ion o i s su ounding. This communica ion enables he ca o a oid acciden s, bu i also enables o he ypes o a ic planning such as a oiding a ic jam queues and minimizing uel consump ion. The eby he connec ed ca enables a ic sa e y, e iciency and eal- ime emo e se ices. This use case is based on es cases eigh and wel e o METIS-I [MET13-D11]. I mainly add esses he use case amilies uMTC and xMBB. 2.2 Re ined equi emen s The main KPIs and equi emen s o he METIS-II use cases a e gi en in Table 2-1. The de ailed equi emen s a e gi en in Appendix C. Table 2-1: Summa y o main KPIs and equi emen s o each METIS-II use case. Use Case (UC) Key Pe o mance Indica o (KPI) Requi emen UC1 Dense u ban in o ma ion socie y Expe ienced use h oughpu 300 Mbps in DL and 50 Mbps in UL a 95% a ailabili y and 95% eliabili y E2E RTT la ency Less han 5 ms (augmen ed eali y applica ions) UC2 Vi ual eali y o ice Expe ienced use h oughpu 5 (1) Gbps wi h 20% (95%) a ailabili y in DL 5 (1) Gbps wi h 20% (95%) a ailabili y in UL bo h wi h 99% eliabili y UC3 B oadband access e e ywhe e Expe ienced use h oughpu 50 Mbps in DL and 25 Mbps in UL a 99% a ailabili y and 95% e ainabili y UC4 Massi e dis ibu ion o senso s and ac ua o s A ailabili y 99.9% De ice densi y T a ic olume pe de ice 1 000 000 de ices/km2 F om ew by es pe day o 125 by es pe second UC5 Connec ed ca s E2E one-way la ency 5 ms ( a ic sa e y applica ions) Expe ienced use h oughpu 100 Mbps in DL and 20 Mbps in UL (se ice applica ions) a 99% a ailabili y and 95% eliabili y Vehicle eloci y Up o 250 km/h Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 18 These use cases, along wi h hei key equi emen s, a e illus a ed in Figu e 2-2. Figu e 2-2: The METIS-II 5G use cases and hei mapping o he 5G se ices. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 19 3 5G spec um scena ios In his sec ion he p elimina y 5G spec um scena ios a e in oduced. Spec um au ho iza ion modes and usage scena ios, spec um bands o he 5G gene ic se ices xMBB, mMTC and uMTC, and spec um demand a e conside ed. Globally ha monized equency bands below 6 GHz, as well as bands be ween 24 - 86 GHz unde s udy in ITU-R o IMT sys ems a e b ie ly e lec ed. 3.1 Spec um au ho iza ion modes and usage scena ios In gene al he use o adio spec um can be au ho ized in wo ways: Indi idual Au ho iza ion (Licensed) and Gene al Au ho iza ion (License Exemp / Unlicensed). Au ho iza ion modes ecognized as ele an o wi eless communica ions a e P ima y use mode, LSA (Licensed Sha ed Access) mode and Unlicensed mode. Fi e basic spec um usage scena ios can be iden i ied o hese au ho iza ion modes: dedica ed licensed spec um, limi ed spec um pool, mu ual en ing, e ical sha ing and unlicensed ho izon al sha ing (see Figu e 3-1, depic ing he ela ions be ween pa s o he domains which a e ei he necessa y (manda o y: con inuous lines) o supplemen a y (op ional: do ed lines)). Figu e 3-1: Spec um usage/sha ing scena ios [MET15-D54]. The gene al conclusion wi h ega d o spec um au ho iza ion o 5G is ha : Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 20  exclusi e licensed spec um is essen ial o he success o 5G o p o ide he expec ed QoS and o secu e in es men s,  sha ed spec um can be conside ed in addi ion, p o ided ha p edic able QoS condi ions a e main ained, e.g. by LSA egime,  license-exemp spec um migh be sui able as a supplemen a y op ion o ce ain applica ions, o ins ance using he LAA (License-Assis ed Access) scheme explained below. In Eu ope LSA is he ecognized app oach by CEPT (Eu opean Con e ence o Pos al and Telecommunica ions Adminis a ions) o adminis a ions wishing o in oduce Mobile/Fixed Communica ions Ne wo ks (MFCN) in he band 2300-2400 MHz while main aining he cu en incumben use [ECC14-DEC02], [ECC14-REP205]. Di e en LSA ials a e cu en ly on-going in Eu ope. In he US, he FCC has announced he i s se o ules o he band 3550-3700 MHz, e med Ci izens B oadband Radio Se ice (CBRS), o enable deploymen o ela i ely low powe ed ne wo k echnologies like small cells [FCC15-47]. The ules equi e a Spec um Access Sys em (SAS) based on exclusion/p o ec ion zones and sensing o implemen a h ee- ie ed egula o y amewo k in which incumben s ope a e in he op ie , P io i y Access License (PAL) use s ope a e in he second ie and Gene alized Au ho ized Access (GAA) use s ope a e in he hi d ie (Figu e 3-2). Figu e 3-2: FCC Te minology a 3.5 GHz [KRB15]. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 21 The incumben s which ha e he highes p io i y spec um access can equi e all seconda y sys ems in a spa ial egion o elinquish all o po ions o spec um a any ime. To he seconda y de ices om he PAL ie exclusi e channels a e alloca ed ha p o ide gua an eed in e e ence p o ec ion, simila o adi ional long e m licenses. The GAA ie p o ides no in e e ence p o ec ion and he e o e, allows mul iple independen ne wo ks o use a gi en channel no used by incumben o PAL use s. LAA is an inno a i e app oach o unlicensed ho izon al sha ing in he Unlicensed mode ha aims a enabling he ope a ion o LTE in he 5 GHz unlicensed spec um o low powe Seconda y Cells (SCells). 3GPP has conduc ed a co esponding s udy [3GPP14-LAA], summa izing he necessa y modi ica ions o LTE and he ele an egula o y equi emen s (e.g. o achie e co-exis ence wi h Wi-Fi). LAA a ge s ca ie agg ega ion ope a ion in which one o mo e low powe SCells a e ope a ing in unlicensed spec um. LAA deploymen s encompass scena ios wi h and wi hou mac o co e age, bo h ou doo and indoo small cell deploymen s, and bo h co-loca ion and non-co- loca ion (wi h ideal backhaul) be ween licensed and unlicensed ca ie s. Figu e 3-3 shows ou LAA deploymen scena ios, whe e he numbe o licensed ca ie s and he numbe o unlicensed ca ie s can be one o mo e. Figu e 3-3: LAA deploymen scena ios [3GPP14-LAA]. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 22 Based on he s udy esul s, 3GPP adop ed a wo k i em o speci y suppo o DL ansmissions o LAA SCells o Release 13. Speci ica ion o suppo o UL ansmissions o LAA SCells is planned o be added in u u e eleases, wi hou modi ica ions o he DL design. 3.2 Spec um bands o 5G gene ic se ices The ollowing gene al conclusions on he sui abili y o spec um bands o he 5G gene ic se ices as de ined by METIS – xMBB, mMTC and uMTC - can be d awn:  Fo xMBB, a mix u e o equency spec um comp ising lowe bands o bo h co e age pu poses and da a a ic, and highe bands wi h la ge con iguous bandwid h o cope wi h he e e -inc easing a ic demand, including wi eless backhaul solu ions, is equi ed. Exclusi e licensed spec um is essen ial o gua an ee he co e age obliga ion and QoS, supplemen ed by spec um au ho ized by o he licensing egimes, e.g. LSA o unlicensed access (e.g. Wi-Fi o load) o new enhanced unlicensed access schemes (e.g. LAA) o inc ease o e all spec um a ailabili y.  Fo some mMTC applica ions, equency spec um below 6 GHz is mos sui able and spec um below 1 GHz is needed in pa icula when la ge co e age a eas and good pene a ion a e needed. Exclusi e licensed spec um is he p e e ed op ion. Howe e , o he licensing egimes migh be conside ed depending on speci ic applica ion equi emen s.  Licensed spec um is conside ed mos app op ia e o uMTC. Fo sa e y V2V and V2X communica ion he equency band 5875-5925 MHz ha monized o In elligen T anspo Sys ems (ITS) [ECC08-DEC01] is an op ion. Ano he op ion is he sub-1GHz spec um, pa icula ly well-sui ed o high-speed applica ions and u al en i onmen s. 3.3 Spec um demand The a ie y o 5G se ices, scena ios and use cases pu s high demands on u u e mobile ne wo ks wi h ega d o co e age, capaci y and eliabili y, o which he a ailabili y o a spec um amoun o se e al GHz is equi ed, o be sough in a combina ion o sui able equency bands alloca ed o he mobile se ice. An example in [MET15-R31] wi h an assumed a ic inc ease o xMBB applica ion 4K ideo up o 70 Gbps/km2 in he yea 2025 shows ha addi ional bandwid h o se e al hund eds o MHz pe ope a o will be needed. Ano he example in [MET15-R31] shows ha uMTC also needs simila bandwid hs as xMBB, i.e. in he o de o se e al hund eds o MHz, due o he o e all impac o la ency and eliabili y equi emen s and he amoun o de ices a he same ime. The use case “Vi ual eali y o ice” has been de ec ed as he mos bandwid h demanding scena io e alua ed in he METIS p ojec [MET14-D53]. The assump ions conside ed o ca y Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 23 ou he analysis we e spec um agnos ic, bu depending on he ac ual SINR dis ibu ions achie able (which a he end will depend on he echnology componen s pe o mance in he selec ed scena io), leading o he ou comes summa ized in Table 3-1. These e alua ions ha e been ocused on he spec um needs, independen o a speci ic equency ange, and he e o e ep esen a good s a ing poin o balance he bandwid h equi emen wi h he echnology componen s pe o mance, and wi h he ne wo k deploymen . Table 3-1: Bandwi h demand o “Vi ual eali yo ice” [MET14-D53]. A ailable bandwid h Mean alues o SINR dis ibu ion needed in he ne wo k 4500 MHz 5 - 10 dB 3000 MHz 10 - 15 dB 1500 MHz ˃ 20 dB The alues p esen ed in Table 3-1 a e om he METIS-I p ojec and migh be e isi ed du ing he METIS-II p ojec . 3.3.1 Spec um below 6 GHz Spec um equi emen s o he yea 2020 ha e been calcula ed o p e-IMT sys ems, IMT-2000, and i s enhancemen s, and IMT-Ad anced [ITU-2290]. Due o di e ences in ma ke s, deploymen s, and imings o he mobile da a g ow h in di e en coun ies, wo se ings ha e been e alua ed o cha ac e ize lowe and highe use densi y se ings. The esul s a e p esen ed in Table 3-2. Table 3-2: To al spec um demand calcula ion o he yea 2020. To alspec um equi emen s o “P e-IMT sys ems, IMT-2000 and i s enhancemen s, and IMT-ad anced”in heyea 2020 Lowe use densi y se ings 1340 MHz Highe use densi y se ings 1960 MHz Spec um bands cu en ly iden i ied o IMT in he ITU-R Radio Regula ions will ha e o cope wi h he mobile a ic demand un il he yea 2020. In Eu ope, o example, he o e all amoun o spec um iden i ied o IMT is 1232 MHz, howe e , no all o hose IMT iden i ied bands a e a ailable o mobile communica ions in e e y coun y. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 24 3.3.2 Resul o WRC-15 on bands abo e 6 GHz WRC-15 has ag eed ha ITU-R will conduc and comple e in ime o WRC-19 he app op ia e sha ing and compa ibili y s udies by 31 Ma ch 2017, o he equency bands:  24.25-27.5 GHz, 37-40.5 GHz, 42.5-43.5 GHz, 45.5-47 GHz, 47.2-50.2 GHz, 50.4-52.6 GHz, 66-76 GHz and 81-86 GHz, which ha e alloca ions o he mobile se ice on a p ima y basis; and  31.8-33.4 GHz, 40.5-42.5 GHz and 47-47.2 GHz, which may equi e addi ional alloca ions o he mobile se ice on a p ima y basis. Some o hese equency bands enable wide con iguous bandwid hs which would enable o cope wi h he spec um equi emen s o high bandwid h demanding xMBB applica ions. Fo he s udy in ime o WRC-19, de elopmen o new me hodology o he spec um equi emen es ima e should be he p io i y, conside ing he 5G en i onmen . Mo eo e , he new me hodology should ake in o accoun 5G sys em equi emen s and i s applica ions om he use cases o ocus on 5G ha can make people enjoy new se ices in e ms o highe da a a e, so called „applica ion-based app oach‟. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 25 4 Ecosys em e olu ions This sec ion builds on he 5G use cases and spec um scena ios desc ibed abo e and on RAN a chi ec u al conside a ions o cons uc p elimina y business models. Sec ion 4.1 s a s by desc ibing he cu en alue chains in he mobile sec o . Sec ion 4.2 p esen s he gene al ends o he e olu ion o he alue chains by iden i ying new ac o s and hei ela ionships. Sec ion 4.3 ocuses on business models o he dense u ban in o ma ion socie y use case. 4.1 Cu en alue chains in he mobile sec o 4.1.1 Iden i ica ion o he main playe s and hei espec i e oles This sec ion p esen s he main exis ing playe s h ough he RAN, which a e in ol ed in he mobile se ice alue chain and hei espec i e oles. De ails a e gi en in Appendix D.1.1. Figu e 4-1 p o ides an o e iew o cu en main playe s in ol ed in adio connec i i y se ices. The oles o MNOs, MVNOs, ou sou cing playe s, de ice manu ac u e s and OTTs/con en owne s a e de ailed in Sec ion 4.1.2 as hey a e di ec ly p esen in he mobile ne wo ks alue chain. O he playe s a e also p esen ed in Figu e 4-1, whose oles a e mo e ela ed o speci ic se ices like in- ligh connec i i y, M2M and IoT and Public sa e y and public p o ec ion playe s; hei alue chains a e de ailed in Appendix D. Figu e 4-1: An illus a ion he main exis ing playe s ha e ol e a ound he RAN. Radio Access Ne wo k MNOs and MVNOs playe s Ou sou cing playe s Public Sa e y and Public P o ec ion and Disas e Relie s playe s IoT/MTC playe s Con en owne s & OTTs playe s De ice manu ac u e s, In- ligh connec i i y Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 32 F om an end cus ome pe spec i e a NW slice is seen as an independen ne wo k. Howe e , in con as o deploying independen ne wo k in as uc u es as i was he case in o me mobile adio gene a ions, i will be ealized oge he wi h o he slices on a common in as uc u e laye (also e e ed o as “ i ual ne wo k”), also including asse s such as spec um. In his way, he in as uc u e and asse s u iliza ion is likely o be much mo e cos and ene gy e icien . The slicing concep is en isioned o u he allow o a much as e se -up o new se ices and applica ions o modi ica ion o exis ing ones. Gene ally, a slice is an abs ac ne wo k ha has o be ins an ia ed o add ess he in as uc u e esou ces. Se e al ins ances o he same slice a e possible, each o hem being cus omized ia a sui able alloca ion o NFs and/o hei pa ame e iza ion ou o a numbe o po en ial al e na i e NFs. No e ha a slice may be s a ically con igu ed, o dynamically ins an ia ed o e-con igu ed o e ime. A slice inally desc ibes an end- o-end ela ion, i.e. i s unc ionali ies also co e he 5G de ice pa . In p inciple, hi d pa ies (e.g. e icals, MVNOs, OTT se ice p o ide s) can c ea e and manage hei own ne wo ks ia he co esponding APIs (known as “any hing as a se ice”, XaaS). Figu e 4-6: Examples o 5G ne wo k slices [NGM15]. E en i NW slices a e seen as sepa a ed NWs (only on logical le el) i is s ill equi ed o e icien ly use common esou ces such as adio spec um, adio in as uc u e, and anspo be ween he slices. Only in special cases (based e.g. on egula o y and/o legal equi emen s) i is assumed ha dedica ed esou ces a e s a ically assigned o a single slice. The de ailed ealiza ion o he NW slicing concep in he 5G RAN is one o he esea ch ields in METIS-II ( isibili y o “ adio slices”, esou ce and QoS managemen , slice isola ion e.g. o secu i y pu poses, e c.). Acco ding o Figu e 4-6, CP and UP NFs o a slice can be lexibly placed acco ding o he needs o se ice cha ac e is ics (la ency, h oughpu , QoS, e c.) on he unde lying in as uc u e laye nodes conside ing he capabili ies o hose nodes w. . . anspo , p ocessing and s o age capaci y. One example is he suppo o low-la ency applica ions whe e some NFs may be Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 33 placed a he mobile edge, e.g. di ec ly on access nodes o al e na i ely in a Cloud-RAN node (see also [MET15-64]). E en i he SDN/NFV concep is so ening he sepa a ion be ween RAN and CN in a u u e 5G sys em, METIS-II s ill assumes a logical spli be ween he 2 pa s as a s a ing poin (see Figu e 4-7). Figu e 4-7: Possible logical 5G RAN a chi ec u e (se es as s a ing poin o he in es iga ions in METIS-II). This logical spli is seen as bene icial o he ollowing easons:  I allows o an independen e olu ion o RAN and CN unc ionali y;  I enables o make he CN unc ions o a leas a subse o hem access agnos ic (e.g. common UP p ocessing);  I acili a es mobili y since some CN unc ions (CP and/o UP) can be kep (ancho ed) when UEs mo e o a new RAN node;  A logical sepa a ion does no o bid c oss-laye op imiza ions in some deploymen s when he unc ions a e co-deployed;  A logical sepa a ion also allows mul i- endo CN/RAN in e ope abili y. METIS-II ini ially assumes ha he new CN/RAN in e ace, called he ein S1*, will ha e he S1 in e ace [3GPP15-36300] as s a ing poin bu wi hou necessa ily backwa d compa ibili y equi emen s. W. . . he CN, he ollowing unc ionali ies a e assumed o be co e ed:  Mobili y ancho ing (e.g. o in e -RAN node mobili y, including in e -RAT mobili y), add essing, oaming unc ionali y; Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 34  QoS con ol, cha ging, policy en o cemen , con ex awa eness;  A ach and au hen ica ion signaling including secu i y key gene a ion. An impo an aspec in 5G will be he enhanced in eg a ion o access echnologies on lowe laye s o he adio p o ocol s ack esul ing in op imized join esou ce managemen and inc eased pe o mance and eliabili y e.g. by applying mul i-connec i i y links on UP and/o CP ac oss di e en 5G ai in e ace a ian s. Whe eas he in eg a ion o e ol ed LTE-A (which is seen as pa o he 5G ai in e ace amily) wi h no el 5G ai in e ace a ian s is expec ed o happen on PDCP/RRC laye (o 5G equi alen ), a igh in eg a ion o no el a ian s may go down o he MAC laye . The inally achie able pe o mance will be also s ongly dependen on he de ice capabili ies, as e.g. mMTC de ices like senso s will – in con as o u u e-p oo ed sma phones o xMBB pu poses – p obably only suppo single ai in e ace a ian s due o cos easons (pa ly also addi ional es ic ion o a single equency band). Mo e sophis ica ed 5G de ices will be able o lexibly ac as an in as uc u e ne wo k node as well, e.g. a sel -backhauled, possibly nomadic access node [MET15-64]). W. . . in as uc u e deploymen aspec s, he key assump ions on he physical RAN a chi ec u e a e s a ed in he o m o ou ep esen a i e scena ios depic ed in Figu e 4-8. These scena ios ha e been chosen and de ined such ha hey e lec in a simpli ied o m he co ne cases o deploymen ha can be expec ed in he 5G ime ame (e.g. deploymen o no el 5G adio si es in addi ion o exis ing LTE-A si es, co-deploymen o LTE-A and no el 5G adio, wide-sp ead usage o sel -backhauling, s and-alone si es and baseband (BB) hos elling e c.) Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 35 Figu e 4-8: Physical RAN a chi ec u e scena ios conside ed. The scena ios a e de ailed in he ollowing:  Scena io 1 depic s a case whe e he same o di e en ai in e ace a ian s a e se ed om wo geog aphically sepa a ed si es, and whe e he connec ion be ween he si es and om he si es o he nex agg ega ion poin a e p one o non-ideal backhaul, which could o ins ance be modeled as in [3GPP14-36932]. Fo simplici y, i is assumed ha CN unc ionali y also esides a he agg ega ion poin , hence he e is no u he la ency in ol ed beyond he agg ega ion poin . As an op ion, mul iple ai in e ace a ian s (e.g. LTE-A and a no el 5G ai in e ace a ian ) could be se ed in a co-loca ed way.  Scena io 2 depic s he case whe e cen alized p ocessing is applied o all cells, and di e en spa ially sepa a ed access nodes a e se ed wi h ideal o non-ideal on haul, depending on he RAN unc ional spli ha is deployed.  Scena io 3 is simila o Scena io 2, bu conside s mul iple clus e s o cen alized p ocessing, each se ing mul iple adio si es ha a e assumed o be connec ed ia ideal on haul. The connec ions be ween he cen alized p ocessing clus e s, howe e , a e assumed o be p one o non-ideal backhaul. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 36  Scena io 4 builds upon Scena ios 1-3, bu now depic s a case whe e wo addi ional access nodes es ablish wi eless backhaul links o si es wi h wi ed backhaul. 4.2.2 E olu ions o he posi ioning o cu en playe s wi h 5G MNOs & MVNOs playe s:  Bo h playe s a e adi ional mobile ope a o s. They cu en ly p o ide a mobile se ice o cus ome s and manage his end cus ome ela ionship.  A mobile i ual ne wo k ope a o (MVNO) does no possess any spec um alloca ion o mobile access in as uc u e, wha e e he echnology used. MVNOs sign a con ac wi h a mobile ne wo k ope a o (MNO) ha holds a equency license and a mobile ne wo k in as uc u e. Table 4-1: Mobile ope a o s & MVNOs posi ioning. Playe Cu en posi ioning Possible e olu ion MNO Builds and ope a es he mobile ne wo k Manages subsc ibe ela ionships o bo h e ail and wholesale ma ke s Con en p o ision Gene aliza ion o he ixed adio access p o ision Mo e ocus on IoT and e ical ma ke s ( o coun e declining e enues on he consume ma ke ) Ge ing highe in he alue chain by p oposing IT / big da a se ices MVNO Buys capaci y o MNOs and manages subsc ibe ela ionships. In some cases MVNO ope a es pa o co e ne wo k May own SIM ca ds May ope a e mo e elemen s o he co e ne wo k O he access ne wo ks:  WiFi access p o ide s use unlicensed spec um and a e usually loca ed in la ge-scale enues whe e people a e likely o use high-speed in e ne (ai po s, co ee shops, ho els, uni e si ies...).Fo a mobile ope a o , a highe densi y o WiFi ho spo s also means mo e possibili ies o o load a ic on he ixed ne wo k and hus educe he a ic load on he adio access ne wo k.  LPWA (Low Powe Wide A ea) playe s use dedica ed echnology o low powe and low expe ienced use h oughpu in unlicensed spec um. Unsu p isingly, many di e en playe s a e s epping in he Low Powe Wide A ea/IoT ield. LPWA pu e p o ide s a e Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 37 ope a o s managing a ne wo k in as uc u e (SigFox). O he playe s (LoRa, Neul) p o ide access connec i i y (no he backhaul connec i i y as a adi ional playe ).  Sa elli e ope a o s use elecommunica ions sa elli es which a e usually placed in Geos a iona y Ea h O bi and Low Ea h O bi wi h a numbe o speci ic echnical equi emen s o elecom sys ems o ope a e (no ably la ge cons ella ions o small sa elli es, equency coo dina ion). Table 4-2: WiFi, LPWA, sa elli e playe s ole. Playe Cu en posi ioning Possible e olu ion WiFi access p o ide Ope a e WiFi access poin s Sells wholesale and e ail se ices Manages subsc ibe ela ionships Small cells ope a ion mainly on unlicenced spec um, wi h possible in eg a ion wi h 5G ne wo ks. Small cell as a se ice (SCaaS) LPWA (Low Powe Wide A ea) ope a o s Deploy and ope a e dedica ed ne wo ks o IoT Possible in eg a ion wi h 5G ne wo ks (cell si e sha ing, RAN in eg a ion? in eg a ion o unlicensed bands?) Sa elli e ope a o Limi ed p o ision o na owband mobile se ices wi h LEO cons ella ions (e.g. I idium, Inma sa ) Possible o e ing in on haul/backhaul in eg a ion wi h e es ial ne wo ks. Use o non-GEO cons ella ions wi h educed la ency. Mul icas app oach in combina ion wi h e es ial ne wo ks. Use o MTC in u al a eas RAN Ou sou cing playe s:  Towe companies a e in posi ion o go u he han hei cu en ole o en ing space o MNOs‟ base s a ions and could ope a e pa o he RAN.  Facili y manage s such as building o s adium manage s a e gene ally in ol ed in mobile co e age/capaci y h ough DAS (Dis ibu ed An enna Sys ems) ope a ion. They usually pay a specialized company o DAS ins alla ion.  U ban u ni u e manage s: Today, small cells a e being deployed on u ban s ee u ni u e such as s ee lamps, u ili y poles, bus s ops, benches o billboa ds h ough Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 38 pa ne ships be ween equipmen endo s and ad e ising companies. Deploymen o small cells in s ee u ni u e is expec ed o enhance co e age and ne wo k capaci y in highly popula ed a eas o mee he u u e cus ome demand.  Equipmen manu ac u e s: A ew yea s ago, elecom ope a o s acknowledged ha hey could no be p esen in e e y segmen o he alue chain and c ea e mo e alue by se ing up pa ne ships wi h la ge-scale playe s. This leads o inno a ions o changes in model o managemen such as, o example, ou sou cing passi e o ac i e in as uc u e o sha ing ixed o mobile ne wo ks. Equipmen manu ac u e s a e well posi ioned o he RAN ou sou cing ole due o hei expe ise in ha domain. Table 4-3: RAN Ou sou cing playe s ole. Playe Cu en posi ioning Possible e olu ion Towe company Build, own, ope a e, main ain owe asse s Build cus omized/g een si es Ren owe s o MNO ( ixed powe and uel cos / adi ional pass- h ough cos ) Could ope a e he RAN / pa s o RANs Facili y manage Ope a e DAS (mainly indoo en i onmen ) Could ope a e small cells (SCaaS) Could o e elecoms access in a building using mobile in as uc u e wi h backhauling ia a ibe op ic a he bo om o he building U ban u ni u e manage Build, own and ope a e u ban u ni u e Ren esou ces on a pay-as-you-go model Could ins all base s a ions, ope a e pa s o RAN (small cells) Equipmen manu ac u e s Ou sou cing and se ice managemen Ope a ion o wholesale mobile ne wo k IoT/MTC playe s:  In eg a o s & IT companies: Sys em in eg a o s o so wa e companies ha e s ong capabili ies in da a and in so wa e, especially in analy ics. As he alue chain is disassembling in many ma ke s and as ha dwa e is now commodi ized, in eg a o s e- assemble ce ain unc ions and p o ide packaged se ices in a “so wa iza ion” mo e. IT Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 39 companies a e mainly posi ioned on he p o essional ma ke s (Machine- o-Machine (M2M) and In e ne o Objec s) on he IT side and da a managemen h ough hei cloud se ices (based on hei da a cen e a ms).  Ve ical ma ke playe s encompass ca manu ac u e s in he au omo i e business, hospi als/doc o s in he heal h indus y…, ha will be pa o he enla ged elecom ecosys em in he nea e m as elecom is sp eading in many e icals. The e icals will de elop new se ices, new p ocesses and new business models, hanks o o e ed obus , a o dable ansmission and au oma ed collec ion o da a. Table 4-4: IoT/MTC playe s posi ioning. Playe Cu en posi ioning Possible e olu ion In eg a o s S ong capabili ies in da a and so wa e Connec i i y p o ision IT companies P o ide Cloud se ices Le e age on da a o p o ide be e se ices and solu ions Ve ical ma ke playe s Buy connec i i y om MNOs Play a ole in se ice c ea ion & ope a ion Public Sa e y and Public P o ec ion and Disas e Relie playe s (PPDR):  Cu en non-b oadband dedica ed public sa e y sys ems mainly use he 400 MHz and he 700-800 MHz bands wo ldwide. Spec um abo e 1 GHz suppo s also a a ie y o PPDR ope a ions o empo a y use only. A numbe o hese PPDR ne wo ks ha e been deployed all o e he wo ld. Table 4-5: Public sa e y playe s posi ioning. Playe Cu en posi ioning Possible e olu ion Dedica ed Public Sa e y Build, own and ope a e Public Sa e y / PPDR ne wo ks P o ide unused capaci y du ing non-c i ical hou s o ope a ion o 5G ope a o s Public Sa e y MVNO Sell mobile se ices o end-use s Ren capaci y o 5G ope a o s Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 40 Con en owne s, OTTs:  Con en owne s a e speci ically in ol ed in con en c ea ion and adap a ion o mobile de ices. They a e esponsible o he accu acy o he con en . The consump ion o con en has g adually e ol ed om a model mainly based on owne ship o con en (physical media o a digi al copy) o a model based on access o a ee (ad-suppo ed) and/o subsc ip ion se ice.  The well-known O e -The-Top (OTT) playe s a e se ing hei cus ome s o e he In e ne ( oice h ough VoIP and messaging a e he wo i s se ices p o ided). They a e p o iding an in eg a ed p oduc o e he In e ne and bypassing adi ional dis ibu ion. Se ices ha come o e he op a e, i no always, lowe in cos han he adi ional me hod o deli e y. Howe e , hey ha e lowe QoS han se ices p o ided by MNOs. Table 4-6: Con en owne s and OTT posi ioning. Playe Cu en posi ioning Possible e olu ion Con en owne s C ea ion o he con en , adap a ion o mul iple de ices No ole expec ed in adio access OTTs P o ision o se ices h ough he in e ne ee o cha ge MVNO p o iding mobile communica ions se ices O he playe s  Ob iously, de ice manu ac u e s a e designing and ma ke ing de ices di ec ly o use s and o elecom ope a o s. De ices a e s ill cen al o he elecom ma ke and a e con inuing o de elop in e m o be e pe o mances.  Today, in- ligh connec i i y on comme cial ai lines is p o ided h ough ai - o-g ound and sa elli e ne wo ks. In- ligh connec i i y is p o ided on a wholesale basis and on a e ail basis. Table 4-7: O he playe s posi ioning. Playe Cu en posi ioning Possible e olu ion De ice manu ac u e s Sell handse s o cus ome s and mobile ope a o s Wi h i ualized SIM ca d, de ice manu ac u e s could manage cus ome s ela ionship In- ligh connec i i y p o ide s Sell in- ligh connec i i y di ec ly o he use o h ough he ai line In- ligh connec i i y p o ide s could manage he use ela ionship widely Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 41 4.2.3 Iden i ica ion o new playe s in he 5G ield In i s 5G Whi e Pape [NGM15], NGMN desc ibes new business models expec ed wi h 5G. New business oles desc ibed in his documen make e e ence o asse p o ide , connec i i y p o ide and pa ne se ice p o ide . In his sec ion, we use he p e ious sec ion on he iden i ica ion o he cu en playe s and NGMN‟s inpu as a basis o iden i y new playe s/ oles in he 5G ield. Connec i i y p o ide s The business models associa ed o connec i i y p o ide s can be di e en ia ed be ween “basic” and “en iched” models and a e he ollowing:  Basic connec i i y p o ide s: in his model, only bes e o IP connec i i y is p o ided. This is he « dumb pipe » model o mobile ope a o and we can include WiFi access p o ide s in he same model. In he yea s o come, we migh see new playe s such as sa elli e se ice p o ide s, LPWA playe s, loons‟ playe s… Wi-Fi i s playe s such as Google wi h i s Google-Fi o e , which use WiFi as he p ima y connec ion op ion and swi ch o mobile ne wo k only as a “backup solu ion” could play a bigge ole in bundling o he access ne wo ks as well (sa elli e, LPWA, loons…). In he ene gy sec o , an example o basic connec i i y p o ide is an e olu ion o he MVNO concep called PVNO (P i a e Vi ual Ne wo k Ope a o ). The ene gy g ids long- e m needs a e no ul illed by exis ing mobile ne wo ks leading he playe s o his sec o o become MVNOs and o ake ull o pa ial con ol o a wi eless ne wo k. The PVNO could con ol elemen s o he co e ne wo k such as cus ome da abase and he SIM ca ds. In coun ies like he Ne he lands and F ance, u ili ies ha e buil cellula ne wo ks and we e awa ded spec um (in he 450 MHz band) o hei own needs.  Enhanced connec i i y p o ide s could inc ease ope a o di e en ia ion h ough ne wo k quali y/ eliabili y/mobili y/con igu abili y. Public Sa e y playe s, new MVNOs p o iding M2M en iched se ices ( o e ical sec o s, o secu i y pu poses…) could appea in his ield. The b oadcas ing sec o could also p opose a new model called TOo5G o „Towe O e lay o e 5G‟ in which he b oadcas ope a o would use i s HTHP (high owe high powe ) in as uc u e; he la e being al eady in place and se es o digi al e es ial ele ision (DTT) se ices. This dedica ed b oadcas ing in as uc u e would p o ide b oadcas and mul icas se ices (such as ideo s eaming) wi h lowe ansmission cos s han in unicas mode, bu he iabili y o his solu ion will depend on he 5G design choices (in eg a ion o DVB-like ai in e aces in 5G) and on he de elopmen o e ol ed MBMS (Mul imedia B oadcas Mul icas Se ices) solu ions. Asse p o ide s The asse p o ide ole co e s bo h ne wo k sha ing models and XaaS models. As explained in Sec ion 4.2.1, XaaS is a collec i e e m ha e e s o e e y hing as a se ice (o any hing as a se ice). Wi h XaaS, e e y hing can be accessed on demand ia he cloud. XaaS gi es a i s Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 48 Figu e 4-12: Pa ne se ice p o ide alue chain. 4.2.5 MTC Ac o s in Sma Ci y and he co esponding Value Ne wo k A sus ainable Sma Ci y is o en de ined as an inno a i e ci y ha uses In o ma ion and Communica ion Technology (ICT) and o he means o imp o e quali y o li e, e iciency o u ban ope a ion and se ices, and compe i i eness, while ensu ing ha i mee s he needs o p esen and u u e gene a ions wi h espec o economic, social and en i onmen al aspec s [ITU- FGSSC-15]. Focusing on he communica ion pa , he ole o ICT on one hand is o collec in o ma ion om he machines and send hem o he applica ions; and on he o he hand i should ans e he “sma ness” om applica ions o machines. Conside ing 5G as he pa adigm shi in ICT ha is supposed o enable Sma Ci ies, his sys em should include many high echnical equi emen s ha a e highly in eg a i e wi h o he indus ies. This in eg a ion hen happens ia Machine o Machine (M2M) communica ion solu ions. The ole o M2M solu ions is hen o sense, analyze and in eg a e he key in o ma ion o co e sys ems in ci ies. As a esul , he ele ance o ICT in Sma Ci ies is wo old: i s , how o enable a Ho izon aliza ion pla o m o o he indus y e icals and second, in eg a ing ICT in as uc u e in o he indus ies in ol ed. Looking in o he Sma Ci y ecosys em, he ne wo k o supplie s [LGM-15] includes ICT p o ide s and subsequen ly M2M p o ide s. Conside ing M2M echnologies as he enabling ICT Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 49 ool o ci ies o become sma , hen MTC would be he pa whe e Cellula Telecommunica ion Ne wo ks come in o he play. This highligh s he ole o Telecom ac o s in Sma Ci ies. A he same ime, hese changes may no occu unless he new alue chain and alue ne wo k can hos speci ic demands and new ac o s o Sma Ci y [GAM-15]. Since MTC co esponds o u ilizing cellula echnologies as he access ne wo k o M2M se ices, adi ional ac o s in he mobile elephony alue ne wo k a e iably ac i e he e as well. Mobile Ne wo k Ope a o s (MNO) as he ypical ca ie s ha con ol and ope a e cellula ne wo ks a e capable o ope a ing he MTC ne wo k. Telecom Equipmen Vendo s (TEV) as he adi ional manu ac u e s o he elecommunica ion equipmen , ypically p o ision he echnical p ocu emen o he MNOs. Bu , acco ding o he shi in he alue chain, he TEVs ha e ecen ly pa icipa ed in di e en oles ha his o ically ha e been assigned o aken ca e o by o he s; such as MNOs. E en oles like p o isioning new demands such as Connec i i y Pla o ms a e now being p o isioned by some TEVs. Figu e 4-13 illus a es he majo business ela ions o MNOs and TEVs in he Mobile Telephony case. A Managed Se ice P o ide (MSP) is ypically an en i y ha o e s end- o-end solu ions; such as ne wo k ope a ion managemen in his case. Based on p oposed MTC Ac i i ies, Resou ces, Ac o s, and mos impo an ly he amewo k in oduced in Figu e D-10 (in Appendix D1.2), i e majo g oups o ac o s can be iden i ied in he MTC alue ne wo k. Besides he End Use s (EU), hese ac o s a e he mos likely en i ies who can own ei he o he esou ces men ioned ea lie in o de o pe o m MTC ac i i ies. These ac o s a e: a) End Use , b) Se ice P o ide , c) MTC ne wo k ope a o , d) De ice P o ide , and e) MSP. Figu e 4-13 Two adi ional Value Ne wo ks o Mobile Telephony. Acco ding o he cases s udied ea lie , we showed ha a he han adi ional Telecom ac o s (i.e. MNOs and TEVs), he e a e o he ac o s who migh be e en mo e compe en in p o isioning any o he ac i i y blocks o MTC. Fo ins ance, a specialized M2M cellula ne wo k ope a o (MTC ne wo k ope a o ) can be conside ed a be e op ion o p o ision MTC ne wo k. Se ice P o ide s o M2M solu ions also in some cases ake con ol o he en i e alue chain by handling he EU; a p e iously dominan posi ion o MNOs in Mobile Telephony (Figu e 4-13 - Righ ). On he o he hand, TEVs and MNOs ha e shown in e es in di e en ac i i y blocks. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 50 Ano he majo ac o in his se up is hen an en i y which pe o ms he ole o p o isioning CDP. I can be seen ha his ac i i y is mainly pe o med by he i ms who ha e a his o y in p o isioning connec i i y in he sense o au oma ing connec ed de ices. Some examples can be ei he ou sou cees o ne wo k ope a ions o MNOs (MSPs) o he ones which ha e been ac i e in au oma ion o indus y e icals (e.g. Gene al Elec ics, Siemens, e c.). Figu e 4-14 Two majo se ups o MTC Value Ne wo k. Figu e 4-14 illus a es wo majo se ups o he MTC alue ne wo k in Sma Ci ies. Acco ding o ou desc ip ion o he MTC ac i i ies, elecom ac o s a e capable o pe o ming mul i ude o ac i i ies in MTC alue ne wo k in Sma Ci ies, bu based on hei esou ces. This di ec ly conce ns he compe ences hey can acqui e and/o ha e. Conside ing “connec i i y domain” as hei main playg ound, p o isioning AEP is also an ac i i y being pe o med by some TEVs in ecen yea s. This way, elecom ac o s mainly co espond o ei he MTC ne wo k ope a o s and/o MSPs (suppo ing ole o p o isioning AEP, CDP). An in e es ing obse a ion he e is he absence o MNOs on “owning” he End Use . In e ms o esou ces, i is impo an o conside ha he ac o /s who owns he end use -as a esou ce- is he mos likely o ha e con ol o e his alue ne wo k. This alue ne wo k should be conside ed when deploying and de eloping 5G sys ems, as i should allow collabo a i e se ups o happen. Finally, i could be concluded ha 5G and ICT will play he enable -suppo ole o making Sma Ci ies happen and no much mo e; so would he elecom ac o s. 4.2.6 E olu ions o he MNO-cen ic alue Ne Ha ing iden i ied, in he p e ious sec ion, he main ac o s and he in e ac ions be ween hem, we ocus in his sec ion on he MNOs. We cons uc he alue ne o hese MNOs and i s e olu ion wi h 5G; he aim being o iden i y hei coope i ion ela ions wi h he o he ac o s. The alue ne model has been elabo a ed by [BRAN96]. This model is a complemen a y app oach o he alue chain amewo k. The objec i e o alue chain and alue ne models is nea ly he same, i.e. iden i y he main playe s on ma ke and he main ac i i ies hey conduc . Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 51 They gi e a gene al o e iew o an indus y du ing a speci ic pe iod o ime. Bu he logic is qui e di e en be ween he wo models. In he case o alue ne , he analysis is mo e comp ehensi e as he main playe s ha e o be in eg a ed in ou ca ego ies: cus ome s, supplie s, compe i o s and complemen o s ollowing a e ical and ho izon al dimension. The ho izon al axis is ela ed o compe i ion and coope a ion issues. Figu e 4-15 shows a gene ic alue chain cen e ed on a gi en company. No e ha playe s can co e – and o en do co e - mo e han one ole in he alue ne . Figu e 4-15: A gene ic alue ne . Figu e 4-16 gi es he cu en alue ne o MNOs. We i s begin by de ining cus ome s as his will gi e us a clea iew abou he posi ioning o he MNO. Two kinds o cus ome s a e iden i ied:  End use s, be hey in he mass ma ke (indi iduals), o o he business cus ome s (p i a e companies o public adminis a ions). Clea ly, hese cus ome s a e con ac ing wi h he MNO as a se ice p o ide .  MVNOs: hey a e cus ome s o he MNO as hey buy he igh o use his ne wo k in o de o se e hei cus ome s. When he MNO sells ne wo k access igh s o MVNOs, he is beha ing as a ne wo k ope a o . Based on his analysis, we can see ha cus ome s g oups can be classi ied in o wo g oups: cus ome s o he MNO as a se ice p o ide and cus ome s o he MNO as a ne wo k ope a o . We will keep his classi ica ion o he es o ou analysis o he alue ne . Nex , we s ay in he e ical dimension and iden i y supplie s. As a ne wo k ope a o , he MNO has he ollowing supplie s:  In as uc u e endo s ha supply him wi h base s a ions, ne wo k con olle s, e c. Examples o such endo s a e E icsson, Alca el Lucen , Huawei, e c. Cus ome s Company Supplie s Complemen o s Compe i o s Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 52  In as uc u e supplie s ha p o ide access o owe s, backhaul capaci y and o he esou ces equi ed o ne wo k ope a ion. These esou ces a e o en sha ed esou ces and en ed o e en p o ided “as a Se ice”.  Ne wo k ope a ion & managemen so wa e make s om whom he MNO can buy ne wo k moni o ing and managemen ools allowing him o p ese e quali y o i s ne wo k. No e ha in as uc u e endo s ha e also hei own so wa e solu ions, bu ope a o s whose ne wo ks a e mul i- endo need hi d pa y so wa e solu ions. On he o he hand, as a se ice p o ide , he MNO has as supplie s de ice manu ac u e s (Apple, Samsung, Nokia, e c.). Indeed, se ice p o ide s usually buy de ices om manu ac u e s and sell hem a lowe p ices o end use s. Le us now mo e o he ho izon al dimension and iden i y compe i o s. As a se ice p o ide , each MNO has as ob ious compe i o s all o he se ice p o ide s, be hey MNOs o MVNOs. O e he op (OTT) playe s, like Skype and Vibe , a e also seen as compe i o s o he MNO as a se ice p o ide as hey p opose subs i u ion se ices ( oice, ideo con e ences). As a ne wo k ope a o , he conside ed MNO has as compe i o s he o he MNOs as hey o e he same se ices o MVNOs. The mos di icul ask is o iden i y complemen o s whose p esence inci es cus ome s o buy mo e se ices om he MNO. Ob iously, con en p o ide s (online game de elope s, Google maps, TV channels) ac as complemen o s as people a e willing mo e o buy mobile da a access in o de o bene i om hei a o i e con en s e e ywhe e. De ice manu ac u e s a e also complemen o s as end use s conside sma phones and able s as aluable de ices by hemsel es, and a sma phone o a able will be mo e use ul wi h a wi eless In e ne connec ion. De ice applica ion de eloping indus y is also a complemen o as he mul i ude o sma phone applica ions inci es use s o buy a sma phone and o subsc ibe o a mobile da a connec ion. No e ha we do no make a dis inc ion be ween complemen o s o he MNO as se ice p o ide o as ne wo k ope a o as hey a e gene ally he same ( hey s imula e he need o a ne wo k access). Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 53 Figu e 4-16: Cu en alue ne o MNOs. Figu e 4-17 shows he e olu ion o he alue ne o MNOs wi h 5G, based on he 5G playe s iden i ica ion in he p e ious sec ions. We s a wi h he e olu ion o he g oup o cus ome s whe e PVNOs join MVNOs as cus ome s o he MNO as a ne wo k ope a o , and whe e e icals, when hey buy di ec ly connec i i y o hei cus ome s, become cus ome s o he MNO as a se ice p o ide . The same e icals become complemen o s as, by mo ing owa ds mo e connec i i y, hey p o ide needs o people (indi iduals and p o essionals) o 5G se ices. As o he supplie s o he MNO, he inc eased he e ogenei y and he i ualiza ion o ne wo ks a e expec ed o di e si y hei lis . The lis s o equipmen endo s and o ne wo k ope a ion and managemen so wa e supplie s a e joined by classical IT companies like IBM, HP, e c., which p o ide p ocessing se e s and i ual ne wo k so wa e (e.g. based on SDN/NFV). Da a cen e playe s may play a ole in managing hos els o BBUs in his con ex , especially o Cloud-RAN a chi ec u es. Asse p o ide s like acili y manage s, u ban u ni u e manage s and owe companies a e expec ed o ha e a la ge ole in he deploymen and he managemen o pa s o he access ne wo k, ein o cing hei posi ion as supplie s o he MNO as a ne wo k ope a o . Wi h he e olu ion o spec um egula ion and he alloca ion o new bands unde inno a i e au ho iza ion schemes (e.g. LSA and LAA), spec um b oke s could play a ole in he u u e and manage spec um esou ces on behal o mobile ne wo k ope a o s in o de o allow eal- ime managemen o he spec um. Finally, as a se ice p o ide , he MNO can make deals wi h Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 54 CDN playe s o con en hos ing nea end use s a he ne wo k edge, making hem supplie s wi h ega ds o his ole as a se ice p o ide . Finally, in he g oup o compe i o s, he ad en o new LPWA ne wo ks and o a ious access ne wo ks based on sa elli es and loons in addi ion o he inc eased in eg a ion o WiFi e olu ions wi hin he 5G ne wo k in oduce a a ie y o playe s as compe i o s o he MNO in he RAN. A possible scena io, as discussed p e iously, is he eme gence o la ge WiFi playe s in he bundling o hese a ious access ne wo ks. Rega ding he se ice p o ide ole o he MNO, PVNOs and MTC ope a o s join MVNOs as compe i o s o o e ing se ices o end use s. 5G is a ge ing o open up he ecosys ems. One o he main ools allowing o suppo PVNOs is he concep o Ne wo k Slicing b ough o wa d by NGMN. I allows PVNOs o ope a e and manage a “ne wo k slice” as an independen i ual ne wo k allowing hem o in oduce hei domain speci ic business cases and o ollow domain speci ic egula ion. Figu e 4-17: E olu ion o he alue ne o MNOs wi h 5G. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 55 4.3 Focus on he dense u ban in o ma ion socie y use case The aim in his sec ion is o iden i y he possible alue ne wo ks o he p o isioning o he a ge ed mobile se ices in he u u e dense u ban in o ma ion socie y, as desc ibed in Sec ion 2.1.1. In he i s pa we will ocus on decons uc ing mobile se ices p o isioning in o a numbe o business oles in iew o he discussed e olu ion pa h o he adio access ne wo k (RAN) A chi ec u e in sec ion 4.2.1, while he second pa is dedica ed o iden i y he possible business ela ionships and coope a ion models be ween he ac o s in he xMBB ecosys em. 4.3.1 E olu ion in RAN a chi ec u e and deploymen op ions The Radio Access Ne wo k (RAN) a chi ec u e is he shi owa ds mo e lexible, dynamic and he e ogeneous a chi ec u e as desc ibed scena ios in Sec ion 4.2.1 and Figu e 4-8. In hese scena ios, he p o ision s a egy o he u u e mobile se ice can be based as a se o ela i ely au onomous esou ces ha can be owned by di e en ac o s. In his con ex , numbe deploymen s a egies a e o be conside ed o he p o isioning o he mobile se ices in u u e dense u ban in o ma ion socie y. In his sec ion, ou deploymen s a egies ha e been chosen based on he RAN a chi ec u e scena ios in Figu e 4-8 and conside ing di e en combina ions o ou doo and indoo ne wo ks deploymen s as ollows:  S a egy A (mac ocells wi h no o limi ed numbe o ou doo small cells): he deploymen o a dense laye o adi ional mac o cell si es and ou doo small cells based on exclusi e spec um use below 6 GHz along wi h use o sha ed spec um esou ces below 6 GHz (i.e. based on di e en au ho iza ion and access op ions such as LSA and/o common au ho iza ion). In o he wo ds, he RAN a chi ec u e can be depic ed as numbe o s andalone mac ocell adio si es ha may be complemen ed by a numbe o ou doo small cells o co e age and capaci y - As an op ion, mul iple adio ai in e ace a ian s (e.g. LTE-A and a no el 5G ai in e ace a ian ) could be se ed in a co-loca ed way. Re e ing o Sec ion 4.2.1, all physical RAN a chi ec u e scena ios (1 o 4) a e sui able o his deploymen s a egy. -  S a egy B (mac ocells wi h massi e deploymen o indoo small cells): he deploymen o he e ogeneous RAN a chi ec u e based on he desc ibed mac ocell in as uc u e in S a egy A along wi h massi e numbe o indoo small cells in public a eas. These indoo small cells may be ope a ed in he sha ed spec um esou ces below 6 GHz (i.e. based on di e en au ho iza ion and access op ion such as LSA and/o common au ho iza ion).  S a egy C (massi e deploymen o ou doo small cells): he deploymen o a dense laye o ou doo small cell si es based in mul iple equency bands and au ho iza ion op ions. In o he wo ds, he ou doo small cells may be ope a ed based on exclusi e and sha ed spec um use conside ing equency bands below and abo e 6 GHz. He e, he Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 56 RAN a chi ec u e can be depic ed as massi e numbe o low-cos and low-powe nodes ha a e suppo ed wi h a cen alized p ocessing capabili y in he cloud. In o he wo ds, all he signals a e sen back ia ideal o non-ideal backhaul links om hese ou doo small cells o he cloud.  S a egy D (massi e deploymen o indoo and ou doo small cells): he deploymen o a dense laye o small cell si es e e ywhe e (bo h indoo and ou doo in public a eas). In o he wo ds, he RAN a chi ec u e can be depic ed as massi e numbe o s andalone ou doo small cell si es as desc ibed in S a egy C along wi h he deploymen o massi e numbe o indoo small cells. He e, he ou doo and indoo small cells may be ope a ed based on exclusi e and sha ed spec um use conside ing highe equency bands abo e 6 GHz. Wi h mo e owa ds mo e lexible, dynamic and he e ogeneous RAN a chi ec u es, he p o isioning o he mobile se ices is g adually sepa a ed om he owne ship o he physical ne wo k in as uc u e as well as om he necessa y ac i i ies o ope a e and main ain he di e en ne wo k elemen s. Hence, he u u e RAN a chi ec u es can be depic ed as a se o ela i ely au onomous oles ha can be pe o med by di e en ac o s. Ye , hese au onomous oles need o be managed based on a amewo k o common p inciples and SLAs o p o ision he a ge ed mobile se ice which open he doo o new business ole. In his espec , i e key business oles can be iden i ied in he u u e mobile se ices ecosys em conside ing he discussed e olu ion pa h o he adio access ne wo k (RAN) A chi ec u e in sec ions 4.2.2 and 4.2.3; namely:  Se ice P o ide : example MVNO can use sub-se o he connec i i y p o ide ne wo k capabili ies based on ce ain SLA o p o ision o e op se ices. He e he se ice p o ide can con ol and manage he ela ionship wi h end-cus ome s.  Connec i i y P o ide : o e ing only B2C connec i i y: Basic connec i i y in ol es bes e o IP connec i i y o B2B capaci y wholesale. He e he connec i i y p o ide can con ol and manage he ela ionship wi h end-cus ome s (B2C).  Asse P o ide s: Ren esou ces on a pay-as-you-go model, In as uc u e as se ices (indoo o ou doo adio base s a ion, agg ega ion node, backhaul link e c.), Spec um as a Se ice e c.  Resou ces B oke : Suppo secu e and seamless esou ces sha ing among di e en Se ice p o ide s, connec i i y Se ice P o ide s, Asse P o ide s. In his con ex , he esou ces de ini ions can be ex ended o include di e en passi e in as uc u e along wi h di e en ac i e ha dwa e and so wa e‟s esou ces. This may include he wi eless ne wo k in as uc u es such as ixed and nomadic adio access nodes and he p ocessing capabili y o hese nodes, backhauling links and he essen ial spec um esou ces. Mo eo e , ce ain co e and anspo ne wo k ope a ion and managemen unc ionali ies can in a simila way be abs ac ed o a esou ce. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 57  Managed Se ice P o ide (MSP): Wi eless ne wo k planning, deploymen and O&M ac i i ies. A hi d pa y such as an in as uc u e endo willing o o e addi ional managed se ices o ei he MNOs o MVNO. E icsson no ably made an announcemen a he Mobile Wo ld Cong ess ela ed o i s own SCaaS p oposi ion. O he playe s such as municipali ies o eal es a e owne s could as well decide o make he in es men o os e be e capaci y while mone izing he in es men . This playe is expec ed o play a ole in public a eas ypically, like shopping malls and ai po s. 4.3.2 Mapping ac o s o he oles: xMBB alue ne wo k Conside ing he iden i ied deploymen scena ios and business oles in sec ion 4.3.1.1, an ac o can play speci ic business oles based on his abili y o con ol o e asse s and con ol o e end- cus ome s. Mo eo e , a playe can pe o m mo e han one business ole i clea s a egic bene i s can be iden i ied as shown in Table 4-9. In his con ex , asse s such as he wi eless ne wo k in as uc u es and he essen ial spec um esou ces can be owned wi h di e en playe s wi h di e en business adi ions. Table 4-9: Mapping ac o s o he oles. Mobile Ne wo k Ope a o (MNO) S a egy A S a egy B S a egy C S a egy D Op ion 1: Ac as Connec i i y and Se ice P o ide by building and ope a ing he mobile ne wo k in as uc u e. He e, he MNO will con ol all he business ela ionships wi h he end-subsc ibe s (B2C). Mo eo e , he MNO may ou sou ce he ne wo k O&M ac i i ies o a MSP. Op ion 2: Ac as Connec i i y P o ide and Asse s P o ide s by selling he ne wo k capaci y o i ual Ope a o s such as MVNOs and PVNOs. Op ion 1: Ac as Connec i i y and Se ice P o ide by building and ope a ing he mobile ne wo k in bo h indoo and ou doo loca ions. He e, he MNO may ou sou ce he ne wo k O&M ac i i ies o a MSP. Op ion 2: Ac as Se ice P o ide only in ce ain indoo loca ions and pu chase ne wo k capaci y om an Asse s P o ide who owns indoo ne wo k in as uc u e. Op ion 1: Ac as Connec i i y and Se ice P o ide by building and ope a ing he mobile ne wo k in as uc u e. He e, he MNO will con ol all he business ela ionships wi h he end-subsc ibe s (B2C). He e, he MNO may ou sou ce he ne wo k O&M ac i i ies o a MSP. Op ion 2: Ac as Connec i i y P o ide and Asse s P o ide s by selling he ne wo k capaci y o i ual Ope a o s such as MVNOs. He e he MNO can ha e a B2B Op ion 1: Ac as Connec i i y and Se ice P o ide by building and ope a ing he mobile ne wo k in bo h public indoo a eas and ou doo loca ions. He e, he MNO may ou sou ce he ne wo k O&M ac i i ies o a MSP. Op ion 2: Ac as Se ice p o ide only in ce ain indoo loca ions and pu chase ne wo k capaci y om an Asse s P o ide who owns indoo ne wo k in as uc u e. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 64 [iJo14-D52] ICT-317941 iJOIN, “Final De ini ion o iJOIN Requi emen s and Scena ios”, No embe 2014. [ITU11] In e na ional Telecommunica ion Union – Radio (ITU-R), “Assessmen o he global mobile b oadband deploymen s and o ecas s o In e na ional Mobile Telecommunica ions”, ITU-R epo M.2243, 2011. [ITU11b] In e na ional Telecommunica ion Union – Radio (ITU-R), “The ou come o he e alua ion, consensus building and decision o he IMT-Ad anced p ocess (S eps 4 o 7), including cha ac e is ics o IMT-Ad anced adio in e aces”, ITU-R Repo M.2198, No embe 2011. [ITU15] In e na ional Telecommunica ion Union – Radio (ITU-R) Wo king Pa y WP 5D: D a New Recommenda ion “IMT Vision - F amewo k and o e all objec i es o he u u e de elopmen o IMT o 2020 and beyond”, Doc. R12-SG05-C-0199 (app o ed by S udy G oup SG5), June 2015. [ITU-2083] Recommenda ion ITU-R M.2083-0, IMT ision – amewo k and o e all objec i es o he u u e de elopmen o IMT o 2020 and beyond, Sep . 2015. [ITU-2290] Repo ITU-R M.2290, Fu u e spec um equi emen s es ima e o e es ial IMT, Dec. 2013. [ITU-2730] Repo ITU-R M.2730, “IMT a ic es ima es o he yea s 2020 o 2030”, July 2015. [ITU-FGSSC-15] ITU-T Focus G oup on Sma Sus ainable Ci ies, “Se ing he s age o s akeholde s‟ engagemen in sma sus ainable ci ies” Ma ch 2015. [KRB15] Chang Wook Kim, Jihoon Ryoo and Milind Madha Buddhiko , “Design and Implemen a ion o an End- o-End A chi ec u e o 3.5 GHz Sha ed Spec um,” 7 h IEEE Dynamic Spec um Access Ne wo ks (DySPAN), S ockholm, Sweden, Sep29-Oc 02 2015. [LGM-15] And es Laya, Ami hossein Ghanba i and Jan Ma kendahl “Tele-Economics in MTC: wha numbe s would no show” EAI Endo sed T ansac ions on IoT, Oc obe 2015. [MAM15-D11] ICT- 619086, “Sys em scena ios and equi emen s speci ica ions”, Janua y 2015. [MCN13-D21] ICT – 318109, “Re e ence Scena ios and Technical Sys em Requi emen s De ini ion”, Ap il 2013. [MM14] S.-S. Man ed and N. Dmi y, “On m2m so wa e pla o ms”, In e na ional Jou nal o Open In o ma ion Technologies, ol. 2, no. 8, 2014. [MET13-D11] ICT-317669 METIS, Deli e able 1.1 Ve sion 1, “Scena ios, equi emen s and KPIs o 5G mobile and wi eless sys em”, Ap il 2013. [MET15-D15] ICT-317669 METIS, Deli e able 1.5 Ve sion 1, “Upda ed scena ios, equi emen s and KPIs o 5G mobile and wi eless sys em wi h ecommenda ions o u u e in es iga ions”, Ap il 2015. [MET14-D53] ICT-317669 METIS, Deli e able 5.3 Ve sion 1, “Desc ip ion o he spec um needs and usage p inciples”, Augus 2014. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 65 [MET15-D54] ICT-317669 METIS, Deli e able 5.4 Ve sion 1, “Fu u e spec um sys em concep ”, Ap il 2015. [MET15-D64] ICT-317669 METIS, Deli e able 6.4 Ve sion 2, “Final epo on a chi ec u e”, Janua y 2015. [MET15-D66] ICT-317669 METIS, Deli e able 6.6 Ve sion 1, “Final epo on he METIS 5G sys em concep and echnology oadmap”, May 2015. [MET15-R31] ICT-671680 METIS-II, Repo 3.1 Ve sion 1, “P elimina y spec um scena ios and jus i ica ion o WRC Agenda I em o 5G bands abo e 6 GHz”, Oc obe 2015. [MiW15-WP] “MiWa eS Whi e Pape - He e ogeneous Wi eless Ne wo k wi h mmWa e Small Cell Access and Backhauling”, Janua y 2015. [MiW13-D11] ICT-608637, “De ini ion o Scena ios and Use Cases”, Decembe 2013. [MMK+11] C. F. Mecklenb äuke , A. F. Molisch, J. Ka edal, F. Tu esson, A. Paie , L. Be nado, T. Zemen, O. Klemp, and N. Czink., “Vehicula Channel Cha ac e iza ion and I s Implica ions o Wi eless Sys em Design and Pe o mance”, P oceedings o he IEEE, ol. 99, no. 7, pp. 1189-1212, July 2011. [MOT13-D21] ICT- 317959, “Use Cases and Requi emen s”, Feb ua y 2013. [NGM10] NGMN Alliance, “Use da a a es in mobile da a ne wo ks”, May 2010. [NGM15] NGMN Alliance, “NGMN 5G Whi e pape ”, Feb ua y 2015. [S 11] E. G. S öm, “On medium access and physical laye s anda ds o coope a i e in elligen anspo sys ems in Eu ope.”, P oceedings o he IEEE, ol. 99, no. 7, pp. 1183–1188, July 2011. [T o13-D21] ICT-318784, “Scena ios and Requi emen s”, July 2013. [We ne 08] M. We ne , P. Mobe g, and P. Skille ma k. “Cos assessmen o adio access ne wo k deploymen s wi h elay nodes”, In ICT-MobileSummi , 10-12 June, S ockholm, Sweden, 2008. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 66 A Key Pe o mance Indica o s (KPIs) This Appendix is based on [MET13-D11] and p o ides an o e iew o he Key Pe o mance Indica o s (KPIs) iden i ied o assess he pe o mance o he echnical solu ions de i ed wi hin METIS-II and he i e use cases de ined in Sec ion 2. Due o he wide a ia ion o he en i onmen al condi ions in he di e en use cases he e is also a co esponding sp ead o he KPI alues which has o be aken in o accoun , i.e. a single KPI alue does no usually i all use cases. The KPIs aken as basis o assessmen o he adio link ela ed equi emen s a e as ollows:  T a ic olume densi y,  Expe ienced end-use h oughpu ,  La ency,  Reliabili y,  A ailabili y and e ainabili y,  Ene gy consump ion (e iciency),  Cos (CAPEX, OPEX). The KPI de ini ions a e gi en wi hin his sec ion. Each KPI has a quali a i e de ini ion, mo i a ion, and ma hema ical de ini ion. The equi emen s on each o he QoS KPIs (e.g. expe ienced end-use h oughpu and la ency) ha e o be ensu ed in space and ime ollowing he a ailabili y and eliabili y equi emen s o he co esponding use case. A.1 T a ic olume densi y Quali a i e de ini ion: The a ic olume densi y desc ibes he o al use da a olume ans e ed o/ om end-use de ices (measu ed on he in e ace be ween Laye 2 and 3) du ing a p ede ined ime span di ided by he a ea size co e ed by he adio nodes belonging o he RAN(s). Fo mul i-hop solu ions each use da a is only coun ed once. Mo i a ion: This KPI is di ec ly ela ed o he gene al 5G goal o suppo a 1000 imes highe a ic olume densi y han oday‟s ne wo ks [MET13-D11] [5GPPP15]. This a ge igu e is expec ed due o s ongly inc easing numbe o mobile de ices wi h high da a a e capabili ies and a ic demand o mul imedia-based se ices wi h ising sha e o ul a-high- esolu ion ideo du ing nex decade [CISCO] [ITU11] [ITU-2730]. Figu e A-1 schema ically shows he expec ed exponen ial g ow h o he mobile a ic olume o e ime o a gi en a ea. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 67 Figu e A-1: Exempla y expec ed inc ease in a ic olume. Today‟s legacy adio ne wo ks in Eu ope, based on a mix u e o GSM, UMTS and LTE as well as o WiFi o ho spo scena ios, will no be able o cope wi h he expec ed a ic olume inc ease, e en hough u he adio echnology imp o emen s will come wi h la es eleases o 3GPP (known as LTE-Ad anced P o). The a ic olume densi y is mos ly in e es ing o he use cases which conside xMBB se ice ypes, i.e., aken in o accoun high use densi ies as well as loca ions wi h high use da a demands. Ma hema ical de ini ion: The a ic olume densi y in a ne wo k can be gene ally compu ed by he sum o a ic olumes each p oduced and consumed by an end-use de ice o e a p ede ined ime span, possibly di e en ia ed be ween downlink and uplink di ec ion, di ided by he ime span and by he o e all se ice a ea co e ed by he co esponding adio nodes. I has o be no ed ha he a ic olume densi y is usually s ongly co ela ed wi h he en i onmen and co esponding use densi ies. E.g. in dense u ban a eas wi h many la ge buildings and high use pene a ion he densi y will be much highe han in u al a eas, see g aphical illus a ion in Figu e A-2 (le diag am). In addi ion, he ins an aneous densi y will a y o e he day wi h he a ic busy pe iod being mos impo an ime slo o s a is ical e alua ion, illus a ed in Figu e A-2 ( igh diag am). The e o e he a ia ion in ime and space has been aken in o accoun a he de ini ion o he KPI alue o he di e en use cases. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 68 Figu e A-2: Exempla y dependency o a ic olume on en i onmen (le ) and day ime ( igh ). A.2 Expe ienced use h oughpu Quali a i e de ini ion: The expe ienced use h oughpu is he da a h oughpu an end-use de ice achie es on he in e ace be ween Laye 2 and 3 (use plane only) a e aged du ing a p ede ined ime span. This me ic is one possible measu e o he quali y o expe ience (QoE) le el a use expe iences o he se ice applied. Howe e , he da a a e o he se ice applica ion i sel is lowe han he expe ienced use h oughpu as addi ional p o ocol o e head and/o a ic con ol on highe laye s, e.g., IP, TCP/UDP/SCTP. The expe ienced use h oughpu depends on he use case en i onmen , bu also on he numbe o use s and he amoun o da a hey gene a e, because hey a ec he cell load and in e e ence om su ounding cells in a adio ne wo k. Mo i a ion: The KPI is di ec ly ela ed o he 5G goal o achie e a 10 o 100 imes highe ypical use da a a e compa ed wi h oday‟s sys ems [MET13-D11] [5GPPP15]. In p inciple i is ele an o all 5G use cases conside ing xMBB se ice ypes. Ma hema ical de ini ion: Le he - h packe (a he in e ace be ween Laye 2 and 3) o he - h use ha e a size bi s. I s dis ibu ion depends on he applica ion. Le be he E2E delay (one ip delay measu ed be ween he in e aces men ioned be o e) o deli e ing he packe s o he des ina ion. I depends on e.g. he RAN solu ion, use posi ion ( adio condi ions), and schedule load. The h oughpu o his packe is . Fo ins ance, i a packe o 1.25 Mby e is deli e ed in 1 s, hen he packe h oughpu is 10 Mbps. The expe ienced use h oughpu is compu ed as he expec ed packe h oughpu : , whe e he expec a ion is aken o e a ime span which is speci ic o he applica ion o use case. Fo ins ance, i could e e o all packe s belonging o a session o when he use is in a p ede ined loca ion. A dis ibu ion o he expe ienced use h oughpu can be de i ed o each use case, based on assumed a ic olume and e alua ed RAN solu ion (p obably di e en ia ed Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 69 be ween DL and UL). Depending on inal se ice a ailabili y (o co e age) de ined in he use cases, he KPI alue may be de ined, e.g. as he 5 h pe cen ile o his dis ibu ion, i.e. 95% co e age o a ce ain da a a e, o as he 1 h-pe cen ile, i.e. 99% co e age. The expe ienced use h oughpu de ined he e is di e en o alues known, e.g. om 3GPP and ITU-R e alua ions on LTE-Ad anced and IMT-Ad anced sys ems whe e use h oughpu cu es a e usually de i ed o homogeneous hexagonal mac o cell s uc u es wi h ixed numbe s o equally-dis ibu ed use s in each cell based on a ull bu e app oach o each use queue, i.e. 100% ne wo k load. In eal mobile adio ne wo k scena ios he expe ienced use h oughpu is di e en due o ypically lowe ne wo k load aken in o accoun he use ‟s andomized access o sha ed adio esou ces acco ding o hei se ice applica ions. NGMN p oposed in [NGM10] o apply ile ans e s as a basis o simula o e alua ions. 3GPP p oposed in [3GPP10-36814] a simila model o enable use h oughpu assessmen a a ious ne wo k loads. In con as o hose me hods he mo e gene al app oach applied in METIS-II can also be used in he e ogeneous ne wo ks wi h unequal dis ibu ion o si es, use s and a ic. A.3 La ency Quali a i e de ini ion: Di e en ypes o la ency a e ele an o di e en applica ions. E.g. he E2E la ency, o one ip ime (OTT) la ency, e e s o he ime i akes om when a da a packe is sen om he ansmi ing end o when i is ecei ed a he ecei ing end. Ano he la ency measu e is he ound ip ime (RTT) la ency which e e s o he ime om when a da a packe is sen om he ansmi ing end un il acknowledgemen s a e ecei ed om he ecei ing en i y, e.g. in e ne se e o o he de ice. The measu emen e e ence in bo h cases is he in e ace be ween Laye 2 and 3. Any p ocessing ime on highe laye s, e.g. o audio and ideo encoding and decoding, on he applica ion laye , is no conside ed he e. The en i e ne wo k ( adio, co e, and backhaul/agg ega ion) ypically a ec s he la ency, al hough his is use case dependen . Only he use da a plane is conside ed in he e alua ion. Mo i a ion: The KPI is di ec ly ela ed o he 5G goal o p o ide a 5 imes educed E2E la ency compa ed o p esen sys ems [MET13-D11] [5GPPP15]. In p inciple all use cases conside ed in METIS-II would bene i om a la ency educ ion, bu main challenges a e wi h espec o ul a- eliable and/o sa e y- ele an se ices o ype uMTC (e.g. o V2V communica ions) ha equi e as eac ions o he in ol ed pa ies. Wi h espec o he METIS-II goal i has o be no ed ha ne wo k en i ies like he mobile co e, backhaul/agg ega ion links as well as in e ne connec ions migh be included in he E2E ansmission chain which a e no in he main METIS-II ocus. La ency imp o emen s a e o be expec ed mainly in he RAN a ea up o he co e elemen s conside ing new ea u es like D2D communica ion, local b eak-ou s o con en deli e y ne wo k (CDN) unc ionali ies. Ma hema ical de ini ion: The RTT la ency TRTT is he ime span measu ed be ween he s a ime TS1 o he ansmission o a da a packe om an end-use de ice (pee 1) o a emo e Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 70 s a ion o de ice (pee 2) and he ime ins an TA1 when he acknowledgemen , sen by pee 2, a i es a pee 1: TRTT = TA1 - TS1. The OTT la ency is he ime span measu ed be ween he s a ime TS1 o he ansmission o a da a packe om an end-use de ice (pee 1) o a emo e s a ion o de ice (pee 2) and he ime ins an TA2 when pee 2 ecei es he message: TOTT = TA2 - TS1. In he use cases a di e en ia ion is equi ed o la ency KPI alues depending on he a ailabili y o an al eady es ablished adio communica ion da a link. Fo example, o mMTC wi h only low da a amoun s and ene gy-e iciency equi emen s he change be ween a non-ac i e o an ac i e s a e has o be as sho as possible, e.g. change om idle o connec ed mode in LTE, which equi es a minimized signaling o e head. A.4 Reliabili y Quali a i e de ini ion: The eliabili y is an assessmen c i e ion o desc ibe he quali y o a adio link connec ion o ul illing a ce ain se ice le el. Mo i a ion: The KPI is impo an o all conside ed use cases, bu main challenge is seen wi h espec o hose co e ing ul a- eliable se ices (uMTC), whe e e y high eliabili y alues a e eques ed, e.g. o sa e y- ele an applica ions. Ma hema ical de ini ion: Reliabili y can be de ined di e sely, i di e en laye s o he ne wo k o di e en applica ions a e conside ed. Gene ally, eliabili y is de ined as he p obabili y ha a ce ain amoun o da a om an end use de ice is success ully ansmi ed o ano he pee (e.g. In e ne se e , mobile de ice, senso ) wi hin a p ede ined ime ame, i.e. be o e a ce ain deadline expi es. The amoun o da a o be ansmi ed and he deadline a e dependen on he se ice cha ac e is ics in he unde lying use case. Typically, he deadline co esponds o he E2E la ency equi emen o he use case, as de ined in Sec ion A.3. Ma hema ically, he eliabili y can be exp essed as ollows: , whe e is he measu ed E2E la ency and is he deadline, which cha ac e izes he deg ee o eal- ime o he communica ion link. Speci ically, i no e ansmission is allowed o mee he deadline , he eliabili y, , is equi alen o p obabilis ic complemen o packe loss a e. Simila o he la ency KPI, a di e en ia ion o a second use case has o be equi ed du ing e alua ion dependen on he use case backg ound. I a low da a a e adio node, e.g. a senso , is usually swi ching o inac i e s a e a e ansmission o a small packe on he use plane due o ene gy sa ing easons, i has o a ach i s o he ne wo k on he con ol plane be o e he nex da a ansmission. The ime needed o a achmen and he co esponding success a e should be conside ed in he inal eliabili y. Ideally he inal METIS-II concep will p o ide a solu ion o minimize he a achmen ime and maximize he success a e. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 71 A.5 A ailabili y and e ainabili y Quali a i e de ini ion: When he eliabili y dec eases below an accep able le el QoE, hen he use may be so dissa is ied ha i may ega d he se ice as una ailable. The a ailabili y is an assessmen c i e ion o desc ibe inside a co e age a ea he pe cen age o places whe e a se ice is p o ided o he end use wi h he use ‟s eques ed QoE le el. Al e na i ely, a ailabili y is de ined as he pe cen age o use s o communica ion links o which he QoE equi emen s a e ul illed wi hin a ce ain geog aphical a ea, e.g. in e ms o eliabili y as de ined in Sec ion A.4. The la e de ini ion is be e sui ed o he case o D2D communica ions. Re ainabili y is a special aspec o he abo e, by which a se ice has been made a ailable as long as he use needs he se ice. The e is a s ong co ela ion o a ailabili y and e ainabili y o eliabili y in Sec ion A.4. Du ing inal e alua ion o bo h KPIs possible deg ada ions ha e o be aken in o accoun which migh occu du ing hando e p ocesses be ween neighbo ing cells when he end-use is mo ing and be ween di e en adio access laye s (di e en echnologies and/o equency laye s) i he end- use da a bea e is shi ed o a di e en laye . Mo i a ion: The KPI is impo an o all conside ed use cases in METIS-II as i will p o ide a measu e o iden i y he se ice a ailabili y o he end-use s in he in ended co e age a ea. Mo eo e , he KPI will con ibu e o he op imiza ion o he ne wo k layou . I inally depends on he se ice ypes and c i e ia unde lying he di e en use cases. Ma hema ical de ini ion: The a ailabili y in pe cen age is de ined as he numbe o places ( ela ed o a p ede ined a ea uni o pixel size) whe e he QoE le el eques ed by he end-use is achie ed di ided by he o al co e age a ea o a single adio cell o mul i-cell a ea (equal o he o al numbe o pixels) imes 100. Al e na i ely, a ailabili y can be de ined as he p obabili y ha he QoE equi emen s a e ul illed o a use o communica ion link wi hin he se ice a ea. Fo he case when he QoE is exp essed in eliabili y e ms, he a ailabili y is exp essed as ollows: , whe e is he measu ed eliabili y and is he QoE equi emen s in e ms o eliabili y o he unde lying use case. Re ainabili y can be de ined as he p obabili y o o emain la ge han he QoE- a ge , , gi en ha he se ice has al eady been made a ailable. Wi h o he wo ds, i is he p obabili y o a use o sa is ac o y comple e a session o a call, once i has been made a ailable. I is he complemen o d op- a e o a session, call o any o he se ice. A.6 Ene gy consump ion/e iciency Quali a i e de ini ion: In oduces an indica o o highligh he ene gy e iciency o any inno a ion in oduced in METIS-II, including he whole METIS-II sys em a chi ec u e. Mo i a ion: The explosion o a ic demand o eseen beyond he 2020 ho izon u u e and he in insic inc ease o esou ces o be deployed o ackle he METIS-II challenges cas s se e e Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 72 equi emen s in e ms o ene gy consump ion o he co esponding sys em. I is qui e s aigh o wa d and cu en ly widely accep ed ha hese demands ha e o be moni o ed no by me ics e e ing o ene gy consump ion only, bu a he o ene gy e iciency, i.e. including he inc eased p o isioning o capaci y ha he new sys em will ensu e. Ma hema ical de ini ion: An elabo a e desc ip ion o he me ics o be adop ed o ene gy e iciency has been p o ided in he Eu opean FP7 p ojec EARTH and i is aken as a e e ence also wi hin METIS-II. Besides, in ITU-R, ene gy e iciency is also iden i ied as one o he key capabili ies o IMT-2020. I s de ini ion is gi en in [ITU-2083]. The de ini ion should be applied o h ee di e en en i onmen s: componen le el ( o ha dwa e inno a i e solu ions), node le el ( o inno a i e solu ion in he ansmi ing nodes) and ne wo k le el ( o e iciency o he whole ne wo k). Mo e insigh is gi en he e on he la e me ic ( he ne wo k one) due o he speci ic METIS-II goals, bu all he de ails abou componen and node le el me ics can be ound in [EARTH12]. Rega ding ne wo k ene gy e iciency h ee me ics a e wo h men ioning:  Ene gy pe in o ma ion bi , exp essed as ollows     bpsWo bi Jin R P I E I//  ha is he mos widely accep ed me ic o ene gy e iciency, especially in u ban en i onmen s (E s ands o consumed ene gy in a gi en obse a ion pe iod T wi h consumed powe P, I is he in o ma ion olume wi h a e R, measu ed a MAC laye ).  In o ma ion bi pe ene gy, which e e s o he quan i y o in o ma ion bi s ansmi ed o and ecei ed om use s, pe uni o ene gy consump ion o he RAN (e.g. in bi /J).  Powe pe a ea uni   2 /mWin A P A  ypically applicable in subu ban o u al en i onmen s (P is he powe consumed and A he a ea co e age). In o de o e alua e he METIS-II goal, he ene gy consump ion mus be modeled and analyzed bo h o he in as uc u e and o he e minals. On he in as uc u e side, models o analyzing and imp o ing he ene gy e iciency o nowadays echnologies ha e been p oposed in he EARTH p ojec [EARTH12]. I is possible ha o many inno a ions in METIS-II no legacy e e ence will be a ailable. As an example, conside he ene gy consump ion o a ca ha could become a adio node in a u u e 5G ne wo k: o cou se he e is no legacy e e ence o his case and i is also qui e ques ionable i a e e ence o a ca wi hou any “mobile ne wo k ela ed” acili y could be easonable. Whene e new ne wo k elemen s a e in oduced in he METIS-II ne wo k, he only easonable Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 73 e e ence is o he o e all ne wo k ene gy e iciency, conside ing hese new elemen s as u he nodes in he ne wo k p o iding mo e capaci y o he whole sys em. So a no pa icula in es iga ions ha e been conside ed wi h he due a en ion ega ding ene gy e iciency o “de ices”, i.e. o D2D, and “machines”, i.e. o M2M, a leas wi hin he amewo k o a mobile ne wo k pe spec i e. I has o be analyzed case by case he e iciency o hese solu ions, making compa isons o he p e ious condi ions, whe e he inno a i e METIS-II unc ionali ies we e no applied o hese de ices o machines. A.7 Cos Quali a i e de ini ion: Unless o he wise s a ed in he use cases, he cos e e s o all he addi ional in es men s and expenses equi ed by he new METIS-II solu ion. Hence, i he METIS-II solu ion euses pa o he legacy in as uc u e o i is a complemen o i , hen he cos o he legacy ne wo k is no included in o he cos o he METIS-II solu ion. Fo a cellula ne wo k solu ion, he cos ypically includes a pa ela ed o in as uc u e, a pa ela ed o he end-use equipmen , and a pa ela ed o spec um licenses. Cos s ha a e no ela ed o he echnical solu ion, such as cus ome ca e and ma ke ing, a e no conside ed. The in as uc u e pa is ypically di ided in o he capi al in es men o acqui e and deploy he ne wo k, called capi al expendi u e (CAPEX), and he cos s o ope a e he ne wo k, called ope a ional expendi u e (OPEX). Fo ins ance, he CAPEX o a mac o si e co e s he si e acquisi ion and p epa a ion; he equipmen acquisi ion, ins alla ion, and con igu a ion; he backhaul ins alla ion; he an enna sys ems; he powe cables. Typically, CAPEX consis s o one- ime expendi u es. Howe e , o p ac ical easons hese expendi u es a e sp ead o e se e al yea s, i.e. annualized. The OPEX o such a si e co e s si e en al; powe consump ion; main enance, op imiza ion, epa a ions, and eplacemen s; backhaul ansmission cos s; so wa e and ope a ion se ices. The METIS-II ocus is on he adio access ne wo k and he e o e he cos s o he co e ne wo k and se ice pla o ms a e ypically no included, unless o he wise s a ed in he use cases. The cos s wi h he end-use equipmen may be signi ican oo. Some new echnical solu ions may equi e changes bo h in he in as uc u e and he end-use equipmen . Thus an in es men in he in as uc u e migh no b ing he expec ed bene i s unless a signi ican pa o he use equipmen has been eplaced. The e o e he ope a o (s) may ha e an incen i e o in es in speeding up he na u al p ocess o enewing he end-use equipmen . Fo new ypes o se ices, applica ions, and echnical solu ions, i migh be ha de o d aw he line be ween he in as uc u e and he end-use equipmen , as is he case wi h small mobile base s a ions, o elays, moun ed on ehicles. Bu simila p inciples may as well be applicable o mos o he use cases. Mo i a ion: METIS-II has an explici goal o p o iding solu ions whose cos s do no exceed he cos o oday‟s ne wo ks, al hough hei pe o mance is subs an ially be e . Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 80 (Augmen ed eali y applica ion in a c owded museum) Scena io 4: Da a dissemina ion wi h O loading in C owds o handle peak o da a a ic demands G oup 2 Scena io 5: Expanded Co e age and Cellula Ne wo k O loading G oup 2 Scena io 6: Con en dissemina ion based on paymen sys em G oup 2 Scena io 7: Vehicle Flee Managemen G oup 2 Scena io 8: Map-Based Ad anced D i e Assis ance Sys ems (ADAS) G oup 2 Scena io 9: Enhancing T a ic E iciency h ough Coope a i e V2X Communica ion Sys ems G oup 2 T opic Gene ic Use case o small cells cloud se ices: Top down scena io – da a dis ibu ion G oup 2 Gene ic Use case o small cells cloud se ices: Bo om up scena io – p ocessing/wo kload dis ibu ion G oup 1 Applica ion/ adio mixed use cases: Downlink ideo s eaming in in as uc u e a ea G oup 2 Applica ion/ adio mixed use cases: Downlink ideo s eaming in in as uc u e- less a ea G oup 2 Applica ion/ adio mixed use cases: Uplink ideo s eaming G oup 2 Co po a e applica ions o e a em o cloud G oup 2 Sma ai po G oup 2 Small cells cloud da a a i plans o mul ina ional co po a ions G oup 2 Co po a e Vi ual Desk op G oup 1 iJoin S adium G oup 2 Squa e G oup 2 Wide-a ea con inuous co e age G oup 2 Shopping Mall / Ai po G oup 2 Apa om he ones included in he able, he MiWa eS and he MCN p ojec ha e also been analyzed, hough, since hey do no ha e de ailed public desc ip ions o he conside ed UCs, hey ha e no been included in he Table B-3. Howe e , o MiWa eS, he de ined UCs a e adi ional xMBB UCs which a e ocusing on he small cell deploymen (in u ban and u al en i onmen s) and on he lexible spec um usage o he mmW equency bands a 57‒86 GHz and we could conside ha may be cap u ed by he “numbe o Use s and Mobili y, ocused on xMBB se ice” g oup (G oup 2) [MiW15-WP]. Rega ding MCN, he ocus is on equipmen sha ing, ne wo k i ualiza ion and con ex awa eness o load p edic ion in cloud enabled xMBB and MTC scena ios, wi hou ha ing de ailed desc ip ion o he iden i ied equi emen s [MCN13- D21]. The desc ip ion o he UCs o he analyzed Eu opean p ojec s is mainly based on he 5G se ice ype, he numbe o use s, hei mobili y, he la ency equi emen s, and he in as uc u e a ailabili y. The analysis o he UCs conside ed in he Eu opean p ojec s leads o he ollowing ou comes:  mainly xMBB scena ios a e being conside ed om he EU p ojec s, Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 81  in some EU p ojec s mMTC UCs ha e been conside ed,  no uMTC UC has been conside ed in he analyzed EU p ojec s apa om METIS-I. As a conclusion, i could be men ioned ha based on he analysis o he EU p ojec s, no addi ional UC g oup seems o be equi ed, since all he UCs could be included in he iden i ied UC g oups p esen ed in Sec ion B.1. B.3 Analysis o NGMN use cases Acco ding o NGMN, [NGM15], he business con ex beyond 2020 will be no ably di e en om oday since i will ha e o handle he new UCs and business models d i en by he cus ome s‟ and ope a o s‟ needs. 5G will ha e o suppo , based on he NGMN ision, apa om he suppo o he e olu ion o mobile b oadband, new UCs anging om delay-sensi i e ideo applica ions o ul a-low la ency, om high speed en e ainmen applica ions in a ehicle o mobili y o connec ed objec s, and om bes e o applica ions o eliable and ul a- eliable ones such as heal h and sa e y. Thus, NGNM has p oceeded in a ho ough analysis o cap u ing all he cus ome s‟ and ope a o s‟ needs. The analysis is based on wen y- i e UCs o 5G g ouped in o eigh UC amilies (Figu e B-1). The UCs and UC amilies se e as an inpu o s ipula ing equi emen s and de ining he building blocks o he 5G a chi ec u e. Figu e B-1: UC amilies conside ed by NGMN wi h ep esen a i e UCs [NGM15]. Acco ding o he NGMN 5G Whi e Pape [NGM15], he UC analysis is no exhaus i e, hough i p o ides a ho ough and comp ehensi e analysis o he equi emen s o 5G. Following he same analysis as in Sec ion B.1 o he METIS-I UCs, we iden i y in his sec ion he key equi emen s and cha ac e is ics o each UC p oposed by NGMN and we a emp o map i in he UC g oups iden i ied using as e e ence METIS-I. The UC desc ip ion p o ided in NGMN ollows simila Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 82 me hodology o ha o METIS-II, hus enabling a ha monized analysis (see Table B-4). The NGMN objec i e howe e is o inco po a e he business aspec s and he ope a o s‟ iews, hus cap u ing addi ional aspec s on he one hand and posing sligh ly ha de equi emen s o he ne wo k on he o he hand. Table B-4: NGMN UCs analysis by hei cha ac e is ics and he dominan 5G se ice. Use Case key cha ac e is ics 5G se ice ype UC numbe UC name Numbe o de ices Mobili y In as uc u e T a ic Type La ency Requi emen s Reliabili y A ailabili y xMBB uMTC mMTC UC1 Pe asi e Video High High Limi ed Yes Con inuous Low - Low X - - UC2 Sma O ice High High No Yes Con inuous Low - Low X - - UC3 Ope a o Cloud Se ices High High Yes Yes Con inuous Medium - Low X - - UC4 HD Video/Pho o Sha ing in S adium/Open- Ai Ga he ing High Medium No Yes Con inuous Medium - High X - - UC5 50+ Mbps e e ywhe e Low Medium High Limi ed Con inuous High - High X - - UC6 Ul a-low Cos Ne wo ks Low Low Medium Limi ed Con inuous Medium - Low X - - UC7 High Speed T ain Medium Medium High Medioc e All ypes Low - Low X - - UC8 Remo e Compu ing Low Medium High Medioc e Con inuous Low - Low X - - UC9 Mo ing Ho Spo s Low High High Medioc e Bu s y Low - High X - - UC10 3D Connec i i y - Ai c a Low Low High Medioc e High Low - High X - - UC11 Sma Wea ables High Low High All ypes Pe iodic Low - High - - X UC12 Senso Ne wo ks High Low Low All ypes Pe iodic Low - High - - X UC13 Mobile Video Su eillance High Medium High High Con inuous Low - High X - X UC14 Tac ile In e ne High High High High Va ious ypes Low High High - X - UC15 Na u al Disas e High Low Low Limi ed Sho messages High High High - X X UC16 Au oma ed T a ic Con ol and D i ing Low High High High All ypes Low High High - X - UC17 Collabo a i e Robo s: A Con ol Ne wo k o Robo s Low Low No High Con inuous Low High High - X - UC18 eHeal h: Ex eme Li e C i ical High Low No/Low High Sho messages Low High High - X - UC19 Remo e Objec Manipula ion: Remo e Su ge y Low Low Low High Con inuous Low High High - X - UC20 3D connec i i y: D ones Low Low High Limi ed Con inuous Low High High - X - UC21 Public Sa e y Low Low Low High Con inuous Low High High - X - UC22 News and In o ma ion High High High High All ypes High - - X - - UC23 Local B oadcas - High High Low High All ypes High - - X - - Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 83 like Se ices UC24 Regional B oadcas -like Se ices High High High High All ypes High - - X - - UC25 Na ional B oadcas -like Se ices High High High High All ypes High - - X - - Table B-5 maps he NGMN UC o he UC g oups, as hey came up in Sec ion B.1. As shown in Table B-5, he NGMN UCs can be g ouped in he conside ed ca ego ies. Howe e , some UCs ei he may no be included in he iden i ied g oups, o ex ensions a e equi ed. These iden i ied use cases a e he ollowing ones:  UC5: 50+ Mbps e e ywhe e, which conside s mobili y combined wi h xMBB se ice, wi h lack o in as uc u e. I hasbeendecided ha  he“50+Mbpse e ywhe e”usecase will be used o he e alua ion o he mechanisms o METIS-II.  UC16: Au oma ed T a ic Con ol and D i ing, which conside s hea y mo ing use s (i.e., ca s) which ha e o exchange (a) in o ma ion ans e wi h high eliabili y and wi h e y low la ency, and, (b) xMBB se ice among he ca s. This use case may be seen as a mix u e o METIS-I UCs 8 (Real- ime emo e compu ing o mobile e minals) and 12 (T a ic e iciency and sa e y). Howe e , i ocuses on he ul a- eliabili y na u e, since i e e s o au oma ed d i ing, whe eas METIS-I UC8 is ela ed o emo e compu ing, ocusing on he xMBB na u e. I has been hen decided ha METIS-I UC “T a ic Sa e y and E iciency” will be upda ed acco dingly so as o inco po a e he co esponding equi emen s. Table B-5: NGMN UCs g ouping. UC g oup desc ip ion NGMN UCs in he espec i e g oup G oup 1 S a ic use s wi h high da a a e equi emen s and xMBB se ice. UC2, UC4 G oup 2 Mo ing use s, indoo / ou doo en i onmen s, wi h high o ela i ely high use densi y, and xMBB se ice. UC1, UC3, UC6, UC7, UC8, UC9, UC10, UC22, UC23, UC24, UC25 G oup 3 Rela i ely low use densi y wi h high a ailabili y equi emen s, and uMTC se ice. UC14, UC15, UC16, UC17, UC18, UC19, UC20, UC21 G oup 4 High o ela i ely high use densi y and mMTC se ice. UC11, UC12, UC13 B.4 Analysis o ITU-R usage scena ios In i s D a New Recommenda ion “IMT Vision - F amewo k and o e all objec i es o he u u e de elopmen o IMT o 2020 and beyond” [ITU15], he ITU-R WP 5D de ined he amewo k and o e all objec i es o he u u e de elopmen o In e na ional Mobile Telecommunica ions (IMT) o 2020 and beyond in ligh o he oles ha IMT could play o be e se e he needs o he ne wo ked socie y, o bo h de eloped and de eloping coun ies, in he u u e. IMT o 2020 Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 84 and beyond is en isaged o expand and suppo di e se usage scena ios and applica ions ha will con inue beyond he cu en IMT:  Enhanced Mobile B oadband: Human-cen ic use cases o enhanced access o mul i- media con en , se ices and da a wi h imp o ed pe o mance and inc easingly seamless use expe ience. This usage scena io co e s a ange o cases wi h di e en equi emen s, e.g., he ho spo case wi h high use densi y, e y high a ic capaci y and low use mobili y, as well as he wide a ea co e age case wi h seamless adio co e age p o iding much imp o ed use da a a e compa ed o exis ing da a a es wi h medium o high use mobili y.  Ul a- eliable and low la ency communica ions: S ingen equi emen s o capabili ies such as h oughpu , la ency and a ailabili y. Examples: Wi eless con ol o indus ial manu ac u ing o p oduc ion p ocesses, emo e medical su ge y, dis ibu ion au oma ion in a sma g id, anspo a ion sa e y, e c.  Massi e machine ype communica ions: Cha ac e ized by a e y la ge numbe o connec ed de ices ypically ansmi ing a ela i ely low olume o non-delay-sensi i e da a. De ices a e equi ed o be low cos , and ha e a e y long ba e y li e. These usage scena ios illus a ed by some examples in Figu e B-2 a e in p inciple a one- o-one mapping o he 3 so-called gene ic 5G se ices xMBB, uMTC and mMTC s a ed in he METIS-I p ojec [MET15-D66]. Due o ha consensus he e is no need in METIS-II o make any changes w. . . se ice ypes which a e he basis o he inal use case de ini ion. Figu e B-2: Examples o usage scena ios conside ed by ITU-R o IMT beyond 2020 [ITU15]. The ITU-R has conside ed 8 pa ame e s o be key capabili ies o IMT-2020:  Peak da a a e (i.e., maximum achie able da a a e unde ideal condi ions pe use /de ice). Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 85  Use expe ienced da a a e (i.e., achie able da a a e ha is a ailable ubiqui ously ac oss he co e age a ea o a mobile use /de ice).  La ency (i.e., he con ibu ion by he adio ne wo k o he ime om when he sou ce sends a packe o when he des ina ion ecei es i ).  Mobili y (i.e., maximum speed a which a de ined QoS and seamless ans e be ween adio nodes which may belong o di e en laye s and/o adio access echnologies (mul i-laye /-RAT) can be achie ed).  Connec ion densi y (i.e., o al numbe o connec ed and/o accessible de ices pe uni a ea).  Ene gy e iciency ( e e s on he ne wo k side o he quan i y o in o ma ion bi s ansmi ed o / ecei ed om use s, pe uni o ene gy consump ion o he RAN, and on he de ice side o he quan i y o in o ma ion bi s pe uni o ene gy consump ion o he communica ion module (in bi /Joule in bo h cases)).  Spec um e iciency (i.e., a e age da a h oughpu pe uni o spec um esou ce and pe cell (bi /s/Hz)).  A ea a ic capaci y (i.e., o al a ic h oughpu se ed pe geog aphic a ea (in bi /s/m2)). The key capabili ies expec ed o IMT-2020 a e shown in Figu e B-3(a) compa ed wi h hose o IMT-Ad anced, bu due o di e en equi emen s he impo ance o he achie ed capabili y alues is no he same o all h ee usage scena ios. This is shown in Figu e B-3(b). Ne e heless, dependen on he dedica ed se ice o en i onmen , he impo ance may also a y wi hin one usage scena io (e.g. di e en mobili y equi emen s o he ho spo and he wide a ea co e age case as men ioned be o e o enhanced MBB). Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 86 a) Expec ed enhancemen s e sus IMT-Ad anced. b) Impo ance o di e en usage scena ios. Figu e B-3: Key capabili ies o IMT beyond 2020 [ITU15]. The ITU-R no ed also o he capabili ies wi h impo ance o he u u e IMT-2020 sys em: Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 87  Spec um and bandwid h lexibili y (i.e., lexibili y o he sys em design o handle di e en scena ios and in pa icula o he capabili y o ope a e a di e en equency anges, including highe equencies and wide channel bandwid hs han oday).  Reliabili y (i.e., capabili y o p o ide a gi en se ice wi h a e y high le el o a ailabili y).  Resilience (i.e., abili y o he ne wo k o con inue ope a ing co ec ly du ing and a e a na u al o man-made dis u bance, such as he loss o mains powe sys em).  Secu i y and p i acy ( e e s o se e al a eas such as enc yp ion and in eg i y p o ec ion o use da a and signaling, as well as end use p i acy p e en ing unau ho ized use acking, and p o ec ion o ne wo k agains hacking, aud, denial o se ice, man in he middle a acks e c.).  Ope a ional li e ime (i.e., ope a ion ime pe s o ed ene gy capaci y which is pa icula ly impo an o machine- ype de ices equi ing a e y long ba e y li e (e.g., mo e han 10 yea s) whose egula main enance is di icul due o physical o economic easons). In he use cases o he METIS-II p ojec – and o he o he 5G PPP p ojec s as well – he capabili ies ecommended by ITU-R should be e e enced as KPIs. B.5 3GPP S udy on New Se ices and Ma ke s Technology Enable s 3GPP SA1 s a ed in Ma ch 2015 a new S udy I em called “S udy on New Se ices and Ma ke s Technology Enable s” (FS_SMARTER) [3GPP15-SMARTER]. The objec i e o his s udy is o iden i y new ma ke segmen s and e icals he 3GPP sys em needs o add ess in he u u e o enable new business models and a ying ope a ional schemes o op imize he use o he ope a o s' ne wo ks (e.g., as e ne wo k and se ice deploymen , lowe ope a ional complexi y and cos ). The wo k will ocus on use cases and equi emen s ha canno be me wi h oday‟s E ol ed Packe Sys em. In a i s s ep SA1 is de eloping use cases o a ious scena ios and iden i ying ela ed high- le el po en ial equi emen s. As second s ep/phase, i is wo king o g oup oge he use cases wi h common cha ac e is ics and ag eed o g oup he di e en use cases in TR 22.891 in o ou building blocks. In o de o accele a e his wo k, i held a dedica ed ad-hoc mee ing du ing 19-21 Oc obe , 2015. The cu en s a us o he FS_SMARTER ou pu wi h desc ip ions o 74 di e en use cases is e lec ed in he D a Technical Repo TR 22.891 [3GPP15-22891] which was p esen ed du ing SA mee ing held in Dec 2015. In he ollowing mee ings, SA1 will con inue o wo k on he indi idual building block s udy i ems o u he de elop each selec ed use case (o g oup o use cases) and o cap u e desi ed sys em equi emen s and capabili ies ha apply ac oss he di e en e icals. This essen ially e lec s a ho izon al iew on po en ial equi emen s o complemen he e ical use cases. The a ge da e o his inal s ep wi h inaliza ion o indi idual TRs is Ma ch 2016. The ou ag eed building block s udies a e: S udy on New Se ices and Ma ke s Technology Enable s – massi e In e ne o Things o FS_SMARTER-mIoT, S udy on New Se ices and Ma ke s Technology Enable s – C i ical Communica ions o FS_SMARTER-CRIC, S udy on New Se ices and Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 88 Ma ke s Technology Enable s – enhanced Mobile B oadband o FS_SMARTER-eMBB and S udy on New Se ices and Ma ke s Technology Enable s – Ne wo k Ope a ion o FS_SMARTER-NEO. The lis o use cases in cu en D a TR 22.891 co e s no only se ice- ela ed use cases, bu also use cases which a e mo e ela ed o u u e ne wo k se -up and ope a ion. Examples o ha a e:  Ne wo k slicing;  On-demand ne wo king;  Mul i-RAT connec i i y incl. selec ion o bes connec i i y pe a ic ype;  Coexis ence wi h legacy sys ems;  Flexible applica ion a ic ou ing including op imiza ions in case o se e changes;  Ne wo k enhancemen s o suppo lexibili y, scalabili y, and au oma ion;  Wi eless backhauling. Such kind o use cases we e no e lec ed by he wo k in METIS-I and a e also no in ended in METIS-II, as he ocus is a de ini ion o end use se ice- ela ed use cases. The se ice- ela ed use cases co e di e en xMBB usage scena ios, e.g. o indoo and ho spo scena ios (incl. i ual p esence) as well as seamless wide-a ea co e age. In addi ion, use cases o low mobili y as well as high mobili y including connec i i y o ca s a e lis ed. Also b oadcas ing is highligh ed by use cases. The ypical mMTC se ice is e lec ed e.g. by “Wide a ea senso moni o ing and e en d i en ala ms” as well as by use cases add essing he con igu a ion o such low-cos de ices. Se e al use cases lis ed a e ela ed o uMTC scena ios, e.g.  Tac ile in e ne ;  Indus ial con ol and ac o y/p ocess au oma ion;  Robo ics;  Localized as well as wide-a ea emo e con ol (e.g. connec i i y o d ones). Fu he examples a e on eme gency and li eline communica ions as well as on highly accu a e use /de ice posi ioning and acking. These se ice- ela ed use cases in D a TR 22.891 a e al eady conside ed o a la ge ex en by he co esponding desc ip ions o METIS-I scena ios and es /use cases [MET13-D11] [MET15- D15], a leas w. . . hei challenging equi emen s. Ne e heless, as TR 22.891 is no inally app o ed, a c oss-check wi h he equi emen s and KPIs aken o he selec ed use cases in METIS-II should be pe o med o achie e consis ency be ween di e en ac i i ies. Excep o he easibili y s udy SMARTER 3GPP SA1 has also ini ia ed 3 o he S udy I ems which ha e some ele ance o 5G, e en i hey a e ocused on LTE echnology: he i s one is Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 89 on V2X use cases (FS_V2XLTE) [3GPP15-V2XLTE], he o he s on mission c i ical communica ions (FC_MCVIDEO [3GPP15-MCVIDEO] and FC_MCDATA [3GPP15-MCDATA]). The D a TR 22.885 o FS_V2XLTE [3GPP15-22885] is in ended o be comple ed in No 2015. I desc ibes use cases in a ehicula en i onmen o V2V, V2I and V2P in he con ex o bo h sa e y and non-sa e y se ices. Acco ding o s udies in METIS-I pe o med e.g. by BMW no all equi emen s especially o sa e y se ices can be e icien ly ul illed by cu en LTE eleases. The wo k in METIS-I on 5G echnology componen s al eady conside ed he mos challenging pa s o he FS_V2XLTE use cases ia UC12 “T a ic e iciency and sa e y” (see Sec ion 3.2 and [MET13-D11]), so i is no explici ly equi ed o conside all lis ed use cases in de ail o he wo k in METIS-II. The wo k on mission c i ical ideo and da a communica ion by SA1 is ela ed e.g. o g oup communica ions sys ems, p oximi y based se ices, o isola ed LTE ope a ion o public sa e y. Di e en use cases which can be cha ac e ized o belong o he 5G uMTC scena io a e lis ed in cu en D a s TR 22.879 [3GPP15-22879] and TR 22.880 [3GPP15-22879] (in ended o be app o ed by No 2015). In mos cases he use cases do no p o ide dedica ed alues o KPIs which can be ans e ed o 5G, bu unc ional equi emen s he wi eless sys em has o ul ill. The main impac on 5G is ha i is assumed ha hose mission c i ical se ices should un on a common pla o m (incl. he de ices) oge he wi h o he comme cial se ices. The use cases a e mainly ela ed o g oup communica ions wi h di e en se ice ypes wi h and also wi hou suppo o he ne wo k in as uc u e, so e.g. ex ensions o D2D communica ion conside ed in METIS-I is equi ed. O he se ice ypes, e.g. emo e con ol ea u es as well as ideo su eillance, a e al eady co e ed by co esponding use cases in SMARTER. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 96 km/h) C.3 UC3: B oadband access e e ywhe e This is a new METIS use case buil on elemen s o a NGMN use case [NGM15] combined wi h some aspec s om he METIS-I es case se en (Blind spo s) in [MET13-D11]. This use case mainly add esses he use case amily xMBB. C.3.1 Backg ound and mo i a ion The demand o e y high expe ienced use h oughpu In e ne access a any ime and a any place is cons an ly inc easing. The ubiqui ous capaci y demands o u u e use s will be challenging o sa is y in a eas wi h spa se ne wo k in as uc u e, such as sca cely popula ed a eas, u al and e en subu ban a eas. A consis en use expe ience wi h espec o h oughpu , needs a minimum expe ienced use h oughpu gua an eed e e ywhe e. The a ge alue o 50 Mbps e e ywhe e should be unde s ood as he minimum expe ienced use h oughpu and no a single use ‟s heo e ical peak a e. Fu he mo e, i is emphasized ha his use a e has o be deli e ed consis en ly ac oss he co e age a ea (i.e. e en a cell edges). Fu he mo e, he ba e y consump ion o sma phones and able e minals in a eas wi h low co e age inc eases signi ican ly due o highe p opaga ion losses. A a ge o 50% ene gy consomp ion educ ion compa ed wi h legacy ne wo k should be achie ed bo h a ne wo k in as uc u e and a end use le el. While cell densi ica ion is p omising o boos ing capaci y in u u e u ban en i onmen , wide co e age solu ions as well as lexible, ene gy and cos e icien solu ions mus be de eloped in u u e wi eless communica ion sys ems o p o ide ubiqui ous co e age in subu ban and mo e o less emo e u al a eas. C.3.2 KPIs and equi emen s The equi emen s and he key pe o mance indica o s a e summa ized in Table C-8, based on [NGM15]. High expe ienced use h oughpu co e age is expec ed a e e y loca ion o he se ice a ea, e en in emo e u al a eas. Mos ly, ideo s eaming and ile downloads a e equi ed, co esponding o a high expe ienced h oughpu pe use . In pa icula , each use should be able o expe ience a h oughpu o a leas 50 Mbps in downlink and 25 Mbps in uplink. Use densi y and a ic olume densi y will a y acco ding o en i onmen ypes. In u al a eas we assume a use densi y o 100 use s pe km2, which esul s in a o al a ic olume o 7.5 Gbps/km2. Lowe densi y alues o mo e emo e u al en i onmen s, ep esen a i e o mo e sca cely popula ed a eas, a e p o ided. Subu ban en i onmen s, ep esen a i e o a eas close o u ban ci ies a e also add essed. The end- o-end (one ip) RAN la ency has o be main ained below 10 ms. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 97 The se ice a ailabili y mus be a leas 99% (space domain). In addi ion, 50 Mbps DL and 25 Mbps UL should be achie ed wi h 95% e ainabili y whe e he se ice is a ailable ( ime domain). Reliabili y is no in he ocus o his use case. Ne e heless, some le el o eliabili y is equi ed in o de o ensu e a seamless consump ion o ideo se ices oge he wi h imely ile deli e y. Fo example, low eliabili y le els can lowe he QoE o ideo se ices due o he p esence o isual e o s and can also delay he download ime o ile deli e y se ices. As a esul , a eliabili y alue o 95% is assumed. Bo h in as uc u al and end use ene gy consump ion should be minimized. In his ega ding, ope a o s associa ed in as uc u e should ha e an ene gy consump ion educed by 50%, whe eas he use s need o spend less han hal he ene gy hey consume in legacy ne wo k. Ne wo k cos including in as uc u al equipmen , si e en al, ene gy consump ion e c., is expec ed o be educed by 50%. Solu ions o low cos backhauling a e hus o be de eloped. Human use s a e gene ally he main a ge o his use case, while a la ge di e si y o se ices should be suppo ed, such as ile downloading and ideo s eaming. Table C-8: KPIs and equi emen s o UC3. Va iable/pa ame e Value (u ban/ u al/highway) Pe o mance a ge s Expe ienced use h oughpu 50 Mbps Mbps (DL) and 25 Mbps (UL) T a ic olume densi y (busy hou ) Fa emo e u al : DL : 0.5 Gbps/km2 Fa emo e u al : UL : 0.25 Gbps/km2 Ru al : DL : 5 Gbps/km2 Ru al : UL : 2.5 Gbps/km2 Subu ban : DL : 20 Gbps/km2 Subu ban : UL : 10 Gbps/km2 End- o-end RAN la ency (including connec ion se up and de ec ion delay) o ecei e s wi hin he a ge ange Less han 10 ms A ailabili y: pe cen age o a ea whe e ansmissions mee he expe ienced use 99% Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 98 da a a e equi emen Re ainabili y : pe cen age o ime whe e ansmissions mee he expe ienced use da a a e 95% Cons ain s Ene gy consump ion (in as uc u e) Low-ene gy ope a ions a e p e e ed mainly o cos and sus ainabili y easons. When no ansmi ing use da a he consumed ene gy should be e y low. Gene ally he ne wo k ene gy consump ion should be compa able o he ene gy consump ion o oday‟s me opoli an deploymen s, despi e he d as ically inc eased amoun o a ic Ene gy consump ion (UE o o he de ices) UE should be able o ope a e on ba e y o se e al hou s Cos (in as uc u e) Ne wo k in as uc u e should be cheap bo h in e ms o ha dwa e cos as well as ins alla ion and main enance cos s Cos (UE o o he de ices) UE cos should be simila o oday‟s sma phones o 3G/4G modems Use case de ini ion Use /de ice densi y Fa emo e u al: 11 use s pe km2 Ru al: 100 use pe km2 Subu ban: 400 use pe km2 Use ype Human Use mobili y A mix o s a ic use s and use s in ehicles wi h eloci y up o 120 km/h Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 99 C.4 UC4: Massi e dis ibu ion o senso s and ac ua o s This use case is based on a es case ele en in [MET13-D11]. I mainly add esses he use case amily mMTC. C.4.1 Backg ound and mo i a ion The impo ance o his use case will g ow oge he wi h he massi e deploymen o hese low cos and o low ene gy consump ion de ices. In o de o ge he maximum o in o ma ion om hese de ices, so as o inc ease en i onmen al awa eness and be e use expe ience, he e is a need o hese de ices o be able o communica e wi h o he de ices and/o wi h he ne wo k. A ange o examples a e p esen ed below. Some examples o cases o he acking o po able objec s a e lis ed in he sequel:  Tools (e.g. d ills), whe e he pu pose o he communica ion node in each ool is o measu e in which en i onmen and how he ool is used (e.g. in o de o no i y he use ha he should a he be using a di e en ool o o de elop ools which a e be e sui ed o hei usage) and de ec ea ly signs o p oduc ailu e.  O he p oduc s whe e he p oduce is in e es ed in imp o ing usabili y. An example could be books (such as manuals), whe e senso s de ec which pages a e opened when and in which o de . Ano he example could be ela ed o a es au an whe e i is measu ed which ables and chai s a e occupied.  P oduc s which equi e ca e (e.g. lowe po s), whe e he communica ion senso /node would send a wa ning i wa e ing o e iliza ion is needed.  F agile p oduc s, whe e he communica ion node could measu e whe he he p oduc is handled wi h su icien ca e (e.g. no oo much accele a ion, and igh empe a u e), and send an ala m i his is no he case.  Po en ially dange ous p oduc s, such as kni es, chemicals (o weapons), whe e he communica ion node could send an ala m i somebody unau ho ized uses o mo es hese p oduc s.  P oduc s ha expi e, such as g oce ies and spices, whe e an ala m could be sen i p oduc p ope ies c oss a ce ain h eshold.  P oduc s whe e s a is ics on he mo emen a e o be collec ed. Fo ins ance, a company may wan o moni o a lee o , e.g. bicycles. In an ex eme case, one could imagine he Eu opean Cen al Bank agging Eu o no es in o de o obse e he low o ha d cu ency.  P oduc s o en subjec o he , such as hand-bags o jewel y, whe e communica ion nodes could aise an ala m in case o unau ho ized mo emen . Simila ly i he cus ome places he p oduc s o a w ong loca ion (shel ), he communica ion node can wa n he cus ome and p obably cus ome se ice. Some o he examples o cases o he moni o ing o , e.g. en i onmen , ma e ials, may include: Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 100  Ag icul u al applica ion: A la ge numbe o senso s sp ead ou o e la ge ag icul u al a eas o measu e, e.g. e ili y and humidi y, o help he a me op imize he igh ime o ha es ing and e ilizing (app oxima ely 10 senso s pe km2, i is accep able i 50% o senso s manage o ge a 20 by e uplink ne payload h ough o he in as uc u e once pe day). I could also be imagined ha such senso s a e used indoo s in, e.g. g eenhouses.  Ma e ial moni o ing (pa icula example: wind mill): Senso s a e placed e e y 5 m on he su ace o he s uc u e o a mode n powe -gene a ing wind mill, epo ing ib a ions and o he measu es ha may gi e an ea ly indica ion o ma e ial damage o subop imal usage (50 senso s pe wind mill, all senso s should be able o ge a 20 by e ne uplink payload h ough o he in as uc u e once pe day). No e ha in his pa icula case he senso s a e mo ing, hough his mo emen is highly p edic able.  Ma e ial moni o ing (pa icula example: high-speed ain): Senso s a e placed in each wheel o he ain and a e able o measu e ib a ions and ea ly indica ions o ack o wheel damage (4 senso s pe ca iage, all senso s should be able o ge a 20 by e ne payload h ough o he in as uc u e pe day). Such senso s – e en a such a low du y cycle - would be able o p e en dange ous ain acciden s ha ha e happened in he pas due o unno iced ma e ial aul s.  Ma e ial moni o ing (pa icula example: building): Senso s a e placed in a building o on he su ace o a building, again o moni o ib a ions and o he ea ly indica ions o po en ial ma e ial ailu e. These senso s will also be use ul o quickly assess he s a e o a building a e a na u al disas e , such as an ea hquake o a hu icane (all senso s should be able o ge a 20 by e ne payload h ough o he In e ne once pe day). C.4.2 KPIs and equi emen s This use case a ge s he 5G goals ela ed o he numbe o connec ed de ices, ene gy consump ion pe de ice by main aining he same ene gy consump ion and cos a he ne wo k side. The cu en s a e o equi emen s o hese uses cases is e lec ed in [3GPP15-45820], o which he pe o mance objec i es a e:  Maximum Coupling Loss o 164.0 dB (wi h a leas 160 bps da a a e on bo h he uplink and downlink)  52547 de ices wi hin a cell si e sec o , based on: 40 de ices pe household, wi h 1517 household pe km2, and a In e Si e Dis ance o 1732 m  10 yea s du abili y o a 5Wh ba e y, o a ic based on 50 by es TX wi h a 24 hou s epo ing in e al in a 164.0 dB coupling loss scena io  Reduced complexi y (wi h e e ence on cu en GPRS modems) in o de o enable cheape de ices, being measu ed based on: o Silicon a ea es ima e, including on-chip memo y. o Indica ion o equi ed ga e densi y should be gi en. o Rela i e a ea o RF and baseband unc ions. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 101 o Lis o ex e nal componen s. o Any special cha ac e is ics o ex e nal componen s speci ic o a sys em p oposal. o DSP cycles / second. Howe e , o 5G on 2020, i is o eseen ha mo e ambi ious a ge s a e needs, as:  Numbe o de ices suppo ed in he sys em 1 000 000 de ices pe km2.  Payload size is assumed o be e y small, om 20 o 125 by es pe message, wi h a e y la ge ansmission cycle, which will depend on he speci ic applica ion. As a e e ence, an expe ienced use h oughpu o 1 kbps is o eseen pe de ice, which is equi alen o 1 message o 125 by es pe second.  Long ba e y li e (on he o de o 10+ yea s) o he wi eless de ice, implying he need o high ene gy e iciency. Ba e y li e is di ec ly ela ed o ene gy e iciency.  Minimum possible signaling o e head.  Keeping he UE complexi y in o de o gua an ee ul a-low cos de ices.  99.9% a ailabili y. Simila ene gy consump ion and cos o he in as uc u e as o he base sys ems o oday. The equi emen s and he key pe o mance indica o s a e summa ized in Table C-9. Table C-9: KPIs and equi emen s o UC4. Va iable/pa ame e Value Pe o mance a ge s Ba e y li e A leas 10 yea s (assuming 5 Wa s-hou ba e y) De ice densi y 1 000 000 de ices pe km2 A ailabili y 99.9% T a ic olume pe de ice F om ew by es pe day o 125 by es pe second Cons ain s Ene gy e iciency (in as uc u e) In p inciple no speci ic cons ain s o he in as uc u e Ene gy e iciency (UE o o he de ices) The powe supply a ailabili y is limi ed, so low-ene gy ope a ion is equi ed. Fo senso ype de ices wi h ba e y powe supply only, he ene gy-op imized Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 102 ope a ion is equi ed. Cos (in as uc u e) In as uc u e cos should be kep on he same le el pe a ea as oday. Cos (UE o o he de ices) Fo senso ype de ices a signi ican cos educ ion compa ed o no mal handse de ices is needed. C.5 UC5: Connec ed ca s This use case is based on es cases eigh and wel e o METIS-I [MET13-D11]. I mainly add esses he use case amilies uMTC and xMBB. C.5.1 Backg ound and mo i a ion The use o emo e se ices is also applicable a highe use mobili y, e.g. while d i ing ca s o using public anspo a ion, and no me ely aking place in slow mobili y o s a iona y se ings. Wi h highe use mobili y enabled on- he-way wo ke s as well as a leisu ed people can enjoy he bene i s o eal- ime emo e compu ing o mobile e minals. A he same ime he connec ed ca also p o ides a sa e and e icien jou ney ia he communica ion o i s su ounding. This communica ion enables he ca o a oid acciden s, bu i also enables o he ypes o a ic planning such as a oiding a ic jam queues and minimizing uel consump ion. The eby he connec ed ca enables a ic sa e y, e iciency and eal- ime emo e se ices. C.5.2 KPIs and equi emen s The connec ed ca needs o be a ic e icien and ale in e ms o sa e y as well as able o p o ide good eal- ime emo e compu ing expe ience o he end-use s. T a ic e iciency and sa e y The main challenges o his use case, when i comes o “ a ic e iciency and sa e y” lie in he equi ed eliabili y, a ailabili y, and la ency o au omo i e sa e y se ices. The equi emen s a e desc ibed below and he key pe o mance indica o s and cons ain s a e summa ized in Table C-10. A maximum wi eless ne wo k end- o-end delay (including wi eless de ice de ec ion, connec ion se up and adio ansmission bu excluding he ime needed om he ehicle o p ocess and gene a e he in o ma ion message) o 5 ms, wi h ansmission eliabili y o 99.999% should be gua an eed o deli e he d i e sa e y se ice. This is a majo challenge.  V2X communica ion needs o be es ablished ac oss di e en ne wo k ope a o s wi h he same equi emen s in e ms o la ency and se ice gua an ee as wi hin a single ne wo k ope a o .  ≈100% a ailabili y equi ed such ha he se ices a e p esen a e e y poin on he oad. Addi ional KPIs and cons ain s a e he ollowing: Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 103  Rela i e posi ioning accu acy below 0.5 m is needed. GPS may no always be a ailable and su icien , and hence cellula based posi ioning echniques could be use ul.  Da a a ic (inspi ed by cu en ETSI Technical Commi ee ITS and IEEE s anda diza ion wo k [MMK+11], [S 11], hough, he pa ame e alues a e mo e challenging han wha is discussed o oday‟s sys ems): o Pe iodic b oadcas a ic consis ing o a leas 1600 payload by e ( o ansmission o in o ma ion ela ed o 10 de ec ed objec s esul ing om local en i onmen pe cep ion and he in o ma ion ela ed o he ac ual ehicle) wi h epe i ion a e o a leas 5-10 Hz. The upda e a e is chosen high enough such ha he ehicle eloci y ec o does no change oo much be ween upda es. The a ic gene a ed by each ehicle has o be deli e ed o all he neighbo ing ehicles wi hin he speci ied ange. o E en -d i en b oadcas a ic consis ing o a leas 1600 payload by e wi h epe i ion a e o a leas 5-10 Hz ( o ansmission o in o ma ion ela ed o 10 de ec ed objec s esul ing om local en i onmen pe cep ion and he in o ma ion ela ed o he ac ual ehicle). o Bo h a ic ypes (pe iodic and e en d i en) can exis a he same ime. No e ha he epe i ion a e o bo h a ic ypes is de e mined by he need o ack changes in he en i onmen . o Fo communica ion be ween ehicles and o he de ices (e.g., sma phones) a payload o 500 by e may be su icien ( o ansmission o he in o ma ion om he ac ual consume elec onics de ice, such as cu en posi ion and addi ional da a om he de ice senso s).  Fou di e en mobili y en i onmen s need o be dis inguished: U ban, u al, and highway. o U ban: maximum absolu e eloci y o 60 km/h and 120 km/h ela i e eloci y be ween ehicles. o Ru al: maximum absolu e eloci y o 120 km/h and 240 km/h ela i e eloci y be ween ehicles. o Highway: maximum absolu e eloci y o 250 km/h and 500 km/h ela i e eloci y be ween ehicles. o Vulne able oad use eloci ies anging om 3 km/h (pedes ian) up o 30 km/h (bicycle).  Use and de ice densi ies depend on he en i onmen and scena io: o Vehicula de ices:  In u ban en i onmen s he use densi y can be up o 1000 use s pe km2  In u al and highway en i onmen s he use densi y can be up o 100 use s pe km2 o Vulne able oad use de ices:  In u al and highway en i onmen s he use densi y can be up o 150 ele an use s pe km2. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 104  In u ban en i onmen s he use densi y can be up o 5000 ele an use s pe km2.  The equi ed communica ion ange is di e en o he a ious en i onmen s: o Up o 1 km in highway scena ios. o Up o 500 m in u al scena ios. o Up o 50 m in u ban scena ios.  Addi ional spec um cons ain s a e: o Use o dedica ed spec um i a ailable, o p e e able equencies below 5 GHz. Real- ime emo e compu ing o mobile e minals The main KPIs o his use case co esponding o xMBB aspec s a e  High a ailabili y o 99%  Low E2E la encies o less han 10 ms. The equi emen s and key pe o mance indica o s o his use case a e summa ized in Table C- 10 and desc ibed in mo e de ail as ollows:  High da a a e connec i i y is expec ed o ehicles (o speci ically, he on-boa d de ices) and also o use de ices inside o ehicles. Real- ime in e ac i e se ices (such as augmen ed eali y and i ual o ice applica ions), loca ion based se ices, and any se ice ha equi es o shi ce ain complex p ocessing asks (usually pe o med locally) o a emo e se e co espond o high da a a e demands wi h eal- ime equi emen s. As a esul we assume ha e e y ac i e de ice in he ehicle equi es a da a a e o 100 Mbps in downlink and 20 Mbps in uplink. Assuming a mos 5 simul aneously ac i e de ices pe ehicle (including he on-boa d de ices) and a ehicle densi y o 100 ehicles pe km2 in mo o ways, his leads o a o al a ic olume o 60 Gbps/km2.  Mul i-ope a o solu ions a e equi ed in o de o se e use s wi h di e en ne wo k ope a o con ac s and hus make his KPIs a ailable.  The conside ed applica ions ha e eal- ime equi emen s. The e o e, E2E la encies lowe han 10 [ms] wi h high eliabili y (i.e. 95% o he packe s should be success ully ansmi ed wi hin his ime) needs o be achie ed.  The ene gy e iciency o he de ices used in he ehicles and o Elec onic Con ol Uni s (ECUs) using emo e p ocessing se ices should be high. Powe consump ion should be minimized in o de ex end he ba e y ime and allow o high p oduc i i y o he use s inside he ehicles.  De ice densi ies depend mainly on he means o anspo a ion sys em. By 2020 and beyond people a e expec ed o ca y mo e han one cellula de ice. Mo eo e , he ehicle i sel ( h ough in- ehicle ECUs) may shi complex p ocessing asks o a emo e Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 105 se e in eal- ime. The numbe o ac i e de ices (passenge de ices plus ehicle on- boa d de ices) scales wi h he size o he ehicle: o A mos 5 simul aneously ac i e de ices pe ca o Up o 50 use de ices simul aneously ac i e pe bus o Up o 300 use de ices simul aneously ac i e pe ain  The use case conside s human- o-machine and machine- o-machine ype communica ion wi h high mobili y making use o eal- ime emo e p ocessing.  Use mobili y equals he ehicle speed. The ca - o-ca communica ion becomes highly ele an in e.g. au oma ed d i ing. Table C-10: KPIs and equi emen s o UC5. Va iable/pa ame e Value (u ban/ u al/highway) Pe o mance a ge s o a ic e iciency and sa e y Expe ienced use h oughpu 1600 by e x 10 Hz, i.e., 128 kbps T a ic olume densi y 0.1 / 0.01 / 0.01 Gbps/km2 End- o-end la ency (including connec ion se up and de ec ion delay) o ecei e s wi hin he a ge ange 5 ms A ailabili y ≈ 100% Reliabili y 99.999% Numbe o deli e ed message To be maximized V2X Range 50 m, 500 m, 1 km Posi ioning accu acy Rela i e accu acy less han 0.5 m Pe o mance a ge s o eal- ime emo e compu ing Expe ienced use h oughpu 100 [Mbps] in downlink 20 [Mbps] in uplink T a ic olume densi y 60 [Gbps/km2] ( o ca s on a highway) La ency Less han 10 [ms] E2E la ency Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 112  OPEX wi h high e ansmission ees. In he u u e, sa elli e ope a o s migh e ol e owa ds p o iding backhaul/ on haul se ices o e es ial ne wo ks o b oadband se ices. The backhaul o mobile ne wo ks is cu en ly mos ly achie ed h ough ibe , e es ial and mic owa e solu ions. As mo e and mo e backhaul op imiza ion echnologies a e equi ed o suppo he g owing a ic, sa elli e becomes mo e and mo e a iable op ion. Wi h new sa elli es, he cos o sa elli e backhaul is expec ed o dec ease signi ican ly. The ela i ely high la ency expe ienced in GEO sa elli e communica ions could also be educed wi h LEO sys ems as his is a basic equi emen o 5G. O he wise he 5G la ency a ge s canno be achie ed. The e o e he e may be a di e en ia ion in he business model (xMBB o emo e a eas wi hou o he in as uc u e (p obably no in main pa s o Eu ope), bu wi h limi ed se o all 5G ea u es). In addi ion, a ailabili y could also be enhanced wi h hese new LEO sa elli es. Sa elli e ope a o s could also mo e owa ds he igh side o he sa elli e alue chain and adop a mul icas app oach in combina ion wi h e es ial mobile ne wo ks. Sa elli e ne wo ks (MEO, GEO and LEO) can be eally help ul d i e he ma ke o MTC in emo e, widesp ead, hos ile o global a eas ( hings can be ae onau ical pla o ms including ai c a s, ma i ime pla o ms, essels…). Table D-2: WiFi, LPWA, sa elli e playe s ole. Playe Cu en posi ioning Possible e olu ion WiFi access p o ide Ope a e WiFi access poin s Sells wholesale and e ail se ices Manages subsc ibe ela ionships Small cells ope a ion mainly on unlicensed spec um, wi h possible in eg a ion wi h 5G ne wo ks Small cell as a se ice (SCaaS) LPWA (Low Powe Wide A ea) ope a o s Deploy and ope a e dedica ed ne wo ks o IoT Possible in eg a ion wi h 5G ne wo ks (cell si e sha ing, RAN in eg a ion? in eg a ion o unlicensed bands?) Sa elli e ope a o Limi ed p o ision o na owband mobile se ices wi h LEO cons ella ions (e.g. I idium, Inma sa ) Possible o e ing in on haul/backhaul in eg a ion wi h e es ial ne wo ks. Use o non- GEO cons ella ions wi h educed la ency. Mul icas app oach in combina ion wi h e es ial ne wo ks. Use o MTC in u al a eas Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 113  P i a e wi eless ne wo k owne . This playe is a co po a e unning a close subsc ibe g oup (CSG) adio ne wo k, p o iding p op ie a y connec i i y, se ices and applica ions, usually wi h addi ional access o public ne wo ks. Towe companies, acili y manage s, u ban u ni u e manage s, equipmen manu ac u e s ( o RAN ou sou cing) In his sec ion, we analyze playe s owning asse s which can be use ul o mobile ne wo k co e age and capaci y ex ensions. Towe companies, acili y manage s, u ban u ni u e manage s and equipmen manu ac u e s all in o his g oup. Towe companies Towe companies such as Ame ican Towe (USA), C own Cas le (USA), AT&T Towe s (USA), China Towe Company (China) o A qi a (Uni ed Kingdom) a e in posi ion o go u he han hei cu en ole o en ing space o MNOs‟ base s a ions and could ope a e pa o he RAN. In he Uni ed Kingdom, A qi a is able o p o ide MNOs wi h access o o e 400,000 s ee asse s, including lamppos s and CCTV came as. CAPEX and OPEX o owe companies a e he ollowing:  CAPEX o acqui e si es, build owe s;  Minimal CAPEX o ope a e (ligh ing sys em and ence epai …), inc ease numbe o clien s (heigh ex ension, mul iple an ennas in eg a ion, sha ed gene a o , s eng hened ounda ion…);  OPEX o main ain and ope a e owe s (g ound en , insu ance, eal es a e axes…); On he o he hand, owe companies en owe s o MNO wi h long- e m con ac s ( ixed powe and uel cos / adi ional pass- h ough cos ). Cha ges a e based on loca ion, e ical space, weigh placed on owe … In he u u e, some owe companies migh conside ope a ing he RAN o a MNO h ough an ou sou cing a angemen in compe i ion wi h equipmen manu ac u e s. Towe companies could also bid o spec um du ing o hcoming auc ions and build RANs ei he o co e u al a eas wi h low equency bands o in o de o p o ide capaci y wi h highe equency bands in u ban a eas. Facili y manage s Facili y manage s such as building o s adium manage s a e gene ally in ol ed in mobile co e age/capaci y h ough DAS (Dis ibu ed An enna Sys ems) indoo ope a ion. They usually pay a specialized company o DAS ins alla ion. Capaci y is sold o elecom playe s. DAS allows ope a o s and acili y manage s o wo k closely. Facili y manage s will no doub inc ease expe ise and expe ience in dealing wi h DAS. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 114 In he u u e, expe ienced and skilled acili y manage s migh e ol e owa ds deploying DAS ne wo ks and ins alling DAS in o he acili ies and enues. They can also ope a e small cells o o e access in a building using mobile in as uc u e wi h backhauling ia a ibe op ic a he bo om o he building. CAPEX and OPEX o acili ies manage s a e he ollowing:  CAPEX o buy buildings;  Minimal CAPEX o ope a e;  OPEX o main ain buildings and p ope ies and ope a e owe s (g ound en , insu ance, eal es a e axes…). In he u u e, acili y manage s can bene i om hei s a egic posi ion and mo e om he p o ision o DAS o p o ision o small cell based connec i i y. Facili y manage s a e expec ed o be willing o expand hei business and u he mone ize p ope ies (p o ision o a dedica ed mobile in as uc u e in he building wi h comple e backhauling ne wo k). U ban u ni u e manage s Today, small cells a e being deployed on u ban s ee u ni u e such as s ee lamps, u ili y poles, bus s ops, benches o billboa ds h ough pa ne ships be ween equipmen endo s and ad e ising companies. Deploymen o small cells in s ee u ni u e is expec ed o enhance co e age and ne wo k capaci y in highly popula ed a eas o mee he u u e cus ome demand. U ban u ni u e manage s build, own and ope a e s ee u ni u e. They p o ide u nkey solu ions o MNOs and ope a ion & main enance se ices. A numbe o ag eemen s we e signed in he las mon hs 2015 be ween s ee u ni u e p o ide s and small cells endo s and MNOs. JCDecaux is one o he i s o in es in he small cell echnology. I leases i s s ee u ni u e o small cell manu ac u e s. CAPEX and OPEX o u ban u ni u e manage s a e he ollowing:  CAPEX o build s ee u ni u e;  Minimal CAPEX o ope a e;  OPEX o main ain and upg ade s ee u ni u e;  OPEX o ope a e (change ads, clean…). In he u u e, u ban s ee u ni u e manage s migh be willing o mo e up in he alue chain. They migh e ol e owa ds p o iding and ope a ing base s a ions and small cells hemsel es. Equipmen manu ac u e s A ew yea s ago, elecom ope a o s acknowledged ha hey could no be p esen in e e y segmen o he alue chain and c ea e mo e alue by se ing up pa ne ships wi h la ge-scale playe s. This leads o inno a ions o changes in model o managemen such as, o example, ou sou cing passi e o ac i e in as uc u e o sha ing ixed o mobile ne wo ks. I was abou inding he bes way o op imizing and secu ing a e u n on hei asse s and, in pa icula , he ne wo k. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 115 Ope a o s con inue oday o ou sou ce hei ne wo k and ield ope a ions also known as managed se ices, o la ge elecom equipmen manu ac u e s. As compe i ion among elecom playe s is s ill inc easing, ope a o s may be emp ed o ou sou ce mo e pa s o hei ne wo ks o e en ually o le equipmen manu ac u e s ope a e hei ne wo ks. CAPEX and OPEX o equipmen manu ac u e s a e he ollowing:  CAPEX o de elop, design and build equipmen on a la ge scale;  OPEX o main ain equipmen ;  OPEX o p o ide managed se ices;  OPEX o manage elecom ne wo ks;  CAPEX o p o ide managed se ices. In he u u e, equipmen manu ac u e s could be in a posi ion o ully ope a e mobile ne wo ks. They could be emp ed o inc ease hei alue p oposi ion and p o ide mobile se ices on a wholesale basis (MVNOs). Table D-3: O he RAN playe s ole. Playe Cu en posi ioning Possible e olu ion Towe company Build, own, ope a e, main ain owe asse s Build cus omized/g een si es Ren owe s o MNO ( ixed powe and uel cos / adi ional pass- h ough cos ) Could ope a e he RAN/ pa s o RANs Facili y manage Ope a e DAS (mainly indoo en i onmen ) Could ope a e small cells (SCaaS) Could o e elecoms access in a building using mobile in as uc u e wi h backhauling ia a ibe op ic a he bo om o he building U ban u ni u e manage Build, own and ope a e u ban u ni u e Ren esou ces on a pay-as-you-go model Could ins all base s a ions, ope a e pa s o RAN (small cells) Equipmen manu ac u e s Ou sou cing and se ice managemen Ope a ion o wholesale mobile ne wo k IoT/MTC playe s IoT (In e ne o Things)/MTC (Machine Type Communica ions) playe s co e ed in his sec ion a e in eg a o s, IT companies and e ical ma ke playe s. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 116 In eg a o s & IT companies Sys em in eg a o s o so wa e companies including SAP o IBM, ha e s ong capabili ies in da a and in so wa e, especially in analy ics. As he alue chain is disassembling in many ma ke s and as ha dwa e is now commodi ized (mo e powe ul and cheape ), in eg a o s e- assemble ce ain unc ions and p o ide packaged se ices in a “so wa iza ion” mo e. IT companies a e mainly posi ioned on he p o essional ma ke s (M2M and In e ne o Objec s) on he IT side and da a managemen h ough hei cloud se ices (based on hei da a cen e a ms). They aim o apply hei expe ise in cloud and big da a o he di e en IoT ma ke s hey wan o add ess. They also wan o le e age hei adi ional indus ial clien ele o help hem o e ol e hei business ans o ma ion, made possible h ough connec ed de ices ollou . Ve ical ma ke playe s Ve ical ma ke playe s encompass ca manu ac u e s in he au omo i e business, hospi als/doc o s in he heal h indus y…, ha will be pa o he enla ged elecom ecosys em in he nea e m as elecom is sp eading in many e icals. The e icals will de elop new se ices, new p ocesses and new business models, hanks o obus , a o dable ansmission and au oma ed collec ion o da a. In he u u e, e ical ma ke playe s will ha e o adap o hese adical changes. They could jump in o he elecom business and p o ide speci ic connec i i y se ices associa ed wi h he e ical. In he ca indus y, he ca could be conside ed as an addi ional – la ge – de ice connec ed o he In e ne . Gene ally, p oduc s a e expec ed o lose hei place a he cen e o he alue chain. The p oduc will become a senso in a la ge alue chain. Table D-4: IoT/MTC playe s posi ioning. Playe Cu en posi ioning Possible e olu ion In eg a o s S ong capabili ies in da a and so wa e Connec i i y p o ision IT companies P o ide Cloud se ices Le e age on da a o p o ide be e se ices and solu ions Ve ical ma ke playe s Buy connec i i y om MNOs Play a ole in se ice c ea ion & ope a ion Public Sa e y and PPDR (Public P o ec ion and Disas e Relie ) playe s Cu en non-b oadband dedica ed public sa e y sys ems mainly use he 400 MHz and he 700- 800 bands wo ldwide. Spec um abo e 1 GHz suppo s also a a ie y o PPDR ope a ions o empo a y use only. In e ms o usage, ha monized alloca ions and echnologies o exis ing Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 117 TETRA/TETRAPOL o simila sys ems (DMR, APCO 25) a e sui able and eliable o na owband da a, and echnical speci ica ions ha e been ex ended o enable wideband da a ansmission. A numbe o hese PPDR ne wo ks ha e been deployed all o e he wo ld. Vi e in Finland and Ai wa e in he UK a e among he oldes ones a Eu opean le el. BDBOS in Ge many is he wides public sa e y ne wo k in Eu ope and he newes one: i has been comple ed ea ly 2015. TETRA-like na owband echnologies ha e se ed PPDR issues h ough dedica ed PPDR ne wo ks using PPDR spec um ex emely well o e he pas decade. As hese ne wo ks a e by na u e na owband, hey only suppo low expe ienced use h oughpu s. They seem inadequa e o public sa e y wi h g owing high b oadband da a and ideo needs which equi e si ua ional awa eness and ideo li e s eaming. The e is now a clea global consensus ha LTE will be he baseline echnology o nex - gene a ion b oadband public sa e y ne wo ks. LTE s ill needs o be adap ed: as om Release 12 o 3GPP LTE s anda ds, LTE will be enhanced o mee public sa e y applica ions equi emen s. Release 12 includes basic PPDR ea u es. I s eeze, howe e , has been sligh ly pos poned and some PPDR ea u es o me ly scheduled in Release 12 will be deal wi h in Releases 13 and 14. A numbe o examples in he wo ld demons a e he end owa d LTE in public sa e y. In he USA, Fi s Ne will be a la ge-scale public sa e y ne wo k using dedica ed spec um in he 700 MHz band which is expec ed o agg ega e local public sa e y “Fi s Ne -compa ible” ne wo ks. Canada announces a qui e simila decision. In Eu ope, Belgium, Finland and he UK a e he i s ones o ein en public sa e y ne wo ks. Va ious o ms o co-ope a ion be ween MNOs and public sa e y can be conside ed: he use o comme cial ne wo ks, ad-hoc ne wo ks, secu e- MVNO (as he Belgian ASTRID i ual b oadband MVNO, Blue Ligh Mobile en s capaci y om he h ee na ional comme cial MNOs; p io i y is gi en by P oximus (a Belgian MNO) o Blue Ligh Mobile subsc ibe s o e p i a e subsc ibe s), and hyb id solu ions a a ious s ages. In e ms o spec um, he use o a dedica ed po ion o he 700 MHz band o he use o comme cial spec um has no been decided ye . CAPEX and OPEX o public sa e y ne wo k ope a o s a e he ollowing:  CAPEX o build a secu e ne wo k ha mee s high equi emen s o public sa e y;  OPEX o ope a e he ne wo k;  OPEX o en capaci y om o he ne wo ks;  OPEX o nego ia e speci ic condi ions (p io i y, SLA…). In he u u e, dedica ed public sa e y ne wo k ope a o s migh e ol e owa ds “de-specializa ion” o hei asse s. They could p o ide in-excess capaci y o comme cial playe s du ing non-c i ical hou s. Secu e-MVNO could also sell in-excess capaci y o comme cial mobile ope a o s. Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 118 Table 6D-5: Public sa e y playe s posi ioning. Playe Cu en posi ioning Possible e olu ion Dedica ed Public Sa e y Build, own and ope a e Public Sa e y / PPDR ne wo ks P o ide unused capaci y du ing non- c i ical hou s o ope a ion o 5G ope a o s Public sa e y MVNO Sell mobile se ices o end-use s Ren capaci y o 5G ope a o s OTTs, Con en owne s OTTs The well-known OTT playe s a e se ing hei cus ome s o e he In e ne ( oice h ough VoIP and messaging a e he wo i s se ices p o ided). They a e p o iding an in eg a ed p oduc o e he In e ne and bypasses adi ional dis ibu ion. Se ices ha come o e he op a e, i no always, lowe in cos han he adi ional me hod o deli e y. QoS o e ed by OTTs is gene ally lowe han he one o e ed by MNOs. They a e also inc easingly elian on WiFi. Apple‟s iMessage can be ou ed o e WiFi ne wo ks – ee o he use and he company – while oice calls can be made o e he In e ne on WiFi da a connec ions using se ices such as Skype. In addi ion, Apple un eiled he iPhone 6 in Sep embe 2014 which ea u e WiFi calling. OTTs a e agg ega ing he sepa a e communica ion ma ke s, p o iding a ious ypes o communica ion h ough hei single pla o m. Such a end means ha in heo y, use s can subsc ibe o jus one IP communica ion p o ide o co e all communica ions. CAPEX and OPEX o OTTs a e he ollowing:  CAPEX o buy, design, build se e s. OTT in es men s a e ocused on se e s mainly a he han ne wo k in as uc u e.  CAPEX o design se ices in some cases;  OPEX o main ain se e s. Some OTTs, among hem Google and Facebook, a e clea ly conside ing an en y in o he cellula business. OTTs may e ol e owa ds p o iding mobile se ices. They ha e a key s eng h: hey do no ely on subsc ibe s o see e enues g ow. They a e able o d i e down p ices o consume s and adop p ac ices ha would be di icul o sus ain by o he ca ie s. Con i ming he announcemen s made du ing he MWC 2015, Google p esen ed on i s blog he “explo a o y” mobile es -se ice named “GoogleFi” in 2015. The new o e elies p ima ily on he a ailabili y o he +1m ho spo s e i ied and alida ed. The “GoogleFi” use is able o use he Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 119 +1million Google ho spo s e i ied and alida ed and ha e access o cellula se ices supplied by Sp in and T-Mobile. Looking sligh ly ahead, OTTs a e well posi ioned o ac as in e media ies o u u e se ices. By aking he pla o m app oach, allowing hi d pa ies o build hei inno a i e applica ions on op o hem, o h ough indi idual pa ne ships, hey can ac as he link be ween use s and u u e inno a ion. As an example, Apple's Si i applica ion combines speech ecogni ion echnologies wi h he use s' pe sonal in o ma ion, answe ing ques ions and making ecommenda ions sui able o he use s' ques ions. He e, Si i ac s as he in e media y be ween he use and o he playe s' con en s and se ices. Ano he example would be webRTC, enabling oice and ideo communica ion h ough any web b owse ; hi d pa ies will hen be able o build applica ions on op o webRTC. Con en owne s Con en owne s a e speci ically in ol ed in con en c ea ion and adap a ion o mobile de ices. They a e esponsible o he accu acy o he con en . The consump ion o con en has g adually e ol ed om a model mainly based on owne ship o con en (physical media o a digi al copy) o a model based on access o a ee (ad-suppo ed) and/o subsc ip ion se ice. Dema e ializa ion had he e ec o dec easing he uni p ice o con en . Wi h dema e ializa ion and i ualiza ion, he echnical building blocks o mul i-ne wo k, mul i- de ice dis ibu ion a e s eadily coming in o place, hanks o p og am pos -p oduc ion, ne wo k con ol and in e connec ion, in e ace managemen and con en s o age unc ions being mo ed o he cloud. This makes ope a ional managemen simple , inno a ion as e and dis ibu ion cos s ha a e be e adjus ed o a pay-as-you-go app oach. CAPEX and OPEX o con en owne s a e he ollowing:  CAPEX o c ea e, design con en and se ices associa ed wi h he con en ;  CAPEX o upg ade con en o new de ices/pla o ms;  CAPEX & OPEX o dis ibu e he con en ;  OPEX o manage he con en . In he u u e, con en owne s migh be emp ed o cash in by jumping in he access side. TV owne s a e al eady jumping in he In e ne side. Table D-6: Con en owne s and OTT posi ioning. Playe Cu en posi ioning Possible e olu ion Con en owne s C ea ion o he con en , adap a ion o mul iple de ices No expec ed ole modi ica ion OTTs P o ision o se ices h ough he MVNO p o iding mobile Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 120 in e ne ee o cha ge communica ions se ices O he playe s De ice manu ac u e s Ob iously, de ice manu ac u e s a e designing and ma ke ing de ices di ec ly o use s and o elecom ope a o s. De ices a e s ill cen al o he elecom ma ke and a e con inuing o de elop in e ms o :  size and sc een esolu ion,  memo y capaci y o s o ing con en ,  chip in eg a ion o ne wo k sha ing. CAPEX and OPEX o de ice manu ac u e s a e he ollowing:  CAPEX o c ea e, design de ices and adap o speci ic needs;  CAPEX o upg ade de ices;  OPEX o adap o e ol ing needs;  OPEX o manage he de ice en i onmen (applica ions). In he u u e, de ices a e expec ed o e ol e in many aspec s (size, o m ac o , unc ion…) and will suppo many echnologies (WiFi, Blue oo h, 3G/4G/5G …) and spec um bands. Fo m ac o s will no doub be e y di e se, spanning om pe sonal de ices such as sma phones and able s, machine de ices, wea ables…, mainly depend on usage. Looking ahead, e-SIM ca ds buil in o he phone will become mains eam. Basically he p inciple o he embedded SIM e e s o SIM ca ds welded in o he modules a manu ac u ing. The possibili y o changing ope a o s in e-SIMs aises he ques ion o subsc ip ion managemen in he de ice. The cus ome will no mo e be ied o a single elecom ope a o and will be able o easily swi ch p o ide s. Telecom ope a o s a e hus expec ed o ha e mino con ol o e he end-use . De ice manu ac u e s could di ec ly sell a package wi h bo h de ice and connec i i y, and as a consequence manage he cus ome ela ionship. Connec i i y could be included pa ly (gi en da a olume included in a pe iod o ime) o en i ely (da a olume is o e ed ee o cha ge du ing he de ice li e ime). F om he elecom ope a o poin o iew, inno a i e da a p icing schemes should also become popula in a nea u u e, such as he on-demand connec i i y based on embedded SIM echnology, ideal o sho - ime jou neys, weekending o aca ioning ab oad, o ins ance. In- ligh connec i i y p o ide s (on comme cial ai lines) Today, in- ligh connec i i y is p o ided h ough ai - o-g ound and sa elli e ne wo ks. In- ligh connec i i y is p o ided on a wholesale basis ( he ai line cha ges o no passenge s o he Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 121 se ice and he se ice p o ide cha ges a se p ice pe plane and a gi en numbe o connec ions), on a e ail basis (bo h he se ice p o ide and he ai line decide on he p ice o cha ge o he passenge and sha e e enues.), o on he Gogo model, an hyb id ATG/Ku in- ligh connec i i y p o ide . Gogo ins alls equipmen (and suppo s all ins alla ion cha ges) and he ai line pays o he connec i i y se ice (no up- on in es men s). Re enues a e sha ed bu Gogo ge s he la ge pa . CAPEX and OPEX o in- ligh connec i i y p o ide s a e he ollowing:  CAPEX o ins all equipmen and build he ne wo k;  CAPEX o upg ade de ices (equipmen on-boa d ai c a s);  OPEX o adap o e ol ing needs;  OPEX o pay sa elli e connec i i y and ATG segmen ;  OPEX o manage cus ome ela ionship (in he Gogo example). In he u u e, in- ligh connec i i y p o ide s migh manage he cus ome ela ionship o he se ice du ing he ligh and migh e ol e owa ds managing i on he g ound. Table D-7: O he playe s posi ioning. Playe Cu en posi ioning Possible e olu ion De ice manu ac u e s Sell handse s o cus ome s and mobile ope a o s Wi h i ualized SIM ca d, de ice manu ac u e s could manage cus ome s ela ionship In- ligh connec i i y p o ide s Sell in- ligh connec i i y di ec ly o he use o h ough he ai line In- ligh connec i i y p o ide s could manage he use ela ionship widely  Na ional Regula o y Au ho i y (NRA). Public au ho i y o go e nmen dealing wi h adminis a i e law, egula ion ulemaking, in he con ex o his documen in he a ea o adio communica ions. D.1.2 De ailed alue chains in he mobile sec o The u ban s ee u ni u e alue chain Today, small cells a e being deployed on u ban s ee u ni u e such as s ee lamps, u ili y poles, bus s ops, benches o billboa ds h ough pa ne ships be ween equipmen endo s and Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 128 MTC Resou ces in he con ex o Sma Ci y When i comes o he ole o ICT in Sma Ci y, a se o esou ces enables he MTC ac o s o pa icipa e and pe o m di e en se s o ac i i ies. The impo ance o hese esou ces lies in he ac ha possession o any o hese esou ces enables playe s o pe o m a speci ic ac i i y. By a esou ce i is mean any hing which could be hough o as a s eng h o weakness o a gi en i m. Mo e o mally, a i m‟s esou ces a a gi en ime could be de ined as hose ( angible and in angible) asse s which a e ied semi pe manen ly o he i m. Examples o esou ces a e: b and names, in-house knowledge o echnology, employmen o skilled pe sonnel, ade con ac s, machine y, e icien p ocedu es, capi al, e c. De ined pedagogically, esou ces can be ca ego ized in o six majo ca ego ies: inancial, physical, human, echnological, o ganiza ional, and epu a ion.  MTC In as uc u e When p o iding communica ion se ices, he need o communica ions ne wo ks appea s na u ally. Wi hin communica ion ne wo ks wo di e en ypes can be iden i ied: Co e Ne wo k and Cellula Access Ne wo k. The co e ne wo k is he cen al pa o he communica ions ne wo k, acili a ing he connec ion be ween di e en sub-ne wo ks. The cellula access ne wo k (also de ined as adio access ne wo k) is he in e ace be ween he end-use and he co e ne wo k, basically using wi eless echnology. The MTC In as uc u e is adi ionally owned by a MNO, since i is he same as he mobile elephony cellula in as uc u e. In he in oduced cases, we also ha e seen eme ging playe s who a e specialized MTC Ne wo k Ope a o s who own hei own in as uc u e.  Connec ed De ice Pla o m CDPs a e so wa e elemen s ha acili a e he deploymen and managemen o connec ed de ices o M2M applica ions o e cellula ne wo ks. CDP allows de ices o connec o Cloud and should be compa ible wi h di e en so wa e pla o ms (e.g. Ja a, And oid, e c.) in o de o include as many de ices as possible. CDP is usually a se ice po al, which should co e billing and policy con ol, bea e se ice, se ice o de ing and subsc ip ion, and SIM-ca d managemen .  Applica ion Enablemen Pla o m A so wa e pla o m ha ac s as a common g ound o de elopmen o se ices and applica ions on op o he physical in as uc u e. AEP can also p o ide an open en i onmen o collabo a ion be ween indus ies and suppo inno a ion in he con ex o sma sus ainable ci ies.  M2M Da a One e y impo an esou ce when in oducing Sma Ci y and M2M se ices is he da a o igina ed o he End Use s on he ci y. Da a can be de ined as all he in o ma ion ob ained om he usage o a numbe o se ices in he ci y en i onmen ; communica ion/In e ne se ices, anspo a ion se ices, ene gy consump ion, ca -sha ing, pa king o logis ics. The Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 129 added alue in Sma Ci y comes om ob aining a big amoun o da a, p ocess he da a and ex ac use ul in o ma ion o decision making in he ci y.  End Use The inal goal o hese se ices is o p o ide use ul in o ma ion and se ices o he End Use , which will be able o make be e decisions on how o in e ac wi h he ci y. In he p o isioning o se ices, a numbe o playe s a e in ol ed and i is no easible ha he End Use has ela ion wi h hem all. The usual ela ions wi h he use a e wi h ei he he se ice p o ide o he M2M de ice p o ide (in some cases). In his sense, he di e en s akeholde s a e sha ing his esou ce e en hough no all o hem ha e di ec ela ion wi h i . I could be concluded ha cus ome s a e he economic esou ce which a e subjec o be cul i a ed by he p oduce MTC playe s in Sma Ci y Since MTC e e s o u ilizing cellula echnologies as he access ne wo k o M2M se ices, adi ional playe s in he mobile elephony alue ne wo k a e s ill ha e key ole in MTC alue ne wo k. In his ega d, he MNOs as he ypical ca ie s ha con ol and ope a e cellula ne wo ks a e capable o ope a ing he MTC ne wo k. Ne wo k Equipmen P o ide (NEP as he adi ional manu ac u e s o he elecommunica ion equipmen , ypically p o ision he echnical p ocu emen o he MNOs. Bu , acco ding o he shi in he alue chain, he NEPs ha e ecen ly pa icipa ed in di e en oles ha his o ically ha e been assigned o aken ca e o by o he s, such as MNOs. O e en oles like p o isioning new demands such as Connec i i y Pla o ms. Based on p oposed desc ip ions o he cases and he gene al se up o o e ing any se ice o he End Use (cus ome ), i e majo g oups o playe s can be iden i ied in he MTC alue ne wo k. Besides he End Use s- hese playe s a e he mos likely en i ies who can own ei he o he esou ces men ioned ea lie in o de o pe o m he ac i i ies. These playe s hen a e: a) End Use , b) Se ice P o ide , c) MTC Ne wo k Ope a o , d) Se ice P o ide , and e) Managed Se ice P o ide . Acco ding o he cases s udied, we showed ha he e a e o he playe s a he han adi ional Telecom playe s who migh be e en mo e compe en in p o isioning any o he ac i i y blocks o MTC (besides MNOs and TEVs). Fo ins ance, a specialized M2M cellula ne wo k ope a o (MTC Ne wo k Ope a o ) can be conside ed a be e op ion o p o ision MTC ne wo k. Se ice P o ide s o M2M solu ions also in some cases ake con ol o he en i e alue chain by handling he End Use s; a p e iously dominan posi ion o MNOs (in Mobile Telephony). On he o he hand, TEVs and MNOs ha e shown in e es in di e en ac i i y blocks. Ano he majo playe in his se up is hen an en i y which pe o ms he ole o p o isioning CDPs. I can be seen ha his ac i i y is mainly pe o med by he i ms who ha e a his o ic backg ound in Documen : METIS-II/D1.1 Ve sion: 1.0 Da e: 2016-01-31 S a us: Final Dissemina ion le el: Public 130 p o isioning connec i i y in he sense o au oma ing connec ed de ices. Some examples can be ei he ou sou cees o ne wo k ope a ions o MNOs o he ones which ha e been ac i e in au oma ion o indus y e icals (e.g. Gene al Elec ics, Siemens, e c.).