Workshop V – The Road to 5G Rami Yaron, NEC/Netcracker Scott Mansfield, Ericsson

Topics What is new in 5G? Transport for 5G Network Slicing 1 What is new in 5G? 2 Transport for 5G 3 Network Slicing 4 Orchestrating 5G Services 5 MEF Services over 5G

Enhanced Mobile Broadband 2017-11-02 5G Is Use-Case Driven Massive MTC Critical MTC LOGISTICS TRAFFIC SAFETY & CONTROL SMART AGRICULTURE FLEET MANAGEMENT INDUSTRIAL APPLICATION & CONTROL REMOTE TRAINING REMOTE MANUFAC- TURING SMART METER TRACKING REMOTE SURGERY LOW COST, LOW ENERGY SMALL DATA VOLUMES MASSIVE NUMBERS Enhanced Mobile Broadband ULTRA RELIABLE VERY LOW LATENCY VERY HIGH AVAILABILITY Enterprise VR/AR Home Mobile/ Wireless/ Fixed Smartphones Broadcasting Venues Non-SIM devices 4k/8k UHD Technical expectations of new 5G performance levels are high and has potential to deliver unparalleled benefits to society and business Peak data rate Maximum achievable data rate under ideal conditions per user/device peak data rate of IMT-2020 for enhanced Mobile Broadband is expected to reach 10 Gbit/s. under certain conditions and scenarios IMT-2020 would support up to 20 Gbit/s User experienced data rate Achievable data rate that is available ubiquitously across the coverage area to a mobile user/device In wide area coverage cases (urban and sub-urban areas) a user experienced data rate of 100 Mbit/s is expected to be enabled. In hotspot cases, the user experienced data rate is expected to reach higher values Spectral efficiency Average data throughput per unit of spectrum resource and per cell Mobility Maximum speed at which a defined QoS and seamless transfer between radio nodes which may belong to different layers and/or radio access technologies (multi-layer/-RAT) can be achieved Latency The contribution by the radio network to the time from when the source sends a packet to when the destination receives it Connection density Total number of connected and/or accessible devices per unit area Network energy efficiency Energy efficiency refers to the quantity of information bits transmitted to/ received from users, per unit of energy consumption of the radio access network Energy consumption for the radio access network of 5G should not be greater than existing networks deployed today, while delivering the enhanced capabilities The network energy efficiency should therefore be improved by a factor at least as great as the envisaged traffic capacity increase Area traffic capacity Total traffic throughput served per geographic area Availability The network availability is characterized by its availability rate X, defined as follows: the network is available for the targeted communication in X% of the locations where the network is deployed and X% of the time. Battery life Battery life of device connected to the 5G network Reliability The amount of sent packets successfully delivered to the destination within the time constraint required by the targeted service, divided by the total number of sent packets. Note that the reliability rate is evaluated only when the network is available. Position accuracy Ability to determine network based positioning in three-dimensional space 3 © Ericsson AB 2017

Technical Expectations of 5G 2017-11-02 Technical Expectations of 5G Peak Data Rate 10k - 1m devices / km2 99.999% (of packets) 1 - 20 Gbps Connection Density Reliability User Experienced Data Rate 10 - 100 Mbps Network Energy Efficiency Position accuracy ×1 - ×100 10m - <1m Strong subscriber authentication, user privacy and network security Spectral Efficiency Area Traffic Capacity 0.1 - 10 Mbps / m2 ×1 - ×3 Security 350 - 500 km/h 99.999% (of time) Mobility Availability Latency 1 - 10 ms Battery life 10 years* *For low power IoT devices Source: ITU-R, NGMN, 3GPP Technical expectations of new 5G performance levels are high and has potential to deliver unparalleled benefits to society and business Peak data rate Maximum achievable data rate under ideal conditions per user/device peak data rate of IMT-2020 for enhanced Mobile Broadband is expected to reach 10 Gbit/s. under certain conditions and scenarios IMT-2020 would support up to 20 Gbit/s User experienced data rate Achievable data rate that is available ubiquitously across the coverage area to a mobile user/device In wide area coverage cases (urban and sub-urban areas) a user experienced data rate of 100 Mbit/s is expected to be enabled. In hotspot cases, the user experienced data rate is expected to reach higher values Spectral efficiency Average data throughput per unit of spectrum resource and per cell Mobility Maximum speed at which a defined QoS and seamless transfer between radio nodes which may belong to different layers and/or radio access technologies (multi-layer/-RAT) can be achieved Latency The contribution by the radio network to the time from when the source sends a packet to when the destination receives it Connection density Total number of connected and/or accessible devices per unit area Network energy efficiency Energy efficiency refers to the quantity of information bits transmitted to/ received from users, per unit of energy consumption of the radio access network Energy consumption for the radio access network of 5G should not be greater than existing networks deployed today, while delivering the enhanced capabilities The network energy efficiency should therefore be improved by a factor at least as great as the envisaged traffic capacity increase Area traffic capacity Total traffic throughput served per geographic area Availability The network availability is characterized by its availability rate X, defined as follows: the network is available for the targeted communication in X% of the locations where the network is deployed and X% of the time. Battery life Battery life of device connected to the 5G network Reliability The amount of sent packets successfully delivered to the destination within the time constraint required by the targeted service, divided by the total number of sent packets. Note that the reliability rate is evaluated only when the network is available. Position accuracy Ability to determine network based positioning in three-dimensional space 4 © Ericsson AB 2017

Critical MTC: Communications Distance vs. Latency 2017-11-02 Critical MTC: Communications Distance vs. Latency Process automation 1 s Inter-substation comm. 500 ms 200 ms Tele-surgery Substation-internal comm. Latency ~150 ms Remote handling w/o haptic feedback 100 ms Hot rolling mill control Latency Automated guided vehicle Remote handling with haptic feedback (e.g. remote mining) 20 ms Robot manufacturing cell/roundtable Autonomous driving 10 ms 3 ms 1 ms Drive Control (packaging, printing) 100 us Wind turbine- internal High speed motion control 2 m 10 m 100 m 1 km 10 km 100 km Technical expectations of new 5G performance levels are high and has potential to deliver unparalleled benefits to society and business Peak data rate Maximum achievable data rate under ideal conditions per user/device peak data rate of IMT-2020 for enhanced Mobile Broadband is expected to reach 10 Gbit/s. under certain conditions and scenarios IMT-2020 would support up to 20 Gbit/s User experienced data rate Achievable data rate that is available ubiquitously across the coverage area to a mobile user/device In wide area coverage cases (urban and sub-urban areas) a user experienced data rate of 100 Mbit/s is expected to be enabled. In hotspot cases, the user experienced data rate is expected to reach higher values Spectral efficiency Average data throughput per unit of spectrum resource and per cell Mobility Maximum speed at which a defined QoS and seamless transfer between radio nodes which may belong to different layers and/or radio access technologies (multi-layer/-RAT) can be achieved Latency The contribution by the radio network to the time from when the source sends a packet to when the destination receives it Connection density Total number of connected and/or accessible devices per unit area Network energy efficiency Energy efficiency refers to the quantity of information bits transmitted to/ received from users, per unit of energy consumption of the radio access network Energy consumption for the radio access network of 5G should not be greater than existing networks deployed today, while delivering the enhanced capabilities The network energy efficiency should therefore be improved by a factor at least as great as the envisaged traffic capacity increase Area traffic capacity Total traffic throughput served per geographic area Availability The network availability is characterized by its availability rate X, defined as follows: the network is available for the targeted communication in X% of the locations where the network is deployed and X% of the time. Battery life Battery life of device connected to the 5G network Reliability The amount of sent packets successfully delivered to the destination within the time constraint required by the targeted service, divided by the total number of sent packets. Note that the reliability rate is evaluated only when the network is available. Position accuracy Ability to determine network based positioning in three-dimensional space Communication distance 5 © Ericsson AB 2017

5G-Enabled Digitalization Revenues for ICT Players 2017-11-02 5G-Enabled Digitalization Revenues for ICT Players +181% +22% • Agriculture • Retail • Automotive • Media &Entertainment • Public transport • Healthcare • Financial Services • Manufacturing • Public Safety • Energy & Utilities “Normal market growth” “Ramp up” $US Billion 5G enabled digitalization revenues for ICT players Revenues generated by new industry digitalization value-producing opportunities that are created or enhanced by the introduction of 5G networks Technical expectations of new 5G performance levels are high and has potential to deliver unparalleled benefits to society and business Peak data rate Maximum achievable data rate under ideal conditions per user/device peak data rate of IMT-2020 for enhanced Mobile Broadband is expected to reach 10 Gbit/s. under certain conditions and scenarios IMT-2020 would support up to 20 Gbit/s User experienced data rate Achievable data rate that is available ubiquitously across the coverage area to a mobile user/device In wide area coverage cases (urban and sub-urban areas) a user experienced data rate of 100 Mbit/s is expected to be enabled. In hotspot cases, the user experienced data rate is expected to reach higher values Spectral efficiency Average data throughput per unit of spectrum resource and per cell Mobility Maximum speed at which a defined QoS and seamless transfer between radio nodes which may belong to different layers and/or radio access technologies (multi-layer/-RAT) can be achieved Latency The contribution by the radio network to the time from when the source sends a packet to when the destination receives it Connection density Total number of connected and/or accessible devices per unit area Network energy efficiency Energy efficiency refers to the quantity of information bits transmitted to/ received from users, per unit of energy consumption of the radio access network Energy consumption for the radio access network of 5G should not be greater than existing networks deployed today, while delivering the enhanced capabilities The network energy efficiency should therefore be improved by a factor at least as great as the envisaged traffic capacity increase Area traffic capacity Total traffic throughput served per geographic area Availability The network availability is characterized by its availability rate X, defined as follows: the network is available for the targeted communication in X% of the locations where the network is deployed and X% of the time. Battery life Battery life of device connected to the 5G network Reliability The amount of sent packets successfully delivered to the destination within the time constraint required by the targeted service, divided by the total number of sent packets. Note that the reliability rate is evaluated only when the network is available. Position accuracy Ability to determine network based positioning in three-dimensional space Source: Ericsson 6 © Ericsson AB 2017

On the Road to 5G Low latency RAN Virtualization Massive MIMO 2017-11-02 On the Road to 5G 2016 2017 2018 2019 2020 3GPP Rel-14 Rel-15 Rel-16 Rel-17 Early deployments 5G new Carrier Type, NR Low latency RAN Virtualization Massive MIMO Massive IoT LTE Advanced 7 © Ericsson AB 2017

Evolution Towards 1Gbps LTE 2017-11-02 Evolution Towards 1Gbps LTE 1000 MBPS cat-16 600 MBPS cat-11 450 MBPS cat-9 300 MBPS cat-6 150 MBPS cat-4 3CC + 256QAM (20+20+20) 3CC 2CC (20+20) (10+10) 2CC 4x4MIMO + 1CC 2x2MIMO (256QAM) 5CC (256QAM) LAA 1 licensed 4x4MIMO + 3 unlicensed CC 2x2MIMO (256QAM) 8 © Ericsson AB 2017

LTE, LTE-E, and NR LTE and LTE-A LTE-E NR 4G 5G 2017-11-02 LTE, LTE-E, and NR 4G 5G LTE and LTE-A LTE-E NR Backward compatible with LTE and LTE-A Unlicensed Low latency Lean design concepts Massive MTC Vehicular Not backward compatible with LTE and LTE-A High frequency bandwidth Large bandwidth Massive beamforming /MU-MIMO Ultra low latency Ultra lean Critical MTC New spectrum -- migrate to existing Up to 3GPP Release 12 Up to 8x8 MIMO Carrier aggregation up to 1Gbps Services VoLTE eMBMS Mid-tier MTC Enhanced dual connectivity Split architecture 9 © Ericsson AB 2017

3GPP R15 – 2 Architecture Tracks for New Radio 2017-11-02 3GPP R15 – 2 Architecture Tracks for New Radio 3GPP Target Q4 17 5G Enabled EPC NextGen Core S1-based New interface Option 1 Option 3 Option 5 Option 2 Option 7 Option 4 Xn Xn LTE NR LTE NR LTE NR/EPC NR/NXGC LTE NR/EPC LTE 10 © Ericsson AB 2017

Why New Functional Splits? Splitting RAN functions Reduce bandwidth demands Enable virtualized RAN Seamless radio resource management Radio coordination & interference mitigation Increase deployment flexibility Availability of transport and sites

5G RAN Network Higher Splits Lower Splits 3GPP 5G C-RAN 3GPP tbd Multiple Functional Splits between RRH and BBU based on Transport and RAN efficiency tradeoff Different Functional Split has different KPIs 12

5G Transport – Dimensioning Examples 2017-11-02 5G Transport – Dimensioning Examples Radio gNB (v)EPC 5G D-RAN (e)CPRI 10-25G/sector < 75µs (e)S1 1-10G < 10ms Radio gNB (v)EPC (e)CPRI 10-25G/sector < 75µs 5G C-RAN (e)S1 10-40G < 10ms Radio DU CU (v)EPC eMBB with V-RAN (e)CPRI 10-25G/sector < 75µs F1 1-10G < 5ms (e)S1 10-40G < 10ms Radio DU CU (v)EPC Example dimensions are traffic model based All latency numbers in the slide are one way Throughput is related to carrier bandwidth and MIMO layers C-MTC with V-RAN (e)CPRI < 30µs (indicative) 5G D-RAN Fronthaul with short distance (e)CPRI E.g., eCPRI 7G/sector is average load for TDD config 2, 64T/R, 16layers, 60Mhz (15G/sector for peak load) 5G C-RAN Fronthaul with long distance (e)CPRI (e)S1 throughput dimension is dependent on the No. of BB on the site eMBB with V-RAN RPU preferably deploy at antenna/hub site (e)S1 throughput dimensions are dependent on the No. of RPU connected to vPP/vRC C-MTC with V-RAN D-RAN, C-RAN can also support C-MTC RRU/RPU/vPP,vRC/EPC needs to be co-located ultra low latency (1ms e2e) Antenna Hub Central Office Switching/Aggr © Ericsson AB 2017

Evolved Architecture Evolved Architecture options: Distributed RAN 2017-11-02 Evolved Architecture L1 L3 L2 high CPRI 5G NR Distributed RAN Architecture Centralized RAN Architecture Centralized Layer 3, common mobility anchor Core Network L2 Low L2 low Evolved Architecture options: Distributed RAN Centralized RAN Virtualized RAN 5G NR RAN 14 © Ericsson AB 2017

5G Main Components & Their Evolution 2017-11-02 5G Main Components & Their Evolution Wireless Access LTE evolution NB-IoT NR Massive MIMO Transport Resource Differentiation RAN Transport Interaction Fronthaul Backhaul 5G Cloud Virtual Data Center NFV SDN PaaS Network Applications Cloud Enabled Distributed Deployment Cloud Native Scalability Management Orchestration Analytics Automation Security Source: Ericsson 15 © Ericsson AB 2017

MEF Work Areas Transport for 5G Network Slicing Orchestrating 5G Services MEF Services over 5G 1 2 3 4

Transport Evolution – Backhaul & Fronthaul Considerations 2017-11-02 Transport Evolution – Backhaul & Fronthaul Considerations Slicing Capacity Latency Security Coordination Radio Access Network Evolved Packet Network # sites Thousands Hundreds Synchronization Sites 17 © Ericsson AB 2017

5G Vision and Requirements User Experience: User Data Rates: 0.1 to 1Gbps Peak Data Rates: 20Gbps High-speed Mobility: 500 Km/h End-to-End Latency: ms level Network Impact: Connection density: 1M connections/Km2 Traffic Volume Density: Tens of Tbps/Km2 Spectrum Efficiency: 3x* Network Efficiency: 100x* *Compared to IMT-Advanced 5G Key Capabilities – User Experience and Network Impact! Figure: ITU-R WP 5D 18

RAN Coordination Requirements 2017-11-02 RAN Coordination Requirements Optimal coordination with co-located basebands Loose coordination with any transport Tight coordination with high performance transport High capacity sites Network wide coverage Capacity sites Distributed RAN Optimal coordination with dedicated fiber transport Virtualized RAN ** Transport capacity and one-way latency C-RAN Radio coordination opportunities Fronthaul: 2.5/5/10/25/(50) Gbps ≤ 75 µs Inter-site: 10/40 Gbps ≤ 23-30 µs Inter-site: BH dimensioning* ≤ 5 ms Inter-site: BH dimensioning* ≤ 30 ms **Optional *Capacity based on backhaul dimensioning 19 © Ericsson AB 2017

Network controller / gateway site 2017-11-02 MEF 22.2 Terminology Network controller / gateway site Remote radio site Base Station site Fronthaul Backhaul RRU Marco RBS SGW,MME Midhaul Backhaul Small cell site Small RBS Backhaul – macro/small cell to core Midhaul – small cell to macro Fronthaul – remote radio to baseband unit 20 © Ericsson AB 2017

Fronthaul Split Options Tradeoff Between Bandwidth and Latency 2017-11-02 Fronthaul Split Options Tradeoff Between Bandwidth and Latency High Layer (HL) or high latency splits Low Layer (LL) or low latency splits RRC PDCP High - Low - High - Low - High - DL Low - PHY RF RLC RLC MAC MAC PHY Data Option 1 Option 2 Option 3 Option 4 Option 5 Option 6 Option 7 Option 8 High - Low - High - Low - High - UL RRC PDCP RLC RLC MAC MAC PHY Low - PHY RF CPRI 3GPP F1 eCPRI Data ~10 fold reduction of required bandwidth compared to CPRI Use of packet transport technologies like Ethernet/IP enabled Required bandwidth can scale flexibly according to the user plane traffic, sharing the network with other traffic possible. Split supports fast control mechanisms for improved radio performance (COMP, Beamforming) 21 © Ericsson AB 2017

MEF 22 Phase 4 - Transport for 5G 5G RRU DU CU (BBU) MFH2 MFH3 CU/DU MFH1 MBH CORE EBH 5G RRU DU CU (BBU) CU/DU MFH1 MFH2 MFH3 DU Distributed Unit CU Centralized Unit Functional Splits can be in 1 or 2 stages (RRH-DU-CU) Three types of Fronthaul Networks Each has its own KPIs RRH, CU, DU, and Core distributed based on Services KPIs Multiple configurations can be co-located Transport Network can provide support for all MEF Ethernet Service Types – Mobile Fronthaul 1,2,3, Mobile Backhaul, Ethernet Backhaul Multiple Ethernet Service Types can co-exist 22

Three Alternative Network Scenarios Modeled 2017-11-02 Three Alternative Network Scenarios Modeled Comparison of three potential network scenarios for deployment of new services One big network Separate networks Network slicing Though it is recognized that network slicing can extend end-to-end across the radio and transport network domains, this study was restricted to evaluating the impacts of network slicing in the core network only. 23 © Ericsson AB 2017

Benefits of Network Slicing

Overall Economic Impact of Network Slicing Network slicing enables: New revenue generation Lower opex Greater capex efficiency Overall impact is a significantly increased economic benefit through new service launches.

Transport Network Slicing – System View MEF Ethernet Service based on EVCs Ethernet Service Slice is a group of Ethernet Services (EVC) A new Group of Ethernet Service is now introduced MFH1 Slice MFH2 MFH3 MBH EBH

Transport Network Slicing Representation Slice Representation Examples - A Group of EVCs Single S-Tag = EVC, Enhanced MEF tools (e.g., Trunk/OVC+, Envelope+) to represent Slice ID Single S-Tag – Higher order bits=Slice ID, Lower order bits=EVC Double S-Tag – Inner Tag=EVC, Outer Tag=Slice ID PBB – S-Tag=EVC, B-Tag+I-Tag=Slice ID MPLS – S-Tag=EVC, MPLS Label=Slice ID Slice EVC 1 EVC 2 EVC 3 EVC 4 27

3GPP Network Slicing – E-2-E Service View 5G America White Paper Network Slicing for 5G Network & Services RRH DU CU Fronthaul is inside the RAN Slice between RRH and BBU or between RRH & DU and DU and CU(BBU) Backhaul is between RAN and Mobile Core Network or between RAN and Wireline Network for Wireline Services There is an East-West association between RAN, Core, and Transport 28

5G Network End-to-End Orchestration RRH DU CU (BBU) MFH MBH EBH CORE UE RF RAN Transport LSO E-2-E LSO Mobile LSO Service Network Resource E-2-E LSO = Mobile LSO + Transport LSO MEF LSO -> Transport LSO MEF LSO extends to E-2-E LSO and/or Mobile LSO 29

Data-Driven Orchestration 2017-11-02 Data-Driven Orchestration Network Slice Catalogue : Service.. n Mobile Broadband Wire-line Internet access Health Care …... …... Health Nomadic Broadband Enterprise Communication …… …… Robotic communication Industry Automation Massive Sensors/Actuator Premium Communication Media Network Service Composition MBB Basic Logical Networks Network Slice Resources: Access/Mobility Service Provider Core Transport Management & Control Service Provider IT Cloud Applications Cloud Infrastructure Access Mobile Fixed Physical Resources (Access, Connectivity, Computing, Storage, ..) 30 © Ericsson AB 2017

Orchestrating 5G Services Ensuring 5G-based services are orchestratable through SDN controllers as part of : Heterogeneous connectivity service Multi-Operator Multi-Technology Full service lifecycle Network resource provisioning Service OAM and SAT Service assurance (e.g. Zero touch telemetry, closed loopback control) OpenCS 5G project Defining use cases, epics and user stories Use case -> Information Model -> Data Model -> standardized open northbound APIs for 5G environments

5G Ready Core @ VNF MEF LSO - Management & Orchestration & Analytics & Exposure Virtualization Software Defined Networking (SDN) Distributed Cloud CORE & RAN Network Slicing Multi-access with new 5G RAT, Fixed Wireless Broadband, Wi-Fi and FTTx SDN based Transport with options for white label switches/routers Local and Private Core Networks User-Plane(s) for various use cases and deployment options (e.g. low latency) 32

MEF Services Over 5G – Why? More and more traffic is now terminated on mobile devices Customers would like the same SLA at the office, home, or traveling Fiber availability is limited Copper plants are old and deteriorating Optimized for low bandwidth T1/E1 replacement Low CAPEX (FedEx vs. Technician) 33

Advantages of MEF Services Over 5G Service on demand (time to deliver = Revenue) Fast activation Out-of-band management for SDN networks Supporting IoT use case 34

MEF Services Over 5G MEF services Carrier Ethernet IP Network slicing allows customer separation for EVC end-to-end 5G adds native support for IP or Ethernet frames/service at the UE Service Provider End-to-End EVC Mobile Network RAN BS site Mobile Network RAN NC site Service Provider Core site CEN UNI UNI Backhaul EVC RAN CE RAN CE

Summary 5G is much more than a new wireless technology 1 5G is much more than a new wireless technology 2 Multiple transport options can support 5G interfaces 3 Application driven networks calls for network slicing 4 LSO can facilitate orchestration of mobile and fixed networks 5 MEF Services can run fully/partially over 5G

Workshop V – The Road to 5G Rami Yaron, NEC/Netcracker Scott Mansfield, Ericsson