© 2015 AT&T Intellectual Property. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Intellectual Property.
AT&T Proprietary (Internal Use Only). Not for use or disclosure outside the AT&T companies except under written agreement.
Intersection of 5G and Open
Reference Platforms
Tom TOFIGH, PMTS , AT&T
Driving Force for 5G and Beyond
Disruptive Network Connectivity
Better, Faster, Flexible, Elastic Access , edge cloud and Services
Emerging Operating Models
– Modular Capabilities
– Open Source Platforms
– Economically Viable Eco Systems
– Digital Traceability / Dynamic Analytics
– AI/ ML , AR/VR, Robotics, Flying objects
– Predictive & Scenario Driven
– Micro Services & Third Party Applications
Empowered Customers
– Seamless data sharing
– Remote Diagnostics
– Increase Trust and Transparency
– Enable Human and Machine Collaboration
– Lower Operating Cost
– Predictive Maintenance
IC
Vendor
ODM OEM Operator
Content
Supplier
Artificial wall
Rise of Open RAN -> Unified SD-RAN
Separation of Control and User plane  SD-RAN
PDCP
High RLC
Low RLC
High MAC
Low MAC
High PHY
Low PHY
Radio/MIMO
User Plane
Power Control
Frequency/Beam Control
Antenna Configuration
Admission Control
Measurement Reporting
AMF Control
Bearer Control
Scheduling Control (E2)
System Broadcast
SON Control (E2)
SMF Control (Core)
Carrier Agg. Control (E1)
Handover Control (E2)
Paging Control
Slice MLB Control (E2)
RRC Control
SD&C Control Interface
SO Interface (A1)
RAN Slicing Control (E2)
Control Plane
FCAP Control ( NEM)
ICIC Control
Radio Resource Manager -> (SD-RAN) -> ORAN
• Common Control & Configuration for Split Architecture
• Extended Control Configuration options to vendors specific features
• Unified vProbe and Analytic Registration process
• Well Known Functional Interface & APIs
• Measurement Configuration, Control and Reporting
• RRC Measurement Config
• RRC Measurement Report
• L2 Measurement Config
• L2 Measurement Report
• Frequency Selections
• Power Level
• Antenna & Beam Forming Commands
• Handoff Actions
• Scheduling and Slicing RAN Control
• UE Admission Control
• Cell Configuration
What Open Reference Platforms Offer
• Offers new user experiences
• AR/VR , battery life , M2M, Edge clouds,
• Provides for new types of connectivity services
• Seamless, carrier aggregation, QOS, Metering, etc
• Provides for autonomous operational flexibility
• Performance, latency scale, security, Reliability, and programmability
• Performs frequent “control” across the RANs
• hyper densified set of technologies and macro, small and micro cells.
RRC Measurement Config
RRC Measurement Report
L2 Measurement Config
L2 Measurement Report
Frequency Selections
Power Level
Antenna & Beam Forming
Handoff Actions
Scheduling ,
Slicing RAN Control
UE Admission Control
Cell Configuration
Maximize
Throughput
Bigger Pipes
More Efficient
Data Handling
Better Faster
Optimal Pipes
Connectivity
and Mobility
Reconfigurable
Dynamic Pipes
Edge Cloud Solutions That Support Increased Data Usage &
A Richer Multimedia Experience
Leads to Adaptive Analytics for 5G with Open Reference Platforms
“The Services of The Future”
LTE/5G
Collaborative
Communications
Compression
Untapped
Spectrum
5G-6G
Wireless
Local Content
Routing
Femto /
WiFi / LiFi
Cognitive
Spectrum
Management
M2M / IOT
Rise of Open Reference Platform & the Need for Real time Observability
Radio Resource Control Split Architecture ControlSlicing Manager Control
5G Applications
… Applied ML Services
Unified Open Mobility Resource Manager & Interface
(SD-RAN, Fronthaul , SD-CORE)
5G Services
Modular and
Disaggregated
Mobility
CORE Functions
Time critical operations offloaded to FPGA
• Disaggregated control from user planes and micro services
• Common Control & Configuration of RUs, DUs, and CUs
• Dynamic vProbe and applied ML
• Open Interface & APIs
• Context Driven Networking ( Closed feed back loops)
Master Orchestrator [Open Network Automation Platform (ONAP)]
RU CORE
Edge
Cloud
Open ToolsRANDU/CU
Configuration Management Slicing Management Services
CORE …..
Virtualized and democratizedOpen Reference Platform
Modular and
Disaggregated
Virtual RAN
Functions:
DU, CU
Common RAN and CORE
Pipeline
Source: The Linux Foundation/AT&T
RAN
CORE
…. + MEC
Open Source Communities & Standards
Open Reference Platforms
X-RAN/ORAN
ONAP
ONF
TIP
Etc.
On-Going Collaboration Efforts
Open Source
EPC, RAN
Non-Open Source
EPC, RANMicro-Services
Applications
3GPP
IETF
IEEE
MEC
ETSI
Impact of Open Platforms
Resources
CaaSSDN
Ctrl
App
VNF
Controller • Defines Authoritative State
• Mediates Trust
• Enforces Policy/Invariants
• Activates Data Path
Ctrl
App
VNF
Adopts Micro-Services with Control as a Service
Source; ONF
Challenges:
Open and Democratized Architecture know-hows
Graph abstraction Representing Radio Networks (Multi-RATs)
Graph: G = ( N, E, L, B, etc.) or ( nodes, Links, Slices, Beams )
N = set of routers = { u, v, w, x, y, z }
E = set of links ={ (u,v), (u,x), (u,w), (v,x), (v,w), (x,w), (x,y), (w,y), (w,z), (y,z) }
L = set of Slices = {sub ((u,v), sub (u,x), sub(u,w), sub(v,x), sub (v,w), sub (x,w), sub(x,y), sub (w,y), sub (w,z), (y,z) }
B = set of beams at time T = Associated beams for { u, v, w, x, y, z }
Beams
Goals:
• Understand principles behind Abstraction
• Understand principles behind service Layers & Multi Tenancy
• Understand principals behind R-NIB and Trigger Functions
SD-RAN: Programmable RAN Operation Examples
RF
Coordinator of Multi RAT Access technologies
Example: SD-RAN & Micro Services
master
Orchestrators
Optimization
Practices
Mobility
Policy
Carrier
Aggregation
vRAN
Applications
SD-RAN (M-CORD)
Radio Resource
Abstracted graph
X.RAN Open APIs
Local RAN Agent Controls
Telemetry &
Observability
Virtualized RAN Micro Services
Unified Platforms & Blue-Prints
Infrastructure VMs, VNFs, Containers, etc.
TaaS ( active & passive)
Edge Operations Tools Edge Operation Tools
Master Orchestartion such as ONAP (DCAE, Policy, Service Models, etc.)
vRAN
5G Components
Events
Akraino
Tool sets
Edge
Service
Design&Creation
Policy Creation
Analytics
Application Design
• AR/VR ( acceleration)
• Massive IOT ( scale )
• Critical IOT ( Delay sensitive, BER)
Edge Example Services
Real-time Life Cycle Management
Akraino Blue Prints
N
O
D
E
T
O
S
C
A
M
O
D
E
L
PLUGINPLUGIN
Accelerators
Adaptive Analytics Service
Vehicular Slice
vRAN Probe Management
Virtual probe
1
2
3
4
vRAN Software Platform
Model driven Passive Monitoring
Anomaly detection & Closed loop automation
Active Test as a service
1
3
4
RAN VNFs
CORE VNFs
1) Assurance as a Service (Passive)
- Feed UP Session Metrics to vRAN RIC
Monitoring Service
Closed loop automation
Stop
QoS
Active PM Testing
Perf Monitoring
Thresholds
Active Test
as a Service
2) vRAN RIC triggers test request for
connectivity/PM session in an slice
3) Active Virtual Measurement Agents
- Deploy Mgmt. Agent VM instances
- Run TWAMP Perf Monitoring sessions
Service
Slices IoT Device Slice
Start
2
Summary Remarks
• Aggressive spectrum allocations for 5G and beyond will require unified, open and
democratized Interfaces.
• Opening up the interfaces, disaggregating RAN, white box deployment models with
coexistence should be a key focus for service providers and open source community
and standardization bodies.
• ORAN’s Radio Intelligent controller provides coordination and when feasible
synchronization of multi vendor and white boxed radio access. It can boost
performance and quality service consistency.
• Open Source + ML + In-band Radio analytics will play a key role 5G and beyond.
Contact:,
tofigh@att.com

5G and Open Reference Platforms

  • 1.
    © 2015 AT&TIntellectual Property. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Intellectual Property. AT&T Proprietary (Internal Use Only). Not for use or disclosure outside the AT&T companies except under written agreement. Intersection of 5G and Open Reference Platforms Tom TOFIGH, PMTS , AT&T
  • 2.
    Driving Force for5G and Beyond Disruptive Network Connectivity Better, Faster, Flexible, Elastic Access , edge cloud and Services Emerging Operating Models – Modular Capabilities – Open Source Platforms – Economically Viable Eco Systems – Digital Traceability / Dynamic Analytics – AI/ ML , AR/VR, Robotics, Flying objects – Predictive & Scenario Driven – Micro Services & Third Party Applications Empowered Customers – Seamless data sharing – Remote Diagnostics – Increase Trust and Transparency – Enable Human and Machine Collaboration – Lower Operating Cost – Predictive Maintenance IC Vendor ODM OEM Operator Content Supplier Artificial wall
  • 3.
    Rise of OpenRAN -> Unified SD-RAN
  • 4.
    Separation of Controland User plane  SD-RAN PDCP High RLC Low RLC High MAC Low MAC High PHY Low PHY Radio/MIMO User Plane Power Control Frequency/Beam Control Antenna Configuration Admission Control Measurement Reporting AMF Control Bearer Control Scheduling Control (E2) System Broadcast SON Control (E2) SMF Control (Core) Carrier Agg. Control (E1) Handover Control (E2) Paging Control Slice MLB Control (E2) RRC Control SD&C Control Interface SO Interface (A1) RAN Slicing Control (E2) Control Plane FCAP Control ( NEM) ICIC Control
  • 5.
    Radio Resource Manager-> (SD-RAN) -> ORAN • Common Control & Configuration for Split Architecture • Extended Control Configuration options to vendors specific features • Unified vProbe and Analytic Registration process • Well Known Functional Interface & APIs • Measurement Configuration, Control and Reporting • RRC Measurement Config • RRC Measurement Report • L2 Measurement Config • L2 Measurement Report • Frequency Selections • Power Level • Antenna & Beam Forming Commands • Handoff Actions • Scheduling and Slicing RAN Control • UE Admission Control • Cell Configuration
  • 6.
    What Open ReferencePlatforms Offer • Offers new user experiences • AR/VR , battery life , M2M, Edge clouds, • Provides for new types of connectivity services • Seamless, carrier aggregation, QOS, Metering, etc • Provides for autonomous operational flexibility • Performance, latency scale, security, Reliability, and programmability • Performs frequent “control” across the RANs • hyper densified set of technologies and macro, small and micro cells. RRC Measurement Config RRC Measurement Report L2 Measurement Config L2 Measurement Report Frequency Selections Power Level Antenna & Beam Forming Handoff Actions Scheduling , Slicing RAN Control UE Admission Control Cell Configuration
  • 7.
    Maximize Throughput Bigger Pipes More Efficient DataHandling Better Faster Optimal Pipes Connectivity and Mobility Reconfigurable Dynamic Pipes Edge Cloud Solutions That Support Increased Data Usage & A Richer Multimedia Experience Leads to Adaptive Analytics for 5G with Open Reference Platforms “The Services of The Future” LTE/5G Collaborative Communications Compression Untapped Spectrum 5G-6G Wireless Local Content Routing Femto / WiFi / LiFi Cognitive Spectrum Management M2M / IOT
  • 8.
    Rise of OpenReference Platform & the Need for Real time Observability Radio Resource Control Split Architecture ControlSlicing Manager Control 5G Applications … Applied ML Services Unified Open Mobility Resource Manager & Interface (SD-RAN, Fronthaul , SD-CORE) 5G Services Modular and Disaggregated Mobility CORE Functions Time critical operations offloaded to FPGA • Disaggregated control from user planes and micro services • Common Control & Configuration of RUs, DUs, and CUs • Dynamic vProbe and applied ML • Open Interface & APIs • Context Driven Networking ( Closed feed back loops) Master Orchestrator [Open Network Automation Platform (ONAP)] RU CORE Edge Cloud Open ToolsRANDU/CU Configuration Management Slicing Management Services CORE ….. Virtualized and democratizedOpen Reference Platform Modular and Disaggregated Virtual RAN Functions: DU, CU Common RAN and CORE Pipeline Source: The Linux Foundation/AT&T RAN CORE …. + MEC
  • 9.
    Open Source Communities& Standards Open Reference Platforms X-RAN/ORAN ONAP ONF TIP Etc. On-Going Collaboration Efforts Open Source EPC, RAN Non-Open Source EPC, RANMicro-Services Applications 3GPP IETF IEEE MEC ETSI
  • 10.
    Impact of OpenPlatforms
  • 11.
    Resources CaaSSDN Ctrl App VNF Controller • DefinesAuthoritative State • Mediates Trust • Enforces Policy/Invariants • Activates Data Path Ctrl App VNF Adopts Micro-Services with Control as a Service Source; ONF
  • 12.
    Challenges: Open and DemocratizedArchitecture know-hows Graph abstraction Representing Radio Networks (Multi-RATs) Graph: G = ( N, E, L, B, etc.) or ( nodes, Links, Slices, Beams ) N = set of routers = { u, v, w, x, y, z } E = set of links ={ (u,v), (u,x), (u,w), (v,x), (v,w), (x,w), (x,y), (w,y), (w,z), (y,z) } L = set of Slices = {sub ((u,v), sub (u,x), sub(u,w), sub(v,x), sub (v,w), sub (x,w), sub(x,y), sub (w,y), sub (w,z), (y,z) } B = set of beams at time T = Associated beams for { u, v, w, x, y, z } Beams Goals: • Understand principles behind Abstraction • Understand principles behind service Layers & Multi Tenancy • Understand principals behind R-NIB and Trigger Functions
  • 13.
    SD-RAN: Programmable RANOperation Examples RF Coordinator of Multi RAT Access technologies
  • 14.
    Example: SD-RAN &Micro Services master Orchestrators Optimization Practices Mobility Policy Carrier Aggregation vRAN Applications SD-RAN (M-CORD) Radio Resource Abstracted graph X.RAN Open APIs Local RAN Agent Controls Telemetry & Observability Virtualized RAN Micro Services
  • 15.
    Unified Platforms &Blue-Prints Infrastructure VMs, VNFs, Containers, etc. TaaS ( active & passive) Edge Operations Tools Edge Operation Tools Master Orchestartion such as ONAP (DCAE, Policy, Service Models, etc.) vRAN 5G Components Events Akraino Tool sets Edge Service Design&Creation Policy Creation Analytics Application Design • AR/VR ( acceleration) • Massive IOT ( scale ) • Critical IOT ( Delay sensitive, BER) Edge Example Services Real-time Life Cycle Management Akraino Blue Prints N O D E T O S C A M O D E L PLUGINPLUGIN Accelerators
  • 16.
    Adaptive Analytics Service VehicularSlice vRAN Probe Management Virtual probe 1 2 3 4 vRAN Software Platform Model driven Passive Monitoring Anomaly detection & Closed loop automation Active Test as a service 1 3 4 RAN VNFs CORE VNFs 1) Assurance as a Service (Passive) - Feed UP Session Metrics to vRAN RIC Monitoring Service Closed loop automation Stop QoS Active PM Testing Perf Monitoring Thresholds Active Test as a Service 2) vRAN RIC triggers test request for connectivity/PM session in an slice 3) Active Virtual Measurement Agents - Deploy Mgmt. Agent VM instances - Run TWAMP Perf Monitoring sessions Service Slices IoT Device Slice Start 2
  • 17.
    Summary Remarks • Aggressivespectrum allocations for 5G and beyond will require unified, open and democratized Interfaces. • Opening up the interfaces, disaggregating RAN, white box deployment models with coexistence should be a key focus for service providers and open source community and standardization bodies. • ORAN’s Radio Intelligent controller provides coordination and when feasible synchronization of multi vendor and white boxed radio access. It can boost performance and quality service consistency. • Open Source + ML + In-band Radio analytics will play a key role 5G and beyond.
  • 18.