5G Technology Elements for Future Internet of Things

Dr. Geng Wu

Chief Scientist, Wireless Standards and Advanced Technologies

geng.wu@intel.com
The future of Internet of Things

Applications Ecosystem

M2M
Vertical Wearables

IoT
Horizontal
1x 10x - 100x
Smart phones Smart things

Future IOT requires new network capabilities to Future IOT requires close partnership among many
support massive number of devices, and efficient industries and ecosystems to offer integrated
SW/HW platforms for both devices and networks compute and communication capabilities

2
IOT expectation gaps

Source Bain: Reflects n=37 interviews with tier 1 leaders in Smart Factory, Smart Building, Smart Fleet, Smart Grid Substation and Smart City Environmental Sensing domain

3
Enabling future IOT

Technologies Standards and regulations

*IDC
**IMC/EDC: The digital universe of opportunities

IOT requires end-to-end design considerations, from Standards collaboration and regulatory compliance
sensors to the cloud are essential

4
5G design goals
5G
IoT 4G mmWave
Data rate Can be very low 100s Mbps 1-10 Gbps
Latency From 2 ms to hours Moderate 2 ms
Battery life Up to 10 years Smart phone Depends on app
Link budget Penetration Coverage High data rate
Multiple access Massive # of devices Smart phones Bursty/opportunistic
Network Overlay Hetnet Hetnet/underlay

5G has to be designed as a system

Compute + communication becomes a necessity

5
Air interface for IOT
5G
Machine-Machine interaction 4G
Latency 1s ms
Direct link or tactile Internet

Wearable underlay networks Human-Human Ultra-high data rate

Varying latency req. Human-Machine High peak to average
Ad-hoc networks Interaction Short range, opportunistic
Het-net carrier aggregation
Latency 10s ms
Sensor Network
Latency often less an issue
Massive # of devices

Platform innovations enable applications/services innovations

6
 Underlay networks for things and wearables

Mesh Data on WiFi Direct

Discovery and

One big intelligent and information network

control on LTE
D2D Cluster #4

Compute, storage, networking

D2D
Coordinator
Pico/Home
Pico/Home eNB
eNB
D2D Unicast
Link
D2D Cluster #0
Macro
Spatial Reuse in Multi-hop
eNB
One Cell Area D2D
Coordinator D2D Unicast
Link
D2D Unicast D2D D2D Cluster #3
Link Broadcast
D2D Cluster #1 Link
D2D Smart Apps
Coordinator
D2D Cluster #2
D2D Broadcasting

Many devices, types of devices, connections

Many moving underlay network clusters Context-based
Storage Processing

7
Virtualization of device and access

5G
Device + Access
Compute Platform

Virtualization
Performance

5G

Portable
Smart
Feature Phone
Phone Wearable

Time

5G device+access virtualization for scalability, versatility, energy efficiency

• Future applications require intensified compute and • 5G services are immersive. Sensing, intelligence and contents
communication but often smaller device form factor require edge cloud and device + access virtualization
• 5G high data rate + low latency radio links enable mobile • Breaking computing barrier through communication may
device + access virtualization across the air interfaces transform consumers’ relationship with network

8
 Future research areas

Peer to peer
Remote
user access M2M
System design
Social Server
network  Compute and communications tradeoffs
MVNO
System Intfc
 Device and access network architecture
tradeoffs
Wide Area network
Local user
 Underlay networks and connectivity design
(cellular or fixed line) view/control

Gateway Gateway tradeoffs

P2P gateway
comms
“medium”
range network
User experience
Star, mesh, tree,
point-to-point
Aggre- Aggre-
gation
Local network
 Privacy and security
gation

 System manageability
Underlay networks “Indirect” Sensors “Direct” Sensors  Subscriber ownership

9
The evolution towards a world of IOT

10
Mobile Edge Computing

Torsten Musiol
© MECSware GmbH 2017
What is Mobile Edge Computing?
Many standardization bodies, consortia, fora, etc. have their own view…

• Wikipedia:
“Mobile Edge Computing (MEC) is a network architecture
concept that enables cloud computing capabilities and an
IT service environment at the edge of the cellular network.”

• MEC
- Mobile Edge Computing
- Mobile Edge Cloud
- Multi-Access Edge Computing
2 © MECSware GmbH 2017
Classical mobile network architecture
Centralized datacenters for Core Network and (Cloud) Application Servers
Mobile Edge

User Equipment Radio Access Network Core Network Application Servers

(Base Stations)
3 © MECSware GmbH 2017
Long data path – non-deterministic QoS
Acceptable for many cloud-based IoT applications

User Equipment Radio Access Network Core Network Application Servers

(Base Stations)
4 © MECSware GmbH 2017
Industry 4.0 goes wireless
Multiple radio technologies are competing

Requirements
• Campus area coverage
• Deterministic QoS
• High-availability
• Strong security
• Seamless mobility

Most demanding use cases require licensed spectrum.

Many use cases can be implemented with LTE – some require 5G.

5 © MECSware GmbH 2017

Industrial M2M applications
Relocating the Base Station doesn’t solve the problem

Industrial Campus The data path needs to be

shortened and controlled!

Machine

Controller Radio Access Network Core Network

(Base Stations)
6 © MECSware GmbH 2017
The hybrid approach – controlled by the operator
Local User-plane, centralized Control-plane
Multiple variants
Industrial Campus • “S1 bump-in-the-wire”
• Distributed S/P-GW
• Control and User Plane Separation (CUPS)
• Network Slicing (5G)
Control

Machine

Controller Radio Access Network Core Network

(Base Stations)
7 © MECSware GmbH 2017
The private LTE approach – controlled by the enterprise
Local User-plane and Control-plane

Industrial Campus

Small
Cells

Machine
App App App App

Control
Computing Storage Networking

MECS Platform
Controller

8 © MECSware GmbH 2017

Central cloud vs. edge cloud
Holistic approach required
• Common programming model
• Security, privacy

App App App App App App App App App

Computing Storage Networking Computing Storage Networking

9 © MECSware GmbH 2017

Role allocation Operator
Different operability and business models Enterprise

WLAN Private LTE Not desired ? ? Public LTE

Radio resource management

Subscriber management

Network element management

Service management

User device management

10 © MECSware GmbH 2017

The complexity challenge Operator
Make LTE as easy as WLAN Enterprise

WLAN Private LTE

Radio resource management

Subscriber management
Plan
Network element management Build
Service management Run
User device management

11 © MECSware GmbH 2017

Vodafone 5G Mobility Lab – opened on 30.08.2017
Mobile Edge Cloud Server (MECS) + 4 LTE Small Cells

MECS

Source: Google

12 © MECSware GmbH 2017

Vodafone 5G Mobility Lab
Live demos

1) Smart Intersection
2) See-through truck
3) Geo-messaging
4) Security information

Source: Vodafone

13 © MECSware GmbH 2017

Other use cases

• Public safety
• Automotive
• Mining
• Construction
•…

14 © MECSware GmbH 2017

LTE goes local!

Picture Credits: Pixabay

© MECSware GmbH 2017