5G Masterclass series:

Towards 6G

Prof. Bart Smolders

Center for wireless technology Eindhoven

Department of Electrical Engineering

Eindhoven University of Technology

Content

• Trends in wireless communication

• What is 5G and 5G New Radio?

• Power consumption dilemma in 5G New Radio

• Looking forward towards 6G

2

Trends in Wireless communications

3

Trend 1: Increase of operational frequency

41900 1920 1940 1960 1980 2000 2020

10-3

10-2

10-1

100

101

102

Year

Fre

quency [

GH

z]

Frequency versus year of introduction

TV

GSM

Satellite

TV

AM

FM

Car radar

60 GHz

WLAN

Trend 2: Increase in bandwidth:Edholm’s Law

Source: IEEE spectrum, 2004 5

Required Bandwidth/datarate doubles each 18 months

Wireless growing faster than wired

1st GENERATION wireless network

• Basic Voice service

• Analog Based Protocols

2.4 Kbps

1G~1980

The foundation of

mobile telephony

2nd GENERATION wireless network

• Designed for voice

• First digital standards (GSM,CDMA)

64 Kbps

2G~1990

Mobile telephony

for everone

3rd GENERATION wireless network

• Designed for voice and data

• First mobile broadband

• Voice through circuit & Data-Packet Switching

2 Mbps

3G~2000

The foundation of

mobile broadband

4th GENERATION wireless network

• Designed Primarily for Data

• IP based protocol

• True Mobile broadband.

100 Mbps

4G~2010

Mobile broadband

enhanced

5th GENERATION wireless network

• 1000x increase in capacity

• Support for 100+ billion connections

• Below 1ms latency

10 Gbps

5G~2020

Embracing a

networked society

6th GENERATION wireless network

• Extension to (sub) mm-wave frequencies

• Real-time cloud computing

1 Tbps

~2030

6G

Enabling a smart

sustainable society

Evolution of wireless standards

6

Compare with Edholms law: factor 2 more BW in 1.5 year 2^(10/1.5)=101

What is 5G?

7

Requires New Radio (NR)

at mm-waves

What is 5G?

• Source: 5G PPP Architecture Working Group, “View on 5G Architecture (Version 2.0),” July 2017 , www.5g-ppp.eu 8

5GURLLC

Ultra-Reliable and Low Latency Communications

Very high reliability

mMTCMassive Machine

Type CommunicationsLow-cost, low-energy

eMBBExtreme Mobile Broadband

High date rate, high-traffic volume

Radio Access in 5G

• Source: 5G PPP Architecture Working Group, “View on 5G Architecture (Version 2.0),” July 2017 , www.5g-ppp.eu 9

Gradual migrationFrom 2020 onwards

LTE Evolution New Radio (NR)

1 GHz 10 GHz3 GHz 30 GHz 100 GHz 1 GHz 10 GHz3 GHz 30 GHz 100 GHz

3.5 GHz30 GHz

“mm-wave New Radio”

NR System embeddingOption 1: Non-standalone operation with LTE master

• Source: 5G PPP Architecture Working Group, “View on 5G Architecture (Version 2.0),” July 2017 , www.5g-ppp.eu 11

Data

LTE

Ctrl

Data

NR

Power consumption in 5G/6Gusing New-Radio

13

Key question

• Can we scale our existing 3G/4G base-station infrastructure to higherfrequencies?

• What happens if we scale from 2 GHz to 30 GHz?

• Let us consider only the downlink (TX) case

14

Existing base-stations for 3G/4G Wireless Communication

• Frequency 0.8-2.5 GHz

• Power consumption ~ 5kW

• Netherlands: 43933 base stations (2G/3G/4G)1

• So currently about 219 MW power dissipation

15• 1 https://www.antennebureau.nl/onderwerpen/algemeen/antenneregister

Spherical wave expansionfrom point source

R

𝑃𝑡

𝑆 =𝑃𝑡

4𝜋𝑅2Power density at surface sphere

Pt: total radiated power

Downlink, Link Budget

17

Pr , Gr

R

Pt , Gt

22

2

0

2 )4(4 R

GGP

R

AGPP rtet tt

r

==

min,

2

2

0

)4( r

rt

P

GGPR t

=

Simple “back-of-the-envelope” calculation, 2 GHz vs. 30 GHz

18

2 GHz

Wavelength: 0=15cm4 Tx antenna elements with 3dB gain each: Gt=8 (9dBi)Rx antenna omni-directional: Gr=1.Output power: 100 W (50 dBm)Sensitivity Pr,min = -70 dBm (=10-10W)

kmP

GGPR

r

rtt 33)4( min,

2

2

0 =

30 GHz

Wavelength: 0=1cm4 Tx antenna elements with 3dB gain each: Gt=8 (9dBi)Rx antenna omni-directional: Gr=1.Output power: 100 W (50 dBm)Sensitivity Pr,min = -70 dBm (=10-10W)

kmP

GGPR

r

rtt 2.2)4( min,

2

2

0 =

Simple “back-of-the-envelope” calculation, 2 GHz vs. 30 GHz

19

30 GHz

Wavelength: 0=1cm4 Tx antenna elements, Gt= 9dBiRx antenna omni-directional: Gr=1.

Output power: 22500 W (73.5 dBm)Sensitivity Pr,min = -70 dBm (=10-10W)

2 GHz

Wavelength: 0=15cm4 Tx antenna elements, Gt = 9dBiRx antenna omni-directional: Gr=1.Output power: 100 W (50 dBm)Sensitivity Pr,min = -70 dBm (=10-10W)

kmP

GGPR

r

rtt 33)4( min,

2

2

0 =

kmP

GGPR

r

rtt 33)4( min,

2

2

0 =

Consequence in Power consumptionEstimations!

20

3G/4G 5G

Frequency 2 GHz 30 GHz

Power per BST* 4+1=5 kW 4+225*5=1129 kW

Total Power in Netherlands* 219 MW 49 GW

#Wind mills (2 MW each) in NL 109 24500

1In May 2017, The Netherlands hosted 43933 2G/3G/4G antennes, see https://www.antennebureau.nl/onderwerpen/algemeen/antenneregisterAt this moment the PA efficiency is about a factor 5 worse at 30 GHz as compared to 2GHz. 10% of RF energy effectively radiated

Uplink case

• Uplink might be even more important than downlink at mm-wave.

• Link budget is not symmetrical!

• Mobile user does not have a lot of power or space.

• We need a large antenna at the base-station with electronic scanning

21

Solution: Use a large antenna array

22

30 GHzWavelength: 0=1cm871 Tx antenna elements, Gt=32.4 dBiRx antenna omni-directional: Gr=1.Output power: 100 W (50 dBm)Sensitivity Pr,min = -70 dBm (=10-10W)

30 GHzWavelength: 0=1cm4 Tx antenna elements, Gt= 9dBiRx antenna omni-directional: Gr=1.Output power: 22500 W (73.5 dBm)Sensitivity Pr,min = -70 dBm (=10-10W)

2 GHzWavelength: 0=15cm4 Tx antenna elements, Gt = 9dBiRx antenna omni-directional: Gr=1.Output power: 100 W (50 dBm)Sensitivity Pr,min = -70 dBm (=10-10W)

kmP

GGPR

r

rtt 33)4( min,

2

2

0 =

kmP

GGPR

r

rtt 33)4( min,

2

2

0 =

kmP

GGPR

r

rtt 33)4( min,

2

2

0 =

Can we come up with solutions in NL?

• The Netherlands has a strong position in:• Phased-arrays, e.g. Thales, ASTRON, TNO, ESTEC and TU’s.

• Silicon-integrated technology and Power amplifiers for base stations, e.g. NXP, TNO, Omniradar, Ampleon and TU’s

23

Phased-array radars from the Netherlands: SMART-L and APAR

PAGE 24

Long range L-band

surveillance

With secondary radar (IFF)

Ref: M.C. Van Beurden, A.B. Smolders, IEEE Trans. AP, 2002, pp.1266-1273

Silicon TechnologiesFt of IC Technology vs Year [ITRS]

26ITRS= International Technology Roadmap for Semiconductors

1990 1995 2000 2005 2010 2015 202010

100

1000

Year

Tra

nsit F

req

uency [

GH

z]

RFCMOS SiGe BiCMOS

Sat

TV

24 GHz

Car radar

60 GHz

WLAN

77 GHz

Car radar

94 GHz

Imaging

NXP

Qubic4Xi

20~30 GHzPoint to point

Do we need more?

• Yes!

• Phased-arrays are too power hungry and too expensive

• Need new antenna concepts

• Further integration in Silicon

• At mm-waves, smaller cells will be used (< 300 m)

27

European project SILIKASilicon-Based Ka-Band Massive MIMO Systems for New Telecommunication Services (5G-NR)

See: www.silika-project.eu 28

Funded by the European Union

Base station cell at mm-waves (28.5 GHz)

29

Scenario: Urban environment

5G New Radio mm-wave Prototype

30/ Electrical Engineering – Electromagnetics group

5G New Radio mm-wave Prototype

31

5G New Radio mm-wave Prototype

32

What about safety in 4G versus 5G?

4G: Based on existing basestation antenna, 10 dipoles, Ptot=50 dBm (100W) Gelem=8 dBi, dual polarization.

5G: Based on EIRP=+65 dBm radiated power, required for a 250 m urban micro cell.

Safety limit: Advies in “Publicatieblad van de Europese Gemeenschappen” L 199/59, 30. 7. 1999 NL.

Looking forward towards 6G

34

Funded by the European Union

Efficient Millimetre-Wave Communications

for Mobile Users (6G)

Kick-Off Meeting01.04.2020

See: www.Mywave-project.eu

Scientific focus MyWave: distributed Massive MIMO

• 36User connected to multiple base stations at same time!

Next step: Feed-array using integrated antennas

Antenna-on-chip (AoC)

• Antenna launcher integrated in Silicon.

• No RF interconnect required anymore

• Wafer thinning can be applied to reduce silicon losses.

Demonstrator in BiCMOS

• AoC+LNA at 30 GHz

37

38

FREEPOWER: Fiber-connected future 5G/6G centralized radio access network (C-RAN)

PAFFPA

FREEPOWER wireless Fronthaul (DU)

Co

re N

etw

ork Central Office (CO)

BBU VNFs Control(pool)

Metro MCF Ringmany cores

A/D

MCF Dropfew cores

A/D

A/D

mm-wave fronthaulalong highway

mm-wave fronthaul/distribution in a factory or event venue

Prof. Ton Koonen

Electro-optical Communication Systems group (ECO)

Institute for Photonic Integration

Dept. Electrical Engineering

Eindhoven University of Technology

Optical Wireless Communication:The benefits of fiber without needing a fiber

European Research Council

[Dominic O’Brien, IPS Summer Topicals 2016]

⚫ Visible Light Communication with wide-coverage beams (<1Gbit/s, shared) ⚫ Beam-steered IR communication (>10Gbit/s, unshared)

Optical wireless communication – basic options

Breaking wireless barriers: free-space beam-steered optical communication

BROWSE’s system concept:

⚫ pencil beams →

no capacity sharing,

long reach

⚫ target:

10Gbit/s per beam

⚫ IR >1400nm →

Pbeam up to 10mW

⚫ passive diffractive beam steerer →

no local powering, easily scalable

⚫ -controlled 2D steering → embedded

control channel

⚫ easily scalable to many beams, just add -s

[Koonen et al, MWP2014]

[Koonen et al., JLT Oct. 2016, JLT May 2018, JLT Oct. 2018]

CCC

OXC

OXC = Optical Crossconnect

CCC = Central Communication

Controller

optical

fiber

access

network

MD = Mobile Device

PRA = Pencil-Radiating Antenna

1221 11

21

11

1112

222121

22

2212

11 22

PRA

MD

fibers

PRA

opticalreceivers

CCC with 2tunable transmitters

PRA

videossent

videosreceived

controllaptop

7 cells captured with IR camera at 2.5m

localization detector

MEMS switch

Lab demonstrator TU/e

⚫ Transfer of two high-def video streams

o Real-time

o Embedded in two 10Gb streams

⚫ Two PRA-s

o 2D array with 128 fibres

⚫ MEMS switch

2D fiber array and assembled beam steerer

Hexagonal 128 fibers array

Co-assembled with 50mmF/0.95 camera lens

Co-assembled with composite C+L band AWGR, 96+48 ports

Summary

• 5G will use mm-wave frequencies • 24.25-43 GHz for base stations in small cells• 71-86 GHz for Front-Haul/Back-Haul

• Existing concepts are too power hungry and far to expensive

• New antenna concepts and high level of integration in Silicon technologies are required.

• 6G will offer even higher capacity• Higher mm-wave frequencies (70-120 GHz)• Optical Wireless using IR or visible light.

44

Thank you !

PAGE

45

Back-up

• Providing 1000 times higher wireless area capacity and more varied service capabilities compared to 2010

• Facilitating very dense deployments of wireless communication links to connect over 7 trillion wireless devices serving over 7 billion people

• Saving up to 90% of energy per service provided. The main focus will be in mobile communication networks where the dominating energy consumption comes from the radio access network

• Reducing the average service creation time cycle from 90 hours to 90 minutes

• Creating a secure, reliable and dependable Internet with a “zero perceived” downtime for services provision

• Ensuring for everyone and everywhere the access to a wider panel of services and applications at lower cost

47 Source: 5G PPP Architecture Working Group, “View on 5G Architecture (Version 2.0),” July 2017 , www.5g-ppp.eu

Key objectives 5G