What is coming next after LTE and IMT-Advanced

What is coming next afterLTE and IMT-Advanced– An Industry Perspective –

Werner MohrHead of Research Alliances

1 © Nokia Siemens Networks 2011

Werner MohrHead of Research Alliances

IEEE Communication Theory Workshop 2011, Sitges, Spain, June 20 to 22, 2011

Outline

• Trends in mobile and wireless communications− Traffic growth and economic impact− Technology

• Requirements on future radio systems• Approach for future radio systems

− Main concepts− Base stations− Ubiquitous connectivity

2 © Nokia Siemens Networks 2011 IEEE Communication Theory Workshop 2011, Sitges, Spain, June 20 to 22, 2011

− Ubiquitous connectivity• Machine-to-Machine / Internet of Things / Sensor-based systems• Energy efficiency• Future directions• Conclusions

The pace of change in the ICT industry

WWW

1st UMTS

IPV6


Prediction is very difficult, especially if it‘s about the future

Network


Source: Nokia Siemens Networks

Bandwidth hunger is ever increasingVideo & TV drive needs

50100 50 100Application bandwidth demands

Multiview 3D SDTV

Multiview 3D HDTV150 200 250 300

Mbps

Ultra HDTV

Super HDTV

Stereo 3D HDTVSharing/conferencing/collaboration


upstreamdownstream

Stereo 3D SDTVHD Video calling

Stereo 3D HDTV

7680 x 43204096 x 2160

1920x 10801920x 1080

HDTVStereo 3D

HDTVSuper HDTV

1920x 1080

Multiview 3D HDTV

Ultra HDTV

HDTV



Worldwide Internet audienceStatus: March 2011

North America:272.1 million (13.0 %)

Europe:476.2 million (22.7 %)


Latin America:215.0 million (10.3 %)

Middle East & Africa:187.2 million (9.0 %)

Asia Pacific:944.3 million (45.0 %)

Source: Internet statistics: http://www.internetworldstats.com/ .

More than 2 billion Internet users world


Internet traffic growth is huge

40

60

80

100

Global FIXED traffic (ExaByte/month)

ResidentialUnicast TV

1.0

1.5

2.0

2.5

Global MOBILE traffic (ExaByte/month)

ResidentialInternet

BusinessInternet

Voice

Handhelddata traffic

LaptopData Traffic

Voice


Source: Analyst reports and Nokia Siemens Networks

0

20

2007 2008 2009 2010 2011 2012 2013 2014 20150.0

0.5

2007 2008 2009 2010 2011 2012 2013 2014 2015

InternetVoiceTraffic

VoiceTraffic

Fixed broadband trafficis 40x mobile in 2015

Mobile data trafficgrows 300 fold


Drivers of the Future Internet

Applications and Content changed

Personal TV

Community Social WebVoIPNavigationGamingIPTV

Future Internet WLANWiMAXHSPA/LTE



Future Internet faces challenges beyond the original design

Mobile and multi-access Devices

WLANWiMAXHSPA/LTE

Mobile broadband driving need for better network quality & efficiency


Source: Analyst reports and Nokia Siemens Networks


Identified frequency spectrum


Potentially, additional allocation in CEPT countries under discussion

Potentially, additional allocation in CEPT countries under discussion

• WRC 2007 identified additional frequency bands for mobile and wireless communications

• Preparation for WRC 2012 ongoing to get agenda item for WRC 2016• WRC 2016 may discuss and identify additional frequency spectrum


Source: ITU-R

Performance evolution by ‘simply’ shrinkingtransistor size becomes more difficult

1µm

10µm

100µmTechnology evolution (half-pitch)

0.7x 100 Mill.

1 Bill.

10 Bill.

100 Bill.

Transistors per chip (Moore‘s law)

2.0x every 2 years


1nm

10nm

100nm

1980 1990 2000 2010 20201970

0.7x every 2 years

10 000

100 000

1 Mill.

10 Mill.

1980 1990 2000 2010 20201970

2 years

Physical limits of atomic structure for scaling transistors while still managing power and thermals drive new approaches

Source: AMD, IBM, Intel processor roadmaps and datasheets


Strong limitations in clock speed & powerImproving performance trough parallelism

Transistors [000]

µP evolution from single-core to multi-core

100000

1000000

10000000

100000000


Cores are the new transistors

Number of cores per chip doubles every two years ““# Cores

Power[W]

Clock[MHz]Transistors [000]

1

10

100

1000

10000

1970 1980 1990 2000 2010 2020

Source: AMD, IBM, Intel processor roadmaps and datasheets


Basic technology evolution - the foundation

Do thelaws hold?

AdvancedIO technologies

UnlimitedProcessing

IT everywhereUnlimitedStorage


SecureNetworks & services

AdvancedIO technologies

Privacyby design



Unlimited processing

Paradigm shift from sequential to parallel processing will enable unlimited processing capacities

Parallelism will have changed SW design principles and application environments

Processing resources get centralized in the cloud –with geographic redundancy


CPU Server Cluster Grid

Virtualization

Core

App App …



Radio networks will be pushed to Gigabits on the move

Spectral Efficiency byextreme beam formingand coordinated transmission

Heterogeneous networksSmart integration of local area technologies for traffic growth and high data rates will provide ubiquitous access


All access standards with programmability. Very compact and widebandfuture radios

Three times licensed bandwidth for mobile communications by 2020



Miniaturization continues and enables abundant computing & storage capacities

32 nm half-pitch 7 nm half-pitch

• 256 cores / CPU

2010 2020Moore’s law

Processing • 8 cores / CPU


Source: IDC digital universe study


• 256 cores / CPU• Human brain

• 100 $ / TB• Located in

clients

• 1$ / TB• 30 % in the

cloud*

Processing

Storage

• 8 cores / CPU• 1000$ PC

computeequivalent: Spider

Innovations ahead in photonics and IOSustainability remains a challenge

• Optics into line cards

• 10 Tbps / fiber

• Optics into CPUs• 60 Tbps / fiber

• HD planar 3D • UHD volumetric

2010 2020Photonics

Advanced IO


• HD planar 3D• 2k pixel/$• Touch input

• UHD volumetric 3D

• 64k pixel/$• Natural language

• 0,8 Gt CO2*• 100 Gbps/kW

• 1.5 Gt CO2*• 600 Gbps/kW

Advanced IO

Sustainability

Source: Global ICT footprint, Smart 2020 report


Traffic growth and changing usage define requirementsDiversity of applications demands flexibility

Majority of all traffic will be streaming of media

Real-time cloud services requirelow latency

Communication technologies

are absorbed in all areas of our lives


Sharing drives upstream requirements

Explosion of communication endpointswith billions of connected objects

But we have to expect the unexpected…



What will the world want from wireless beyond 2020?

Support up to 1000 times more traffic

Rock solid, ubiquitous connectivity

Mobile data traffic forecast

2010 20151 EB

1000x1000 EB

2020

©2011 Intuitive Surgical, Inc.


Gbps peak speeds

Millisecond latency for true “local feel”

Super HDHD

Ultra HD3D HD

3D SDTV

100Mbps 200Mbps 3000Mbps

Apps bandwidth demand

©2011 Intuitive Surgical, Inc.



By 2020 – radio can reduce latency 10x

The main benefit is realized when

Radio latency can be pushed to 1 ms in 2020 by using a shorter frame length

15 ms

20 ms

Latency in radio networks


realized when content is close. 1 ms limits fiber length to 100 km.

Enable low-latency M2M solutions0 ms

5 ms

10 ms

HSPA LTE “B4G“

Air interfaceBase station UE



By 2020 - radio can improve in spectral efficiency 10x

Efficiency is not limited by

Spectral efficiency can be improved by managing inter-cell interference.

6

8

Spectral efficiency [bps/Hz/cell]


limited by Shannon but by inter-cell interference.

Cell edge data rates improve twice as much

0

2

4

LTE2x2

LTEAdvanced

4x4

+ CoMP + UEinterferencecancelation

+ furtherinnovations

HSPA today



By 2020 - there can be 10x more spectrum available

10 times more spectrum can be

Evolution of available radio spectrumUnlic 5GHzUnlic 2.4GHz3700 MHz3400 MHz2600 MHz

1500MHz


spectrum can be made available if we drive for it.

Cognitive radio enables optimized spectrum usage over multiple operators

2600 MHz2300 MHz

2100 MHz1800 MHz900 MHz800 MHz700 MHz450 MHz

2010 2012 2015 2020

500MHz

1000MHz

2100 MHz TDD



By 2020 - there will be 10x more base stations

80 % will be microcell or

Number of cellular base stations grows to over 50 Million

Global base station forecast

30 Mn

40 Mn

50 Mn


microcell or smaller

Additionally more than 500 Million WiFi APs2010 2020

0 Mn

10 Mn

20 Mn

30 Mn

2015 Micro, Pico, Femto BTS Macro BTS



Bringing all together allows up to 1000 times more capacity

10 xPerformance

10 x Spectrum

10 x Base stations

1000xcapacity


capacity



How can we build this?

Radio frequencyBandwidth capability of base stations

2MHz

200MHz

60MHz

20MHzContinuous growth of

System on Chip enables small radios, low power consumption & integration of intelligent functions. SDR is no problem for digital processing!

Laws of physics determine antenna size


Switched mode power amplifier, high-voltage GaAs HBT and GaN technologies for wideband radios with multicarrier capabilities

2MHz

1990 2000 2010 2020

Continuous growth of computing power with Moore‘s law

2Mn

60Bn

60Mn

1990 2000 2010 2020

2Bn

For same performance, antenna size does not get smaller. Size even increases if beamforming is required

1990 2000 2010 2020



How can we deploy, commission and maintain all the radios, frequency bands and layers?

All cells and frequency layers automatically managed by advanced SON

All spectrum underunified RRM for instant

Virtually one ubiquitous connectivity



All spectrum underunified RRM for instant capacity and coverage optimization

Cognitive networks will reduce errors, improve quality and lower operation and energy costs

Heterogeneous access networks will enableubiquitous broadband coverage

Gigabits on the move with “beyond 4G”

Licensed & unlicensed spectrum


FTTx for homes, enterprises & backhaul

Large & small cells

FTTx DSL



Gigabit ExperienceReliable, Efficient and Personal

Unified RadioResourceManagement

CognitiveNetworks

UbiquitousConnectivity


Liquid RadioWidebandMultiradio

Multi-Standard

HeterogenousNetworks

Self Organized Networks

Multi-Carrier

Active Antennas



M2M / Internet of Things / Sensor based systems

Billionsof devices Evolution of Internet

2010 …..

Everythingconnected

• Mobility, pervasiveness, scalability• Data deluge (Web 2.0, P2P, M2M, 3D, …)• Heterogeneity of devices (e.g. RFIDs, sensors)

Reports

Example:Fleet Monitoring – Functional Architecture


Source: EU Commission / Nokia Siemens Networks

2010 2014 …..EU Commission Send notification in response

to pre-defined triggers

Control Room Operator

GPS location

Transport Units

Containers with RFID

LocationNotification

Container Sensors

Vehicle Sensors

GIS Spatial Database

Data Centre

Location Map

Console Unit

Nokia Siemens Networks

• Such systems essential for Future Internet applications and solutions for societal challenges such as− Smart energy− Smart traffic− eHealth− etc.

Estimate for global annual CO2 emissions of mobile network elements

Secondary energy

Mobile CS Core

Mobile PS Core

Transport

IP Core

10 kW1 kW0.1 W


Secondary energyusage of deliveryand supply chainUsers

5 billion0.33 GW

3.5 million x 1.5 kW5.25 GW

12000 x 10 kW0.12 GW

1.7 MtCO2/year

23.3 MtCO2/year

Secondary energyusage for construction

0.6 MtCO2/year

Mobile PS Core IP Core



Physical layer – Shannon law• Associated with traffic growth energy consumption in communication networks is

expected to increase• Energy consumption results in significant economic cost factor for network operation• Today’s mobile communication systems designed under paradigm

− of high spectral efficiency and peak throughput− with respect to the limited available frequency spectrum

• Physical layer is approaching the Shannon limit• Future ICT networks and systems to be designed

− to be more energy efficient and−


− to be more energy efficient and− with reduced operational cost



0

2

4

6

8

10

12

14

-20 -15 -10 -5 0 5 10 15 20 25 30 35 40

T/B

S/(N ΔCIR) [dB]

0

2

4

6

8

10

12

14

-20 -15 -10 -5 0 5 10 15 20 25 30 35 40

T/B

S/(N ΔCIR) [dB]

Operating rangeof today‘s systems with adaptive modulation

S/(BN0F) [dB]

Physical layer – Shannon lawS2/S1 [dB]

-15,0

-10,0

-5,0

0,0

5,0

5 10 15 20 25 3530 40 45 50-20 dB-10 dB0 dB10 dB

FNBS

⋅⋅ 01

1 [dB]Parameter


−

⋅⋅

+⋅⋅⋅

⋅= 112

1

01

1

1

01

1

2

1

2B

B

FNBS

SFNB

BB

SS

B2/B1

-25,0

-20,0 20 dB30 dB



• Conflicting requirements between− high spectral efficiency and− low energy consumption

Overall primary power of network• Signal processing and

transport taken into account in addition to RF power amplifier

Entire signalProcessing

exceptRF amplifier

RF amplifierPt

Pprim, RFPprim, ex RF

Pprim

η

η⋅= RFprimt PP ,

ηt

exRFprimRFprimexRFprimprimPPPPP +=+= ,,,

ηα t

exRFprimPP ⋅=,



Same peak throughput, cellular system

-8,00-6,00-4,00-2,000,002,004,00

n = 2n = 2.5n = 3n = 3.5n = 4

α = 0.01

0 2 4 6 8 10 12 14 16 18 20 Δ [dB]Σprim, total [dB]

Propagationcoefficient

Same area capacity, cellular system

-20,0000

-15,0000

-10,0000

-5,0000

0,00001 4 7 10 13 16 19 22 25 28 31 Serie

s1Series2Series3Series4Series5

Δ [dB]5 10 15 3020 250Σprim, total [dB]

-4.50

-9.00

-13.50

-18.00

0.00

α = 0.01

n = 2n = 2.5n = 3n = 3.5n = 4

Multi-hop systems

• Single-hop system (LOS)

• Multi-hop system (LOS)

• Total transmit power savings for same antenna gains and different path loss exponents

[ ]dB =Σ

[dB]



[ ]

[ ]dBNN

dB

n

hopmultitotalprim

+

+⋅

⋅=

=Σ −

1

1

log10

,,

α

α

-14-12-10

-8-6-4-20

1 2 3 4 5 6 7 8 9 10

N – Number of hops

Propagationcoefficient

n = 2n = 2.5n = 3n = 3.5n = 4

α = 0.01

Σpr

im, t

otal

mul

ti-ho

p[d

B]

• New challenging requirements for improved user experience− to increase cell edge throughput and− average cell capacity

New directions I

Interference avoidance through high reuse factors

Interference shaping and exploitation through distributed

MIMO and relaying

Interference suppression through classical MIMO


BMBF EASY-C project


Source: Nokia Siemens Networks / BMBF EASY-C project

• More balanced throughput across deployment area requires− interference mitigation techniques and− new deployment concepts combined with complex signal processing (CoMP –

Cooperative Multipoint)

• New concepts like cognitive radio getting more attention due to− limited available identified frequency spectrum and− a more flexible regulatory approach for the allocation of frequency bands

• SON – Self-Organising Networks to manage complexity• Future Internet applications increasingly require

− short-range communication systems− M2M / Internet of Things / Sensor based systems

• Increasing data traffic as well as concerns on climate change and CO2

New directions II


• Increasing data traffic as well as concerns on climate change and CO2emissions result in requirements to improve energy-efficiency of ICT systems

• Societal challenges on the agenda to investigate, how ICT systems can support• more efficient use of energy and resources• in other sectors of the economy and society

• ICT research subjects will address more interdisciplinary topics



Conclusions

• Traffic in mobile and wireless communication systems is significantly increasing• Traffic growth in relation to flat rate billing models result in economic challenges• Technology developments in chip technology and architectures supporting

increased system complexity• Future systems for ubiquitous connectivity will provide

− increased throughput,− short-range communications,− M2M / Internet of Things / Sensor based systems and


− M2M / Internet of Things / Sensor based systems and− heterogeneous networks

for the Future Internet• Future systems based on optimized integration of major building blocks

− multicarrier radio systems, multi-standard networks, heterogeneous and self-organized networks

− active antennas, wideband radio systems and liquid radio− unified radio resource management and cognitive networks

• Energy efficiency is a new major concern in research and development



Acknowledgement

The presenter would like to acknowledge the contributions from his colleagues at Nokia Siemens Networks to this presentation.

