March 30th 2019

Spectrum and its application to 5G and the Power of Massive MIMO

Tajit Mehta

2

The jump from LTE to 5G will be like…

3

4

5

Existing networks & technology must evolve to handle the demands of the connected future

Between 2018 and 2022, the traffic generated by

smartphones will increase by 10 times

6

5G IS TRULY REVOLUTIONARY …& will satisfy customer demands 10X

5G will handle 10X connection density 10X

5G will deliver 10X experienced throughput

10XLower latency

100XIncreased area traffic

capacity

2G 3G 4G 5GSource: ITU, IMT-2020

7

5G3 Million

new US jobs(800K in construction)

$275 Billioninvestment

$500 Billioneconomic

growth

UNTOLD POTENTIAL

for innovation

Even more important – 5G can deliver a MASSIVE economic benefit

Source: CTIA

8

EM Intro video

ElectromagnegticSprectrum

Current Wave Magnetic Wave 7 Types of wave

Electric field to give a direction To give the field Radio wave

Microwave

Infrared

Visible

Ultraviolet

X-ray

Gamma

9

10

Spectrum Waves and Output

11

The Sweet Spot

Capacity Demands and Deployment

Coverage shown is for illustrative purposes only

Capacity need

LTE/LTE-Advanced

LTE-Advanced Pro/eLTE5G sub-6 GHz/mmWave Small cell

5G sub-6 GHz Macro

Map data: Google

13

Sample Intra-band breakouts

FCC Spectrum Allocation of Bands of Interest

750 - 950 MHz

2300 - 2700 MHz

Sam Wetsel

Nextel Spectrum Resources

2496

Current

Allocation

On Top

1900 - 2170 MHz

2320

2345

2360

2305

WCS

(Down)

WCS

(Up)

DARS

Aeronautical

Telemetry

Amateur

Radio

Cellular

A & B

(Uplink)

900 SMR (Up)

Amateur

Radio /

Radiolocation

Public

Safety

2 MHz

Guard

Band

5 MHz 3G

10 MHz 3G

Fixed

Microwave

800 SMR/

NPSPAC

(Uplink)

806

824

1 MHz

Guard

Band

782

776

777

Public

Safety

762

764

5 MHz 3G

10 MHz 3G

1 MHz

Guard

Band

752

747

792

794

849

851

Commercial

Aviation Air

Ground Systems

Cellular

A & B

(Downlink)

800 SMR/

NPSPAC

(Downlink)

869

894

902

901

896

929

928

Paging

Fixed Microwave

900 SMR (Down)

941

940

932

935

MAS Fixed

Microwave

Narrowband

PCS

2500

2506

2512

2518

2524

2530

2536

2542

2548

2554

2560

2566

ITFS D3

2572

2578

2584

2590

2596

2602

2608

2614

2620

2626

2632

2638

2644

2650

2656

2662

2668

2674

2680

2686

2690

Mobile – Satellite

(Downlink) /

Radiodetermination

Satellite

Big LEO paired

w/ 1610-1626.5

(Iridium &

Globalstar)

Lower Guard Band

ITFS C4

ITFS D4

ITFS C3

ITFS D1

ITFS C2

ITFS D2

ITFS C1

ITFS B3

ITFS A4

ITFS B4

ITFS A3

ITFS B1

ITFS A2

ITFS B2

ITFS A1

MMDS H3

ITFS G4

R Channel

ITFS G3

ITFS G2

ITFS G1

MMDS H2

MMDS H1

MMDS E1

MMDS F1

MMDS E2

MMDS F2

MMDS E3

MMDS F3

MMDS E4

MMDS F4

ITFS B4

ITFS C4

ITFS D4

ITFS A4

ITFS G4

ITFS F4

ITFS E4

K Guard

J Guard

MDS 2

BRS E1

BRS E2

BRS E3

BRS F1

BRS F2

BRS F3

BRS H1

BRS H2

BRS H3

ITFS G1

ITFS G2

ITFS G3

ITFS D1

ITFS D2

ITFS D3

ITFS C1

ITFS C2

ITFS C3

ITFS B1

ITFS B2

ITFS B3

ITFS A1

ITFS A2

ITFS A3

MDS 1

2495

2502

2507.5

2513

2518.5

2524

2529.5

2535

2540.5

2546

2551.5

2557

2562.5

2568

2572

2578

2584

2590

2596

2602

2608

2614

2618

2624

2629.5

2635

2640.5

2646

2651.5

2657

2662.5

2668

2673.5

2679

2684.5

2690

Fixed - Satellite /

Radio Astronomy /

Space Research

Re-Allocation

On Bottom

PCS

Block C

(Uplink)

PCS

Unlicensed

2150

Currently

MDS Ch. 1

1895

Async

(data)

Iso

(voice)

1910

1930

PCS Block D (Down)

PCS

Block B

(Down)

PCS

Block C

(Down)

PCS

Block A

(Down)

1945

PCS Block E (Down)

PCS Block F (Down)

1950

1975

1970

1965

1990

2110

2000

2025

2020

2162

2156

Reallocated To:

NTIA 3G Advanced Wireless

(Downlink)

PCS Block

G (Up)

PCS Block

H (Up)

PCS Block

G (Down)

PCS Block

H (Down)

(ICO / Iridium/

Celsat / Boeing)

MSS (Uplink)

NXTL White

Paper (Up)

Currently TV Aux Broadcast (BAS)/Cable TV

Relay/Local TV Transmission Channels 3-7

Fed: Space Operation/Earth Exploration Satellite/Space Research

Currently BAS

Channels 1 & 2

Reallocated TV Aux Broadcast (BAS)/Cable TV

Relay/Local TV Transmission Channels 1-7

Fed: Space Operation/Earth Exploration Satellite/Space Research

Currently

MDS Ch. 2

2155

To Be

Reallocated

Currently Public

Microwave

Currently

Public

Microwave

1995

2000

1915

1920

14

The Three Ps & a Bonus C

PowerPropagationPenetrationCapacity

Dense urbanMM wave

MetroMid-band

15

NationwideLow band

New T-Mobile

Multi band use requirements met by band not Single band to multi use

Multi Band Spectrum use

16

MU Mimo video

17

MU-MIMO and Beamforming

LTE

LTE+5G

LTE32T32R 5G NR

32T32R

Split Mode Massive MIMO Beamforming Multi-User MIMO Pairing

Vertically paired users

Horizontally paired users

The image part with relationship ID rId2 was not found in the file.

Massive MIMO Highlights

Higher Spectral Efficiency

5GKey building block of

Network

Better Coverage Better Capacity

Horizontal and Vertical Beamforming

19

Capacity 3x20 MHz LTE + 60 MHz 5G NR

Single Sector (Gbps) 3

3-Sector Site (Gbps) 9

8x Peak Sector Throughput

Avg. Sector Throughput

6x

3x Cell EdgeThroughput

Expected Capacity with LTE + 5G NR

7256 QAM, 8 MU-MIMO layers and 32T32R each for LTE and 5G NR

Massive MIMO PerformanceExpected 32T32R - Split Mode Performance over 8T8R

Air Interface EnhancementsCapacity - Massive MIMO Benefits

Multi-User MIMO increases spectral efficiency

8x NetworkCapacity

Ave. Sector Throughput

6x

4x Cell EdgeThroughput

4dB UL Coverage Gain

Expected 64T64R 16 Layer Performance over 8T8RMassive MIMO MacroLegacy Macro

Massive MIMO for 5G below 6 GHz, Emil Björnson, Linköping University

Horizontal and Vertical Beamforming

Spec

tral

Effi

cien

cy [b

it/s/

Hz]

1000

500

0

1000

500

0

8

6

4

2

0Position [m] Position [m]

Spec

tral

Effi

cien

cy [b

it/s/

Hz]

1000

500

0

1000

500

0

8

6

4

2

0Position [m] Position [m]

912

24

33

0

5

10

15

20

25

30

35

3x20 MHz LTE + 60 MHz 5G NR 3x20 MHz LTE + 100 MHz 5G NR 3x20 MHz LTE + 100 MHz 5G NR 180 MHz 5G NR

3-Se

ctor

Site

Thr

ough

put (

Gbps

)

LTE/NR Site Configuration

Near Term

Mid Term

Long Term

32T32R 32T32R

64T64R

64T64R

SPLIT MODE

8

Peak Site Capacity

22

• Interference• Cross purpose absorption • Ducting• Seasonal• UFO

• Edge of Network handoff• Network densification• Market expectation vs market opportunity

Challenges

23

Opportunities

24

https://www.pbs.org/wgbh/nova/article/thanks-nanoparticle-injections-these-mice-can-see-infrared-are-we-next/

https://www.pbs.org/wgbh/nova/article/thanks-nanoparticle-injections-these-mice-can-see-infrared-are-we-next/

25

Mouse Video

• https://www.eurekalert.org/pub_releases/2019-02/cp-nmi022019.php

https://www.eurekalert.org/pub_releases/2019-02/cp-nmi022019.php

26

Appendix

Massive MIMO Antenna Arrays

- Conventional 8T8R Antenna:

- Each column uses 8-12 antenna elements to create a vertically fixed directional pattern.

- The 8 columns, each used with a single transceiver for signal transmission, are designed for horizontal beamforming only.

- 64T64R Massive MIMO Radio/Antenna:

- Has 64 transceiver units, each mapped to 2 antenna elements.- All 128 antenna elements create the antenna pattern, with array

design enabling both vertical and horizontal beamforming.- The addition of the vertical dimension, with per antenna element

adjustments, turns a column of antennas into an antenna array allowing many more layers and much finer adjustments.

- TDD is better than FDD due to channel reciprocity; with FDD, feedback overhead increases with antenna elements.

The image part with relationship ID rId2 was not found in the file.

Benefits of TDD in Massive MIMO

• TDD operates on the same frequency on DL and UL⇒ DL/UL channel reciprocity: The channel estimate of the UL at the Tx can be utilized for DL beamforming, thus less overhead

• FDD operates on different frequencies on DL and UL⇒ No DL/UL channel reciprocity: Two way pilots and feedback needed

Much higher overhead, which greatly limits the total number of allowable antennas in case of high mobility scenario when the coherence block τ is small.

• To ensure reliable channel feedback, the reference signal symbols should be sent within a coherence block (𝜏𝜏):

𝜏𝜏 = Channel coherent bandwidth x Channel coherent Time

• TDD requires 2𝐾𝐾 UL sounding signal symbols for channel estimation𝐾𝐾 < 𝜏𝜏/2

• FDD requires 2𝑀𝑀DL reference signal symbols and 2𝐾𝐾 UL reference signal symbols for channel estimation and feedback

𝑀𝑀 + 𝐾𝐾 < 𝜏𝜏/2

𝑀𝑀 is the total number of antennas at the eNB, 𝐾𝐾 is the total number of served UEs.

Source: Massive MIMO: Ten Myths and One Critical Question, Emil Bjornson, Erik G. Larsson, and Thomas L. Marzetta

FDD

TDD

(100,25)for τ= 200

Number of terminals (K)

τ2

τ2

τ2

τ2

Number of antennas (M)

Number of antennas (M)0%

10%

20%

30%

40%

50%

Overhead

29

Dallas

Chicago

Los Angeles

New York City

Washington, D.C.

Atlanta

Phoenix

Houston

Kansas City

NOW FOCUSED ON 9 MAJOR MARKETS

Air Interface EnhancementsCapacity - Massive MIMO and Beamforming

Massive MIMO for 5G below 6 GHz, Emil Björnson, Linköping University

UserSignal goes in alldirections

User

Substantial side-lobes Main

lobe

User

Narrow main lobeTiny side-lobes

Fully Digital Beamforming

Hybrid Beamforming

highly focused beams

Single User MIMO

co-scheduled on same time/frequency resourceMulti-User MIMO

Massive MIMO systems enable sharper beamforming and null-forming

>

Adapted from Ericsson

M- # of Tx

Small Cell Deployment ChallengesBackhaulSiting/Real Estate/Zoning & PermittingSheer # of sites needed

These remain as open challengesfor the Industry today for LTE small cells. 5G will onlyaccentuate the problem.

5G will NOT be successful if we cannot adequately address roadblocks to large-scale small cell deployments that exist today.

Large variability in local laws, landlord policies, and other site-specific factors result in cost distributions that limit small cell deployments today to the 1,000’s to 10,000’s, and will greatly inhibit the adoption of high-frequency 5G in the US, where 100,000+ units may be necessary per operator (depending on frequencies used, and other company-specific business and technical design criteria).

Large-scale small cell site deployment capital cost model provided courtesy of Nokia.

Slide Number 1Slide Number 2Slide Number 3Slide Number 4Slide Number 5Slide Number 6Slide Number 7EM Intro videoSlide Number 9Spectrum Waves and OutputThe Sweet Spot Capacity Demands and Deployment Sample Intra-band breakoutsThe Three Ps & a Bonus CSlide Number 15MU Mimo videoMU-MIMO and BeamformingMassive MIMO HighlightsSlide Number 19Air Interface Enhancements�Capacity - Massive MIMO BenefitsPeak Site CapacityChallengesOpportunities��Mouse VideoAppendixMassive MIMO Antenna ArraysBenefits of TDD in Massive MIMOSlide Number 29Air Interface Enhancements�Capacity - Massive MIMO and BeamformingSmall Cell Deployment Challenges