LTE-U Technology and Coexistence

sherifhanna

vVjZqx

Qualcomm Technologies, Inc.

LTE-U Technology

and Coexistence

May 28 th , 2015

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

1


Disclaimer and Copyright Notification

Copyright © 2015 LTE-U Forum Companies: Alcatel-Lucent, Ericsson, LG Electronics, Qualcomm Technologies Inc., Samsung

Electronics, Verizon. All rights reserved.

This document provides initial information related to Long Term Evolution (LTE) technology operation using unlicensed

spectrum as a Supplemental Downlink to LTE technology operation using licensed spectrum. All information provided herein is

subject to change without notice. The information provided herein was considered technically accurate by the LTE-U Forum

Companies at the time the documents were initially published, but LTE-U Forum Companies disclaim and make no guaranty or

warranty, express or implied, as to the accuracy or completeness of any information contained or referenced herein. LTE-U

FORUM COMPANIES DISCLAIM ANY IMPLIED WARRANTY OF MERCHANTABILITY, NON-INFRINGEMENT, OR FITNESS FOR

ANY PARTICULAR PURPOSE, AND ALL SUCH INFORMATION IS PROVIDED ON AN “AS-IS” BASIS. Other than as explicitly

provided below, LTE-U Forum is not providing any licenses under any intellectual property of any kind owned by any person

(whether an LTE-U Forum Company or otherwise) that may be necessary to access or utilize any of the information contained

herein, including but not limited to any source materials referenced herein, and any patents required to implement or develop

any technology described herein. It shall be the responsibility of the developer to obtain any such licenses, if necessary.

LTE-U Forum disclaims liability for any damages or losses of any nature whatsoever whether direct, indirect, special or

consequential resulting from the use of or reliance on any information contained or referenced herein.

Permission is given to reproduce this document in its entirety solely for the review, analysis, and implementation of LTE-U. Use

of this document for any other reason without the permission of LTE-U Forum Companies is explicitly prohibited.

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

2


Outline

• Overview of LTE-U coexistence algorithms design

− Channel selection, CSAT and OSDL

− How to coexist with Wi-Fi BSSs within and outside CCA-ED

• Lab & OTA results for coexistence above and below CCA-ED

− Different traffic applications

− Wi-Fi/Wi-Fi, Wi-Fi/LTE-U fairness study

• Demos

− Stress room demo with three different Wi-Fi AP vendors

− OTA on-campus demo with two different Wi-Fi AP vendors

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

3


Introduction

• Throughout the LTE-U design and testing, coexistence has been a main principle that has

been studied thoroughly with different applications, devices, and deployment models

• We couldn’t rely just on 802.11 spec due to divergence by many implementations

− Wi-Fi benchmark based on testing of Wi-Fi AP and chip vendors with the highest market shares

• What is presented here is Qualcomm’s design in accordance with LTE-U Forum spec

− The spec provides guidelines and testing scenarios to be passed but doesn't dictate specific

implementations

− This is obviously not the only way for implementation

• The design covers a larger set of scenarios than those mandated by spec

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

4


LTE-U Coexistence Solutions State Diagram

Initialization

& Configuration

Triggered by channel switch

condition

Channel Selection

(ChS)

CSAT Enabled

LTE-U UE added

(or high traffic load)

OSDL

(LTE-U Off)

No LTE-U UE

(or low traffic load)

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

5


Channel Selection

Wi-Fi AP

UE3

Wi-Fi AP

STA

UE2

Co-located LTE-U

SC/Wi-Fi AP

Wi-Fi AP

UE1

LTE-U and WiFi NL(Network Listening) assisted with UE reporting

LTE & Wi-Fi measurements and CQI/BLER to detect hidden nodes

LTE-U SC

LTE-U SC

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

6


Channel Penalty Function

Wi-Fi AP Beacon RSSI

Measurements, and whether

channel is primary or secondary

LTE-U UE WiFi

Measurements over RRC

δ 1

×

δ 2

×

δ 3

LTE-U NL

Measurements

LTE-U UE RRM

measurements

LTE-U UE CQI/

BLER

×

+

δ 4

×

δ 5

×

Total Channel Penalty

• Different penalty weighting for different interference sources, e.g. WiFi primary channel has higher penalty

compared to secondary channel

• Quantized penalty for robustness to measurement errors and to reduce bias due to outliers

• Measures % of coverage area with some desense due to other AP interference

• Built in mechanisms to detect and mitigate frequent channel switching

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

7


Channel Selection- Summary

1. Has to be both during initial boot up as well as dynamically during operation

2. Choose the cleanest channel in general

3. If possible avoid primary channels of WiFi

4. If possible avoid channels occupied by other LTE-U operators and choose channel occupied

by the same LTE-U operator (Forum coexistence spec 6.1.2)

• Channel selection is aided by Small Cell NL(Network Listening) and/or UE measurements

• Can be done through passive or active scan- our design uses active

• Scanning typically happens every few seconds

• Channel selection can happen at any time, but typically within 10s seconds

• Channel selection is conservative in that it considers any found beacon to be active

− Taking channel loading in consideration is for further optimization

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

8


Carrier Sense Adaptive Transmission (CSAT)

• A TDM transmission pattern on the LTE-U SCell with T CSAT cycle



This pattern leverages existing LTE carrier aggregation MAC signaling ( MAC control element for activating

and deactivating an SCell)

This pattern is triggered based on transmissions measured from Wi-Fi APs

• Wi-Fi medium utilization measurement is performed during LTE-U SCell OFF time, to adjust

ON/OFF duty cycle

− Adaptive trade-off between LTE-U and other systems (e.g. Wi-Fi) performance.

LTE-U ON

Gating Cycle (T CSAT )

LTE-U OFF

T on /T off

adaptive

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

9


Wi-Fi Medium Utilization (MU) Estimation

Input to CSAT loop

• Wi-Fi MU monitoring is done where all LTE-U small cells are required to turn 5GHz RF

transmission OFF in this period

− guarantees overlapping OFF time across neighbor LTE-U small cells for sensing Wi-Fi

• Sensing time is selected sufficiently large to allow the co-located Wi-Fi NL module for

reliable MU monitoring of the operating channel. MU is also averaged across multiple

observations

• Wi-Fi NL decodes the preamble of any WiFi packet detected during this time and records its

corresponding received signal strength indicator (RSSI), duration in μs (or NAV), modulation,

coding scheme (MCS) and source/destination address

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

10


Wi-Fi MU Estimation

Averaged weighted observed activity over a monitoring window

• The algorithm divides packets into different categories based on their attributes and each category is assigned

a weight

• If the packet is found to be an ACK, or CTS, and the preceding data packet is not detected, the packet is

classified as generated from a hidden node and count the packet as full duration (nominal value of around

4msec)

• Let W i and D i denote the weight and duration of i-th detected packet during MU monitoring time, assume that

K packets were detected during the monitoring time. The instantaneous MU is computed as

MMMM(nn) =

1

KK

MMMMMMMMMMMMMMMMMMMMMMMMMMMM ii=1

WW ii × DD ii

• WiFi medium utilization is filtered as follows:

MMMM nn = αα MMMM MMMM nn + 1 − αα MMMM MMMM nn − 1

where filter coefficient αα MMMM could be set small for smaller monitoring time to allow more averaging

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

11


CSAT Adaptation

Outer loop to decide on Ton in a given CSAT cycle

• CSAT algorithm will increase/decrease TT OOOO if MU is below/above a certain threshold

• CSAT adaptation is done as follows:

TT OOOO nn + 1 = miiii TT OOOO nn + ∆TT UUUU , TT OOOO,mmmmmm iiii MMMM nn < MMMM_TTTTT 1

TT OOOO nn + 1 = TT OOOO nn iiii MMMM_TTTTT 1 ≤ MMMM(nn) ≤ MMMM_TTTTT 2

TT OOOO nn + 1 = maaaa TT OOOO nn − ∆TT DDDDDDDD , TT OOOO,mmmmmm iiii MMMM nn > MMMM_TTTTT 2

where

− TT OOOO,mmmmmm bounds LTE-U ON interval to ensure LTE-U receives a fair share of the medium



TT OOOO,mmmmmm bounds LTE-U OFF interval to allow time to sense Wi-Fi activity

∆TT UUUU and ∆TT DDDDDDDD determine the convergence speed of CSAT adaptation

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

12


Adjust T ON,min for Fairness

Selecting T ON,min independent of the number of neighbor Wi-Fi APs and LTE-U SCs might result

in unfair resource sharing. To mitigate this issue, T ON,min is selected as below after every AP

scan:

TT OOOO,mmmmmm = mmmmmm TTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTT,

NN + 1 × TT CCCCCCCC

NN + 1 + MM + NNNNNNNNNNNNNNNNNNNNNNNN

where





TTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTT is a configurable parameter controlling minimum duty cycle below ED

N is the number of LTE-U SCs with the same PLMN ID measured by the specific LTE-U SC

M is the number of LTE-U SCs with different PLMN ID

Number of Wi-Fi Devices (how to weigh in STA activity (visible to the algorithm) is still being optimized)

− Always go 1/num_of_Tx vs 1/num_APs vs a mix

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

13


Protecting Latency Sensitive WiFi Applications - Subframe Puncturing

Gaps inside Ton

• Subframe puncturing introduces frequent gaps in the LTE-U ON duration to help Wi-Fi flush

delay-sensitive data and management packets that may be queued at the AP because of

LTE-U transmissions

• In the specified punctured pattern, the subframes which are punctured are not used for

scheduling any data

• Typically few msec every few 10s of msec (2/20msec is typical)

eNB Tx

Power

T ON 80ms

Few ms SF puncturing every

20ms

e

• Another way to accommodate latency sensitive applications is to run short CSAT cycles


Example: Punctured CSAT Waveform

Recent design allows LTE-U to go to as low as 20msec of Ton

T OFF

80ms

20ms

Tim

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

14


CSAT Cycles

• Typical CSAT cycles

− 80msec total ON/OFF duration (i.e. like 40/40) & 160msec (i.e. like 80/80msec)

• CSAT cycles on primary WiFi channel are typically shorter to enable beacon transmissions

and connection setup activities

• In general CSAT cycles are constant but can change in scenarios where LTE-U share is too

small

− i.e. 10% LTE-U share may be 40msec/360msec as opposed to 10msec out of 100msec

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

15


Benefit of WiFi NL

Not fundamental

• Most coexistence functionalities can be achieved without NL albeit

more complexity and less optimal trade-offs

• Examples:

− NL can help though reacting to some less common scenarios like hidden

nodes

− NL may allow better tradeoff of coexistence since it allows distinction

between one node vs multiple nodes when both scenarios show same

medium activity

− NL allows distinction between LTE-U and WiFi nodes allowing the ability to

run reuse 1 if need be on the former

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

16


Lab and OTA Testing

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

17


LTE-U / Wi-Fi Coexistence Setup

• Extensive lab test on LTE-U coex

− Wi-Fi: Multiple AP and STA brands

LTE-U: Test eNB and UE MTP

CSAT Wi-Fi Testing

Logical Setup Diagram

• Current focus: Non-LBT/CSAT design





LBT-based test depend on 3GPP progress

Key Non-LBT coex feature is CSAT

CSAT has dynamic duty cycle, max 100-ms ON time, and long + short

OFF gaps

Note LTE-U coexistence in reality would be better with proper channel

selection

TBS eNB

PL eNB,UE

PCell

SCell

FC

FC

PL eNB,APUT

PL eNB,UE

PL eNB,STAUT

AP UT

PL APUT,AP

PL APUT,STAUT

PL AP,STAUT

PL STAUT,STA

PL APUT,STA

AP

PL AP,STA

• Cabled-up test in controlled environment

− 80 MHz max bandwidth assumed for Wi-Fi

LTE-U on one of secondary channels of Wi-Fi

LTE-U UE

STA UT

1x1 AWGN Cabled Up, 802.11ac, 80 MHz

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

STA

18


Full Buffer TCP Throughput Results (Below ED)

Below ED

AP ID CINR TCP Throughput

1 0 dB 0.9x

1 5 dB 1.1x

1 15 dB 1.8x

2 0 dB 0.9x

2 5 dB 0.95x

2 15 dB 1.85x

Baseline throughput = x (50% of isolated Wi-Fi throughput – optimistic assumption)

CINR = Wi-Fi signal energy / LTE signal energy at Wi-Fi STA

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

19


Wi-Fi YouTube Streaming Results – Wi-Fi/Wi-Fi vs. Wi-Fi/LTE-U

Below ED

Metrics

Baseline

(w/ Wi-Fi)

AP 1 AP 2

w/ LTE-U Baseline

(w/ Wi-Fi)

w/ LTE-U

No. of video stalls 6.00 1.67 12.33 2.67

Playback interrupt time (s) 7.94 7.40 4.77 2.00

Initial Buffer Time (s) 0.67 1.33 0.33 0.33

Avg. Thpt (Mbps) 3.49 4.49 3.17 5.89

Avg. PHY Rate (Mbps) 19.81 39.88 65.65 83.55

LTE-U hit is an erasure

LTE-U with CSAT is a better neighbor than Wi-Fi

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

20


Overview of OTA Campus Network

OTA Campus Network

WC

LTE-U eNB

WB

Wi-Fi AP1

(Above CCA-ED)

Wi-Fi AP2

(Below CCA-ED)

Inset Video Footage

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

21


Carrier Wi-Fi VoIP: Primary CH, Above ED, 160msec CSAT, 2/20

Puncturing

AP backs off for energy during the

LTE-U ON period except for gaps

in the waveform provided by

MBSFN when packets are flushed.

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

22


Carrier Wi-Fi VoIP: Primary CH, Below ED, 160ms CSAT, 2/20

Puncturing

Sequence Number Time Trace

2645

2640

Sequence Number

2635

2630

VoIP packets sent during gaps

in the CSAT ON duration

2625

2620

CSAT OFF duration

2615

13.3 13.4 13.5 13.6 13.7 13.8

Time (s)

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

23


Carrier Wi-Fi VoIP: Primary CH, 160ms CSAT, 2/20 MBSFN

• Wi-Fi VoIP Calling Statistics†

Scenario |Jitter| > 50ms Packet Loss > 3 Consec. Packet Loss

Above ED 6.6% 1.1% 0.1%

Below ED 1.0% 0.7% 0.0%

1

0.9

Primary CH, Below ED

Primary CH, Above ED

DL VoIP Packet-to-Packet Jitter CDF

1

0.9

Primary CH, Below ED

Primary CH, Above ED

DL VoIP Cycle-to-Cycle Jitter CDF

0.8

0.8

0.7

0.7

0.6

0.6

CDF

0.5

CDF

0.5

0.4

0.4

0.3

0.3

0.2

0.2

0.1

0.1

0

-60 -40 -20 0 20 40 60

DL VoIP Packet-to-Packet Jitter (in ms)

0

-60 -40 -20 0 20 40 60

DL VoIP Cycle-to-Cycle Jitter (in ms)

†Calculated per WFA-VE guidelines; Packet arrival one approx. every 20ms

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

24


Carrier WiFi VOIP: Secondary CH, Below ED, 640ms CSAT, 2/20 MBSFN

30

DL Data Rate Time Trace

1

Smart BW mgmt. forces VoIP to go

on the clean primary channel for

PER. No degradation in voice

quality.

DL Packet BW CDF

DL Data Rate (Mbps)

25

20

15

0.9

0.8

10

0.7

0.6

0.5

0.4

0.3

0.2

93.1% of the time packets are

sent on the primary 20MHz

channel.

5

0 1 2 3 4 5 6 7 8 9 10

Time (s)

0.1

0

0 5 10 15 20 25 30 35 40

DL Packet BW

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

25


Fast CSAT: Low latency MAC Activation/Deactivation

• MAC CE for activation is sent xms before start of on time

• If deactivation time (Toff) is < Thr, UE is expected to be ready to decode PDCCH in 1-2ms

(short cDRX behavior)-no minimum 8ms for activation as in spec

• MAC CE for deactivation is sent before end of Ton

• Error events of losing or misinterpreting activation/deactivation have to happen ith low

probability by design (to the level the can be handled and absorbed by UE tracking loops)

CRS ON or T on CRS OFF or T off T on Activation Overhead

δ

• To be accompanied with sending CTS2S with MAC activation after contention



MAC CE

Activation

MAC CE Deactivation

To protect against BI

To protect WiFi rate control

MAC CE

Activation

MAC CE Deactivation

Time UE is ready fro

grants

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

26


Stress Room Testing

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

27


Stress Test Chamber

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

28


Stress Test Chamber Setup

1 WiFi AP and 1 LTE-U node under test

− The WiFi AP-UT is inside an RF shielded box

AP1

AP2

AP3

AP4

AP is inside

RF shielded box

8 Reference WiFi nodes

− 4 APs inside RF box to allow below ED scenarios

STA1

STA2

STA3

STA4

3 Enterprise grade WiFi APs + Controller

− Referred to as Vendor A, B, and C

AP under test

STA8

STA under test

STA7

STA6

STA5

Gigabit

Ethernet

switch

Results presented in these slides

− All links above CCA-ED

− 8WiFi+WiFi vs. 8WiFi+LTE-U

LTE-U duty cycle = 1/9

AP8

AP7

AP6

AP5

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

29


Demo GUI Summary (Vendor A)

LTE-U outperforms

Wi-Fi

LTE-U maintains

Wi-Fi performance

5.8

6.7

LTE-U

6.6

6.8

Wi-Fi

LTE-U

Wi-Fi

Wi-Fi

Baseline

CSAT

Baseline

CSAT

Unit Under Test

Unlicensed throughput

Average Wi-Fi throughput

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

30


Demo GUI Summary (Vendor B- more aggressive in winning medium)

LTE-U outperforms

Wi-Fi

LTE-U maintains/improves

Wi-Fi performance

6.6

4.8

LTE-U

3.9

4.9

Wi-Fi

LTE-U

Wi-Fi

Wi-Fi

Baseline

CSAT

Baseline

CSAT

Unit Under Test

Unlicensed throughput

Average Wi-Fi throughput

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

31


Demo GUI Screenshot (Mixture of Vendor A and B)

Baseline:

8 Wi-Fi + Wi-Fi

Stress Test:

8 Wi-Fi + LTE-U

Vendor B

Vendor A

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

32


Demo GUI Summary (Mixture of Vendor A and B)

LTE-U outperforms

Wi-Fi

LTE-U maintains

Wi-Fi performance

6.7

LTE-U

5.8

5.8

1.9

Wi-Fi

Baseline (A)

LTE-U

CSAT

Unit Under Test

Unlicensed throughput

Wi-Fi Wi-Fi

Baseline CSAT

Average Wi-Fi throughput

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

33


Demo GUI Screenshot (Mixture of Vendor A and C)

Stress Test:

8 Wi-Fi + LTE-U

Baseline:

8 Wi-Fi + Wi-Fi

Vendor A

Vendor C

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

34


Demo GUI Summary (Mixture of Vendor A and C)

LTE-U outperforms

Wi-Fi

LTE-U maintains

Wi-Fi performance

7.1

6.6

6.7

4.5

LTE-U

Wi-Fi

LTE-U

Wi-Fi

Wi-Fi

Baseline

CSAT

Unit Under Test

Unlicensed throughput

Baseline

CSAT

Average Wi-Fi throughput

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

35


Summary of Stress Room Testing

Main Observations

In general LTE-U preserves or enhances Wi-Fi user experience

If Wi-Fi nodes are fair in medium sharing, LTE-U gains are lower, as its

link level gains are not exposed by this dense setup

If Wi-Fi nodes are aggressive in medium occupancy, frequent collisions occur and

LTE-U provides better user experience for WiFi nodes by deterministically

clearing the medium after its fair share decreasing relative number of collisions

LTE-U gains over Wi-Fi are more pronounced in this case and mostly due to MAC

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

36


Fairness Testing

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

37


Introduction to fairness study

We studied how different Wi-Fi APs coexist with each other

− Later compared to coexistence with LTE-U as well

Conducted measurements using commercially available Wi-Fi

Certified WLAN devices

− 5 top-selling retail APs (11ac) based on ‘best-seller lists’ determined from top

industry magazines and online-retailers

− Product diversity of 5 OEMs and 3 chipset-vendors

− Labelled – A thru E

− STA (11ac, 1x1) – a top-selling mobile-device

− Same brand of STA device used with all APs to limit combinations

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

38


Test Setup

Four Networks with 1 AP

having 1 STAs each

− All APs set to VHT20 in

Ch#153

− All combinations of APs

from set of 5 considered

RF Screen Room

Traffic

Source

AP1

STA1

STA2

AP2

Traffic

Source

Both APs are doing fullbuffer

downlink traffic

(iperf) to the STAs

Average throughput

determined over a 90sec

period

Traffic

Source

AP3

STA3

STA4

AP4

Traffic

Source

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

39


Test Results

Different APs achieve different levels of throughput share

Wide variation in channel aggregate throughput of different devices despite similar link

conditions (high SNR regime)

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

40


Fairness between Wi-Fi/Wi-Fi and Wi-Fi/LTE-U

Test Setup

AP1

STA1

STA under test

AP under test

Gigabit

Ethernet

switch

Tests are performed using two Wi-Fi connection pairs across different AP models, and one LTE-U link with one Wi-Fi AP link across different AP models

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

41


Consumer grade Fairness with and without LTE-U (Air time)

Testing pair-wise airtime sharing

across 4 Wi-Fi AP models and

between Wi-Fi/LTE-U

1

0.9

0.8

W in 1W+1W

W in 1W+1L

L in 1W+1L

Duty

Duty Cycle CDF

0.7

0.6

CDF

0.5

0.4

• Points corresponding to

coexistence with AP model A

• AP model A does not obey

CCA-ED rule and does not

backoff to LTE-U

Points corresponding to all 4 Wi-Fi AP models

0.3

0.2

0.1

0

0 10 20 30 40 50 60 70 80 90

Percentage

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

42


Thank you

Follow us on:

For more information, visit us at:

www.qualcomm.com & www.qualcomm.com/blog

©2013-2015 Qualcomm Technologies, Inc. and/or its affiliated companies. All Rights Reserved.

Qualcomm is a trademark of Qualcomm Incorporated, registered in the United States and other countries. All trademarks of Qualcomm Incorporated are used with

permission. Other products and brand names may be trademarks or registered trademarks of their respective owners.

References in this presentation to “Qualcomm” may mean Qualcomm Incorporated, Qualcomm Technologies, Inc., and/or other subsidiaries or business units within the

Qualcomm corporate structure, as appli

.

Qualcomm Incorporated includes Qualcomm’s licensing business, QTL, and the vast majority of its patent portfolio. Qualcomm Technologies, Inc., a wholly-owned subsidiary

of Qualcomm Incorporated, operates, along with its subsidiaries, substantially all of Qualcomm’s engineering, research and development functions, and substantially all of its

product and services businesses, including its semiconductor business.

©2013-2015 Qualcomm Technologies, Inc. All Rights Reserved.

43

More magazines by this user
Similar magazines