Emerging Wireless Standards 09 - Cognitive Radio Technologies

http://www.crtwireless.com/EWS_09.htmlAvailable online

Emerging Wireless StandardsWireless Summer SchoolWireless Summer School

June 4, 2008Session C2

1:30‐5:00 PM

James [email protected](540) 230‐6012www crtwireless comwww.crtwireless.com

Presentation Overviewhttp://www.crtwireless.com/EWS_09.htmlAvailable online

Presentation OverviewEmerging Standards(11) Cellular (other than LTE and ( ) (

WiMAX)(10) LTE(11) WiMAX(11) WiMAX(11) 4G(24) WLAN( 8) Interoperability Standards( 8) Interoperability Standards( 9) WPAN(11) Cognitive Radio Standards

http://www.wisoa.net/members_logos/mobile_internet‐big.jpg

(10) Summary and Trends

Break & Poster SessionBreak & Poster Session2:45‐3:30

CellularCellular

(other than LTE and WiMAX)

Cellular OverviewCellular Overview• Two primary competing approaches to 3G

– 3GPP Family• GSM GPRS EDGE WCDMA TD SCDMA

3GPP Declared IP

• GSM, GPRS, EDGE, WCDMA, TD‐SCDMA (WCDMA‐TDD), HSCSD, HSPDA, LTE, LTE Advanced

• Promotional www.gsmworld.com• Standards www.3gpp.org

– 3GPP2 Family• CDMAOne (IS‐95a,b), 1xRTT, 1xEVDO, 1xEVDV, UMB

• Promotional http://www.cdg.orgd d• Standards www.3gpp2.org

– One vision• Voice + high speed data + mobility

– One dominant IP holder (Qualcomm)Oth

3GPP2 Declared IP

• Other– Mobile WiMAX and WiMAX II (802.16m)– Standard http://wirelessman.org/– Promotional http://www.wimaxforum.orgL IP

4 /114

– Lower cost IP• 350 companies own essential IP• http://www.eetimes.eu/design/197007324

Source: “3G Cellular Standards and Patents”, David J. Goodman and Robert A. Meyers

GSM Dominates the LandscapeGSM Dominates the Landscape

http://www.coveragemaps.com/gsmposter_world.htm

• 3GPP (GSM/WCDMA) has most of the market (77% in 2005, 83% in 2006, 86.6% in 2008)– Most of that lead is in GSM3GPP2 ( d 2000) i j 3GPP• 3GPP2 (cdma2000) got a massive jump on 3GPP– 418/431 million of CDMA is 3G (www.cdg.org)– 3GPP2 = 11.4%, 3GPP = 5.6%

• WiMAX just cranking up but will be deploying years ahead of LTEWiMAX just cranking up but will be deploying years ahead of LTE

3GPP Technologies3GPP Technologies

• Generic Access Network (UMA)Supports handoffs between GSM networks

• High Speed Downlink Packet Access – W‐CDMA downlink– Supports handoffs between GSM networks

and 802.11 or Bluetooth networks• Packet Switched Handoffs

– Enables easier handoffs between different 3GPP networks

M lti di B d t/M lti t S i

– W‐CDMA downlink • 8‐10 Mbps (and 20 Mbps for MIMO systems) over

a 5MHz bandwidth–Adaptive Modulation and Coding (AMC), –MIMO (Release 6)–Hybrid ARQ

ll k• Multimedia Broadcast/Multicast Services– Simultaneous broadcast of data streams to

multiple recipients

–All IP core network• (Release 4)• Originally ATM

• High Speed Uplink Packet Access (Enhanced UpLink)

–Similar technologies to HSDPA on uplink–Similar technologies to HSDPA on uplink–AT&T in 350 markets

• http://www.mobileburn.com/news.jsp?Id=4660• Loosely coincides with launch of 3G iPhone

HSPA+“HSPA Evolved”Adds MIMO28/42 Mbps DL12 Mbps YULLatency reduction (2ms on TTI)AT&T upgrade

http://www.informationweek.com/blog/main/archive/2009/05/ 72 b h l j i id 35T02OIs/2009/05/att_says_72mbps.html;jsessionid=35T02OIX4PFBAQSNDLOSKH0CJUNN2JVN

Table from: http://www.umtsworld.com/technology/images/hsdpa.png

TD‐SCDMATD SCDMA• Time Division – Synchronous CDMA

– Synchronized uplink channels aided by joint detectionChina’s 3G technology

• 3GPP

LCR TDD（R4）

LCR TDD（R5）

LCR TDD(R6) MC-CDMA

TDDOFDMA

TDDSC-FDMA

/OFDMA TDD

LTE TDD

LCR TDD(R7)

• 3GPP

LCR TDD（R4）

LCR TDD（R5）

LCR TDD(R6) MC-CDMA

TDDOFDMA

TDDSC-FDMA

/OFDMA TDD

LTE TDD

LCR TDD(R7)

– China’s 3G technology• Core network is almost the same as WCDMA

– Requires mature 2G (GSM) network for implementation

• CCSA

TD-SCDMAN F TD-SCDMA

TD-SCDMAStage III

（R6/R7）

• CCSA

TD-SCDMAN F TD-SCDMA

TD-SCDMAStage III

（R6/R7）

• Part of the 3GPP (3rd Generation Planning Partnership Project)

• Multiple chip rates– LCR: 1.28 Mcps, 1.6 MHz BWHCR: 3 84 Mcps 5 MHz BW

Multi-carrier

TD SCDMA Stage I

（R4 2003/03）

N FrequencyBands Cell

SCStage II（R5） TD-SOFDMA

Current status Short Term Evolution Long Term Evolution

Multi-carrier

TD SCDMA Stage I

（R4 2003/03）

N FrequencyBands Cell

SCStage II（R5） TD-SOFDMA

Current status Short Term Evolution Long Term Evolution

– HCR: 3.84 Mcps, 5 MHz BW• TDD link

– Does not use paired frequency bands• Optimum for symmetric and asymmetric data

servicesb d d h ll fl bl

TD‐SCDMA Multiple Access Options

ZTE Corporation, “3GPP Specification Evolution”2005 20072005 2007

– 1.6 MHz bandwidth allows flexibly spectrum allocation

• Frequency band: – 2010 MHz ‐ 2025 MHz in China (WLL 1900 MHz ‐ 1920 MHz)

7 /114

)• Partially motivated by avoiding paying Qualcomm royalties

B. Li, D. Xie, S.Cheng, J. Chen, P. Zhang, W.Zhu, B. Li; “Recent advances on TD‐SCDMA in China,” IEEE Comm. Mag, vol 43, pp 30‐37, Jan 2005

Significant Issues DeployingSignificant Issues Deploying• Standardized in 1999• Was going to roll out in 2004

– http://www.commsdesign.com/news/market

• However, China has made it a point of national pride to have the network running for the 2008 Olympics

– http://www.highbeam.com/doc/1G1‐150687033.html– Is already being tested in 10 cities (includes the Olympic

iti ) b t ti id li t b i d bhttp://www.commsdesign.com/news/market_news/OEG20030102S0009

• Then 2005– http://www.chinadaily.com.cn/english/doc/2004‐06/23/content_341749.htm

• Then 2006

cities) but nationwide licenses may not even be issued by the Olympics

• http://www.thestandard.com.hk/news_detail.asp?pp_cat=1&art_id=54099&sid=15557306&con_type=1

– First commercial trials supposed to begin April 1, 2008• http://www.tdscdma‐

alliance.org/english/news/list.asp?id=4426First public demos in May went badly• Then 2006

– http://www.accessmylibrary.com/premium/0286/0286‐9623636.html

• Then 2007– http://www.theage.com.au/news/Technology/Chi M bil l h 3G bil i

– First public demos in May went badly • http://www.pcworld.com/businesscenter/article/146128/china

_shows_off_olympic_techsort_of.html• China won’t allow 3G or WiMAX until TD‐SCDMA takes off

• Developed a bad reputation– http://homepage.mac.com/dwbmbeijing/iblog/SiHu/C520China‐Mobile‐to‐launch‐3G‐mobile‐services‐

end2007/2007/02/12/1171128898337.html• Now will reportedly issues licenses in 2008

– http://news.zdnet.com/2110‐1035_22‐6207356.html

http://homepage.mac.com/dwbmbeijing/iblog/SiHu/C520534961/E20060302210839/index.html

– Unnamed China Mobile engineer – “you GIVE me a TD‐SCDMA network, and I wouldn't take it.“

– 419,000 Users end of 2008• But 320,000 used it for free in Nov 2008• http://www.marbridgeconsulting.com/marbridgedaily/2009‐01‐

• Delays make Chinese state‐owned service providers unhappy

– Grumblings about forgoing TD‐SCDMA from ChinaMobile (primary deployer)

– http://www.forbes.com/markets/feeds/afx/20

09/article/22889/china_mobile_records_419k_td_scdma_users

Nonetheless, iPhone and Android coming to China apparently with support for TD‐SCDMA

http://www.theregister.co.uk/2009/06/01/chinese_iphone_rumored_for_july/http://www smartbrief com/news/ctia/storyDetails jsp?issueid=71C0

8 /114

p // / / / /06/01/31/afx2489964.html

http://www.smartbrief.com/news/ctia/storyDetails.jsp?issueid=71C09C05‐825C‐40FF‐B9ED‐3AE0CEA92342&copyid=18BF787F‐8CBC‐4772‐9E89‐C8ABD10C4B1D

3GPP2 Technologies3GPP2 Technologies• cdma2000 1xRTT

Packet switched (always on)• EVDO Rev A

Wide deployment– Packet‐switched (always on)– Maximum of 144kbps

• Typical 40‐60 kbps– 2G / 3G

• 1x EVDO

–Wide deployment• Verizon, Sprint, Kindle best known

–Features• Higher modulation uplink• Multi‐user packets (time‐slots)• Lower Latency– CDMA EVolution Data Only

• Designed to support only data applications– VOIP

• Also known as:– CDMA 1x EV‐DO– CDMA EV‐DO

Lower Latency• Couple new data rates downlink (changed code rate)

• Promotional–http://www.evdoinfo.com

– Can offer data rates of 384kbps ‐ 2.4Mbps• Does not mix voice traffic with data traffic• Changes modulation, # timeslots

• EVDV (Voice + Data)– Dead on arrival

EVDO and HSPA Coverage

– Dead on arrival• http://telephonyonline.com/mag/telecom_evdv_

dead/index.html• Qualcomm halted work on the standard in 2005

– http://news.com.com/Cell+phone+makers+to+adopt+Internet+calling/2100‐7352_3‐5618191 html

9 /114

5618191.html– Slow to field

http://www.verizonwireless.com

EVDO Rev B (TIA‐856 RevB)EVDO Rev B (TIA 856 RevB)

• Adds Multiple carriers – 2xEVDO, 3xEVDO,…U 15 1 25 MH i i hi 20 MH– Up to 15 1.25 MHz carriers within 20 MHz

• Adds support for 64‐QAM modulation• DL 73.5 Mbps • UL 27 Mbps• Dynamic non‐contiguous carrier allocation• Support for single carrier and multiple carrier subscribers• Standardized 2006• Trials mid‐2007 • Software upgrade (at BTS) to Rev A• Commercial deployments?

– Alcatel‐Lucent announced products April 2008http://www cdg org/news/search/2008/04/vendor/040108– http://www.cdg.org/news/search/2008/04/vendor/040108_ven_f.html

– No subscribers listed in 4Q 08 by CDMA Development Group• http://www.cdg.org/worldwide/cdma_world_subscriber.asp

EVDO Rev C (UMB)EVDO Rev C (UMB)• Spec published Sep 24, 2007

h // d / / /200• Killed when Verizon went with

LTE– http://www.cdg.org/news/press/2007/Sep24_07.asp

– 3GPP2 (UMB) beats 3GPP to market again

– Chipsets available nowish

LTE– http://www.phoneplusmag.com/

hotnews/79h20122346.html– Dead on ArrivalChipsets available nowish

• http://www.qualcomm.com/press/releases/2007/070327_complete_solution_ultra.html

• Data rates, mobile with 20 MHz bandwidth

• http://www.abiresearch.com/products/research_brief/Wireless_Infrastructure_Research_Briefs/112

– Qualcomm initially differsbandwidth– DL: 288 Mbps– UL: 75 Mbps

• Key technologiesOFDMA MIMO b f i

– Qualcomm initially differs (ineffectually)

• http://www.fiercebroadbandwireless.com/story/qualcomm‐ceo‐umb‐not‐dead‐yet/2008‐01‐14

– OFDMA, MIMO, beamforming– Flexible spectrum allocation– Enhanced QoS– Support for multiple access

technologies

y

– Qualcomm announced end of UMB work, shift to LTEhttp://www.forbes.com/feeds/afx/2008/11/13/afx5692165.html

technologies– Reduced latency

3GPP2 Statistics3GPP2 Statistics• “More than 301 operators in 108 countries have selected CDMA2000® as their 3G platform ofselected CDMA2000® as their 3G platform of which 280 offer commercial services”

http://www cdg org/technology/3g/resource/cdg markettrendsfacts english pdfhttp://www.cdg.org/technology/3g/resource/cdg_markettrendsfacts_english.pdf

http://www.cdg.org/technology/3g/resource/CDMA450%20Market%20Facts.pdf

IEEE 802.20IEEE 802.20 • Allegations of process abuse brought to a screeching halt when standard suspended in

• Fill performance gap between “high data‐rate, low mobility 802 standards” and “high mobility cellular

k ”g p

September• Project Launched 2004• Looked to be dead in the water

– Flarion leading proposall l d h ld

networks”• From QTDD/QFDD Proposal• OFDMA data channel• CDMA control channel• Bandwidths

– Qualcomm leading vote holder• Turned around when Qualcomm bought Flarion(Aug 05)

– http://www.dailywireless.org/modules.php?name=News&file=article&sid=4532

– 5 MHz – 20 MHz• MIMO

– Single, multiple code word– Pseudo‐ Eigen beamforming

• Space Division Multiple Access• Went to proposal downselection process

– Qualcomm (Flarion) TDD, FDD– ETRI– BEST‐WINE (Kyocera)

R d i D 06

p p– Separate mode from MIMO

• Data Rate 260 Mbps–MIMO, 20 MHz

• Turbo coding• Time‐frequency hopping • Reapproved in Dec 06

• First meeting Jan 2007• Approved June 2008

– http://www.ddj.com/mobile/208403913

• Available:

Time frequency hopping• Supposed to support inter Radio Access Technology handoffs

• Similar to UMB– UMB is effectively an upgrade to MBFDD version– IEEE C802.20‐07/14 Available:

– http://standards.ieee.org/getieee802/802.20.html

• Working on conformance, minimum performance requirements

/







3GPP Long Term Evolution(LTE)

3GPP Long Term Evolution (LTE)3GPP Long Term Evolution (LTE)

• Evolved Universal Terrestrial Radio Access• Downlink: Adaptive multilink OFDM (AML‐

Approximate Deployment ScheduleE‐UTRA Air Interface

Downlink: Adaptive multilink OFDM (AMLOFDM), which means different bandwidths based on demand

– Variable prefix size• 4.7 ms to 16.7 ms• Intent to support up to 120 km cells

C ll d Hi h S d OFDM P k t A HSOPA– Called High Speed OFDM Packet Access or HSOPA• Uplink: SC‐FDMA (more later)• DL 100 Mbps in 20 MHz (5 bps/Hz)• UL 50 Mbps in 20 MHZ (2.5 bps/Hz)• Reduced transition time between states (such as

http://www.motorola.com/staticfiles/Business/Solutions/Industry Solutions/Service Providers/Wireless Operators/LTE/_Document/6993_MotDoc.pdf

(between idle and active states)

• Variable bandwidth allocations: 1.25 MHz, 1.6 MHz, 2.5 MHz, 5 MHz, 10 MHz, 15 MHz and 20 MHz in both the uplink and downlink

• Frequency reuse factor of 1

http://www.3gpp.org/ftp/Specs/html‐info/36‐series.htm

q y– Fractional frequency reuse

• 200 users/cell• Load sharing/policy across radio access technologies

• Support for antenna arrayspp y– Beamforming, MIMO– Space Division Multiple Access http://hgmyung.googlepages.com/scfdma.pdf

LTE NetworkLTE Network• Flat access network

– No central control

– Directly connected via X2 link

– Cooperative resource management

• Evolved Packet Core– All IP Core Network

• All voice is VOIP

– Serving Gateway• Anchor for local mobility between eNodeBs

• Some administrative work for roaming

– Mobility Management Entity• Setup / management of bearer services

• Security

– PDN GatewayIP dd ll i UE

http://hgmyung.googlepages.com/3gppLTE.pdf

HSS• IP address allocation to UE

• QoS Enforcement

– Home Subscriber Server• Hosts user subscription data

PCRF

MME

HSS

PCRFS1‐MME

– PCRF• Sets policy for bearers

UE eNodeBServing Gateway

PDN Gateway

Services / IP Cloiud

S1‐US1‐U

LTE The UMTS Long Term Evolution, S. Sesia, I. Toufik, M. Baker, eds., John Wiley and Sons, 2009

Protocol StackProtocol Stack

• NAS = Non‐access

Control Plane

stratum– Protocols running between core network and UE

• RRC = Radio Resource Controlesou ce o o

• PDCP = Packet Data Convergence Protocol

User Plane

Protocol• RLC =Radio Link Control

http://www.home.agilent.com/agilent/eventDetail.jspx?nid=‐35356.0.08&lc=eng&cc=US&ckey=1381781&id=1381781

Resource ManagementResource Management• Fine grained allocation / scheduling

of subcarriers / timingof subcarriers / timing– Also applies to SCFDMA– Can be based on current fading

conditions• User diversity

• Handoffs– Hard Handoff– Over X1 and over S1

• Multiple levels of packet priorities http://hgmyung.googlepages.com/scfdma.pdf

– QCI = QoS Class Indicatorp g y g g g p g p

B. Furht, S. Ahson, “Long Term Evolution: 3GPP LTE Radio and Cellular Technology” CRC Press, 2009

LTE The UMTS Long Term Evolution, S. Sesia, I. Toufik, M. Baker, eds., John Wiley and Sons, 2009

Single Carrier FDMASingle Carrier FDMA• SC‐FDMA on the uplink

Wh ?• Why?– Gain frequency domain equalization

• Little extra complexity to OFDM• But extra steps to effect SC‐FDMA

– ~2 dB reduction in PAPR• Van Nee and Prasad, OFDM for Wireless

Multimedia Communications, Artech House Publishers, 2000.

– Less power consumption• Better battery life• Better for handsets• Better for handsets

– Less sensitivity to carrier offset– Possibly worse performance in fading channels– Priyanto, B.E.; Codina, H.; Rene, S.;

Sorensen, T.B.; Mogensen, P., “Initial

http://hgmyung.googlepages.com/scfdma.pdf

J Andrews A Ghosh, ; g , ,Performance Evaluation of DFT‐Spread OFDM Based SC‐FDMA for UTRA LTE Uplink,” VTC Spring 2007, pp. 3175 –3179.

J. Andrews, A. Ghosh, R. Muhamed, Fundamentals of WiMAX, Prentice Hall, 2007

More LTE DetailsMore LTE DetailsFrame Structure

• Support for paired / unpaired

http://www.motorola.com/staticfiles/Business/Solutions/Industry Solutions/Service Providers/Wireless

Support for paired / unpaired spectrum

• Multicast / Broadcast Services– Optional now, but expected to

Operators/LTE/_Document/6993_MotDoc.pdf

– Subframes 0,5 must be DL

– Otherwise arbitrary TDD structure

• Other Features

p , pbe supported

• DRX– Aggressive sleep modes guided

b N d B – Interference Mitigation– Extensions

by eNodeB• Latency reduction influenced by

state reduction

UMTS LTE

21 /114http://www.motorola.com/staticfiles/Business/Solutions/Industry Solutions/Service Providers/Wireless Operators/LTE/_Document/6993_MotDoc.pdf

Adrian Scrase, “Overview of the Current Status of 3GPP LTE,” Mobile World Congress, Feb 11‐14, 2008

Deployments• Nokia officially abandoned WiMAX for LTE

// / / / /– http://gigaom.com/2009/03/04/nokia‐plans‐lte‐devices‐for‐2010/

• Japan & China to work on LTE‐flavor– Government level agreement (TD‐SCDMA / LTE?)

– http://tech.yahoo.com/news/afp/20090428/tc_afp/japanchindi l t h l t ladiplomacytechnologytelecom

• First FCC Cert (March 2009)

– LG infrastructure for 1700 AWS

– http://www.engadgetmobile.com/2009/03/18/fcc‐approves‐lte‐gear‐from‐lg‐wimax‐says‐welcome‐to‐the‐party/

h f f• iPhone for LTE for Verizon– http://www.usatoday.com/tech/wireless/phones/2009‐04‐26‐

apple‐verizon‐iphone_N.htm?csp=34

• VodaPhone LTE in German TV Spectrum– http://www.telegeography.com/cu/article.php?article_id=282

96& il h l96&email=html

• LSTI – LTE / SAE Trial Initiative– http://www.lstiforum.org/– Drive industrialization of 3GPP, demonstrate capabilities– Began Interoperability testing May, 09

S. Hattangady, “The promise of 4G LTE (Long‐Term Evolution),” April 2009


Verizon Open Development for LTE(What Hath Google Wrought?)

• Open platform was a requirement of 700 MHz auctionof 700 MHz auction

• Website:– https://www22.verizon.com/opend

ev/• C‐band LTE spec available

– Issued April 2009– Available to “developers”– Anyone can register as a developerAnyone can register as a developer

• I did! (We’re not building LTE devices)

• LTE Specification Web Conference May 13. Developers, online– wwwverizonwireless‐– www.verizonwireless‐

opendevelopment.com Verizon Wireless, “DEVICE REQUIREMENTS LTE 3GPP BAND 13 NETWORK ACCESS” VERSION 0.9 Issued: April 2009

Potential Snags?Potential Snags?• Concern about VOIP voice quality

– http://www.unstrung.com/document.asp?doc_id=176867

• Backhaul Issues?– HSPA site requires 45 Mbps (3x14.4)

– E1 line is 2‐8Mbps– Circuit switch fallback?• http://www.unstrung.com/document

.asp?doc_id=173090

• IPR Issues?H t 3G i il l

E1 line is 2 8Mbps• http://www.unstrung.com/document

.asp?doc_id=176867

– Could be IP over microwave

M t t th t f b kh l– Hurt 3G, similar players• http://www.unstrung.com/document.asp?doc_id=176867

– Starting to work on patent pooling• Sisvel, VIA Licensing and MPEG LA

– Movement to ethernet for backhaul• “Over half of cellular backhaul

capacity will use Ethernet by the end of 2011”

• http://www instat com/press asp?SkSisvel, VIA Licensing and MPEG LA• Started in 2008• http://www.fiercewireless.com/story/lte‐patent‐pool‐efforts‐heat/2009‐05‐26

– Qualcomm made about $8B in 2008

• http://www.instat.com/press.asp?Sku=IN0904603GW&ID=2533

– Also have to work with security and latency requirements

2008 • LTE Setup more complex (flat, inter‐BS management)

– Automation desired

– http://www unstrung com/document– http://www.unstrung.com/document.asp?doc_id=177252

http://files.shareholder.com/downloads/QCOM/315627873x0x189123/2f1a890f‐ef39‐43f6‐bfd3‐d673c59c16c3/QCOM_stockholder_031108_FINAL.pdf







WiMAXWiMAX

802.16d,e,j

802.16 Family (WiMAX)802.16 Family (WiMAX)802.16 Apr 2002 LOS 10‐66 GHz802 16a Apr 2003 2 11 GH802.16a Apr 2003 2‐11 GHz 802.16c Jan 2003 2‐11 GHz 802.16d Oct 2004 Combined 802.16,a,c802.16e Dec 2005 Mobile WiMAX

Commercialization Roadmap

802.16f Dec 2005 Net Management Database (MIB)

802.16g Spring 2007 Network management plane

802 16h 2009 Li t802.16h 2009 License‐exempt Coexistence

802.16i Fall 2008 Mobile Management Information Base

802 16j 2009 Mobile Multihop Relay802.16j 2009 Mobile Multihop Relay802.16k Aug 2007 Network Management802.16m 2010 4G

WiMAX Forum (2006): Mobile WiMAX – Part I: A Technical Overview and Performance Evaluation.

Available at www.wimaxforum.org

Projections based on data at http://grouper.ieee.org/groups/802/16/published.html

802.16e (Mobile WiMAX, 802.16‐2005)802.16e (Mobile WiMAX, 802.16 2005)

• Ideally 802 16 + mobilityPHY Spec Overview

Ideally, 802.16 + mobility– Really intended for nomadic or low mobility– Not backwards compatible with 802.16‐2004

• http://www.unstrung.com/document.asp // g /p?doc_id=76862

• Direct competitor to 3G, 4G, 802.20 though WiMAX Forum once said otherwise

• Numerous ongoing deployments and working systems, particularly for WiBRO

• PHY– Scalable OFDM + Optional MIMO– Convolutional turbo codes– Optional block turbo codes, LDPC

28 /114WiMAX Forum (2006): Mobile WiMAX – Part I: A Technical Overview and Performance Evaluation. Available at www.wimaxforum.org

Other Mobile WiMAX FeaturesOther Mobile WiMAX Features• Frame‐by‐frame resource allocation • Security

f ff d l d• Hybrid Automatic Repeat Request (HARQ)

• UL and DL Scheduling• Variable QoS

– AES for traffic and control data– EAP– Privacy and Key Management

Protocol Version 2 (PKMv2)Variable QoS• Three handoff methods

– A traditional Hard Handoff (HHO)– Fast Base Station Switching (FBSS)

– 3‐way handshake on handoffs• IP Core Network (supports Voice

Over IP)• Multicast Broadcast Services• A list of reachable base stations is

maintained by mobile and base stations, but base stations discard packets if not the active BS

i i ( HO)

• Multicast Broadcast Services– Like cellular multicast services

• WiBRO– Defines a set of options for

– Macro Diversity (MDHO)• Same list is maintained, but all

base stations in the list can participate in the reception and transmission of packets

pMobile WiMAX for Korean deployment

transmission of packets.

WiMAX SpectrumWiMAX Spectrum• WiMAX Spectrum Alliances• Regulatory DatabaseRegulatory Database

–AT4 Wireless– Launched November 2006–http://www.wimaxforum.org/join/spectrum_demo/

• WiMAX Global Roaming AllianceWiMAX Global Roaming Alliance–Brought together unlicensed providers to promote global roaming

–Now defunct –Will probably come back in some form–http://www.theregister.co.uk/2006/09/29/oz_wi i lli /imax_roaming_alliance/

• WiMAX Spectrum Owners' Alliance–http://www.wisoa.com/–Promotes roaming agreements–Participants:U i d A t li N t k Pl M iti UK

• Recent reports of interference with with C‐B d VSAT

http://www.wimaxforum.org/news/downloads/supercomm_2005/WF_Day_in_a_Life_with_WiMAX_Final.pdf

–Unwired Australia, Network Plus Mauritius, UK Broadband, Irish Broadband, Austar Australia/Liberty Group, Telecom New Zealand, WiMAX Telecom Group, Enertel and Woosh Telecom

• 700 MHz band

Band VSAT– http://www.suirg.org/pdf/SUIRG_WiMaxFieldTestReport.pdf

• Officially declared 3G so 3G spectrum– http://www.wirelessweek.com/WiMAX‐is‐

–http://www.xchangemag.com/articles/501/79h13917183935.html?cntwelcome=1

3G.aspx

Mobile WiMAX DeploymentsMobile WiMAX Deployments• First Mobile WiMAX products certified

April 2008– Now 103 (wimaxforum)

• Mostly 3 5 GHz

802.16d + 802.16e + WiBRO (WiMax Forum)

• Mostly 3.5 GHz, – Already more 802.16e than 802.16d

• Less than 3 million subscribers (3Q08)– ~500,000 on ClearWire– Slightly smaller than largest

Not being adopted by incumbents– Not being adopted by incumbents• April 09 – WiMAX Forum 70% of operators

were greenfield– Success for cellular appears tied to Sprint‐

Nextel / Clearwire effort• Narrow band / high power for private

radio systems (utilities)

All bands

radio systems (utilities)– http://www.wimax.com/commentary/blog/b

log‐2009/may‐2009/introducing‐private‐mobile‐wimax‐0524 http://www.wimaxmaps.org/

3.5 GHz Onlyy

(WiMAX Forum)

Clearwire/SprintClearwire/Sprint • Fixed WiMAX based wireline replacement

Clearwire Coverage

• http://www.clearwireconnections.com/pr/pressreleases/050708.pdf

Merged Company

wireline replacement service to home + portability within coverage area

• 2 Mbps data + voice

• Clearwire + Sprint WiMAX unit– Called Clearwire

• Investors– $3.2 Billion from Google (500 M), Comcast (1 05B) Time Warner (550M) Bright House

S i t

http://www.clearwire.com/

(1.05B), Time‐Warner (550M), Bright House (100M), Trilogy Equity (10M)

– Sprint owns 51%– Clearwire owns 27%– Investors own 22%

N i id fSprint• Mobile WiMAX• Rapid deployment to major cities

– 10,000 sites in preparation– 1750 base stations delivered in 2007, 20,000

• Nationwide focus– 120‐140 million coverage by 2010

• Commercial agreements– Intel will put WiMAX in chipsets

• Had been planning on thatantennas• Incorporated into numerous devices (cameras and televisions)

• Open Network (support Android)• Federal government connectivity via WiMAX

http //www wimaxday net/site/2007/06/05/sprint plans

Had been planning on that– Google services to be carried (and search provider)

– Support Android– Sprint, Comcast, TimeWarner, and Bright House will be wholesale

32 /114

– http://www.wimaxday.net/site/2007/06/05/sprint‐plans‐wimax‐for‐gov%e2%80%99t‐services/

House will be wholesale– Sprint contributes its 2.5 GHz holdings

• Looks like a slow roll out

WiBroWiBro• Korean version of 802.16e

– Phase 1 standardized by TTA of Korea (2004)y– Phase 2 standardized in 2005

• Korean spectrum allocated 2002– 2.3 GHz (100 MHz)H i ti 802 16 /WiB d N 2004• Harmonization 802.16e/WiBro agreed Nov 2004 – Samsung joined WiMAX Forum Dec 2004– May indicate Samsung’s guess on 4G direction

• KT & SK Telecom launched June 30 2006 in Seoul• KT & SK Telecom launched June 30, 2006 in Seoul http://kt.co.kr/kthome/kt_info/pr/news_center/news_view.jsp?page=1&no=397&gubun=1

• Just 200,000 subscriptions since launch (12/08).– Possibly influenced by data‐only – http://www.intomobile.com/2008/12/28/wibro‐voice‐calls‐

coming‐to‐south‐korea‐wimax‐networks‐in‐2009.html

How does WiBRO relate to 802.16e?How does WiBRO relate to 802.16e?

• WiMAX Forum: (http://www.wimaxforum.org/news/press releases/WiBro and Mobile WiMAX Backgrou(http://www.wimaxforum.org/news/press_releases/WiBro_and_Mobile_WiMAX_Backgrounder.pdf)

– “WiBro is the service name for Mobile WiMAX in Korea. WiBro uses the Mobile WiMAX System Profile. The system profile contains a comprehensive list of features that the equipment is required or allowed to support, and, as a result, WiBro offers the same capabilities and features of Mobile WiMAX ”Mobile WiMAX.”

– It’s Mobile WiMAX, just with a different profile (frequency, bandwidth…)• Vendors: WiBRO is compatible with 802.16e, but there’s more to Mobile WiMAX than just

802.16e compatibility and many choices in WiBRO are different from what is mandatory in 802 16e802.16e

– From (http://www.nortel.com/solutions/wimax/collateral/wimax_wibro_white_paper.pdf)• Important implementation differences from Nortel white paper

– Mandatory Handoff • 802 16e = HHO• 802.16e = HHO• WiBRO = FBSS

– HARQ• 80.16e = Chase combine HARQ• WiBRO = Incremental redundancy HARQy

– Likely (though unclear) network layer differences

802.16j Mobile Multi‐hop Relay• Expand coverage, capacity by adding

relay stations

802.16j Mobile Multi hop Relay• Relays controlled from base

stationsrelay stations• Intended for licensed operation• Not intended as a mesh network

– Actually a tree

stations• Fixed Relay

– Permanent installation– Useful for coverage holesN di R l• Support mobile units • Nomadic Relay

– Temporary fixed installation– Extra capacity for special events

(military SDR conferences)• Mobile Relay• Mobile Relay

– Placed on mobile platform to support users on the platform

– Useful for public transport (buses, trains)(buses, trains)

Modified from Fig 1 in IEEE 802.16mmr‐05/032

LTE vs WiMAXLTE vs WiMAX• Technically, very similar• LTE a little later, a little more refined,• Came from different industries

– WiMAX more from the computing industry

– LTE more from the cellular industry– Probably biggest factor in deployment

trends and IP• LTE upgrades, WiMAX deploys new

– Easier spectrum for LTE, More niche markets for WiMAX– More niche markets for WiMAX

• Both seem to be going to more “open” systems

– Development, not so much open‐source• Will WiMAX voice like issues also be• Will WiMAX voice‐like issues also be

experienced in LTE?• Both working towards more

distributed control and more complex topologies


topologies







4G4G

IMT‐Advanced, 802.16m, LTE‐Advanced

4G (IMT‐Advanced)4G (IMT Advanced)• Been looking towards 4G for a while…

– 1 Gbps

• ITU Requirements (Nov / Dec 2008)

• Report ITU‐R M.2134: “The key features of IMT‐Advanced are:

– a high degree of commonality of ITU Requirements (Nov / Dec 2008)

– Report ITU‐R M.2134: Requirements related to technical performance for IMT Advanced radio interface(s) (2008 ) http://www.itu.int/publ/R‐REP‐M 2134‐2008/en

g g yfunctionality worldwide while retaining the flexibility to support a wide range of services and applications in a cost efficient manner;REP‐M.2134‐2008/en

– Report ITU‐R M.2135: Guidelines for evaluation of radio interface technologies for IMT‐Advanced ( 2008) http://www.itu.int/publ/R‐REP M 2135 2008/en

manner;– compatibility of services within IMT and with fixed networks;

– capability of interworking with other radio access systems;REP‐M.2135‐2008/en y ;

– high‐quality mobile services;– user equipment suitable for worldwide use;

– user‐friendly applications, services y pp ,and equipment;

– worldwide roaming capability;– enhanced peak data rates to support advanced services and applications • (100 Mbit/s for high and 1 Gbit/s for low mobility were established as targets for research)3G Americas, “Defining 4G: Understanding the ITU Process for the Next

Generation of Wireless Technology,” July 2007 Available online: http://3gamericas.com/PDFs/3G_Americas_Defining_4G_WP_July2007.pdf

IMT‐Advanced Requirements (1/2)Report ITU‐R M.2134

• Capacity • Mobility classes are defined:i k /h– Stationary: 0 km/h

– Pedestrian: > 0 km/h to 10 km/h– Vehicular: 10 to 120 km/h– High speed vehicular: 120 to 350

k /h

– In terms of voice channels, actually

km/h

Cell Spectral Efficiency

In terms of voice channels, actually not that impressive

• Assume 10kHz• 1 MHz / 10 kHz = 100 users

• Efficiencies– Peak Link efficiencies

• DL (4x4) 15 bit/s/Hz• UL (2x4) 6.75 bit/s/Hz

Cell‐edge Spectral Efficiency

Tables from: Report ITU‐R M.2134. Available: http://www.itu.int/publ/R‐REP‐M.2134‐2008/en

IMT‐Advanced Requirements (2/2)IMT Advanced Requirements (2/2)

• Handover interruption • Bandwidthsa do e e up o a d d s– 40 MHz

– 100 MHz research target

• Latency – Control (idle to active) =

100 ms

– User (time to move IP packet across link) = 10 ms

Declared Intention to Support IMT‐d dAdvanced

•Several candidates already emerging–LTE‐Advanced–802.16mNTT D C M ’ 5 Gb t t–NTT DoCoMo’s 5 Gbps prototype•http://www.nttdocomo.com/pr/files/20070209_attachment02.pdf

–China’s home grown standardg•http://www.forbes.com/markets/feeds/afx/2007/09/25/afx4151478.html

•Common techniques• Perhaps an impractical

number of antennas for direct

42 /114http://www.nttdocomo.com/pr/files/20070209_attachment01.pdf

Common techniques–OFDMA, MIMO, small cell sizes optimized for low speed, but support for high speed, IP b kb

communication with a handset– 12x 12 MIMO– Large implied bandwidths to

achieve goals – 100 MHzbackbone achieve goals – 100 MHz• Hard to find

802.16m802.16m• TGm System Requirements Document

– http://wirelessman org/tgm/docs/80216• Minimum Peak Rate

– Downlink 6.5 bps/Hz

Requirements

– http://wirelessman.org/tgm/docs/80216m‐07_002r4.pdf

– http://wirelessman.org/tgm/docs/80216m‐07_003.pdf

– http://wirelessman.org/tgm/index.html• Key functionalities to be added (not

p /– Uplink 2.8 bps/Hz

• Latency less than 802.16e• Radio Resource Management

– Reporting, interference management– Multicast broadcast service• Key functionalities to be added (not

defined yet)– Routing – Self Organization – Multi‐Carrier

– “High‐resolution” location determination• Internetworking with:

– 802.11 3GPP, 3GPP2• Coverage optimized for 5 km, functional to 30‐100 kmO i i d f l bili ( 15k h)– Multi‐Radio Coexistence • Optimized for low mobility (<15kph), maintain connection up to 350 kph

• Optimized for contiguous spectrum but support discontiguous

• Reuse/share bandwidth with legacy systemssystems

• Direct migration from 802.16e• Recognition of niche markets:

– “IEEE 802.16m shall be able to support public safety first responders, military and emergency services such as call prioritizationemergency services such as call‐prioritization, pre‐emption, and push‐to‐talk.”

IEEE C802.16m‐07/002r1

TGm TimelinesTGm Timelines• Formal announcement ofannouncement of 802.16m for IMT‐Advanced (Feb 2009)h //– http://www.reuters.com/article/pressRelease/idUS140783+16‐Feb‐2009+BW200902162009+BW20090216

• Mid‐2009 System Description

• Sep 2009 First IMT l

2009‐03‐12

proposal• Final spec – May 2010

LTE‐AdvancedLTE Advanced

• RequirementsRequirements– http://www.3gpp.org/ftp/Specs/html‐/ftp/Specs/htmlinfo/36913.htm

– 3GPP TR 36.9133GPP TR 36.913 V8.0.1 (2009‐03)

• Targeting spec

T. Nakamura “Update of E‐UTRA and IMT Advanced Requirements,” ATIS‐3GPP LTE Conference, January 26, 2009

Targeting spec completion end of 20102010

Requirements q3GPP TR 36.913 V8.0.1

• Available: http://www 3gpp org/ftp/Specs/archi

• Efficiency:DL 30bps/Hhttp://www.3gpp.org/ftp/Specs/archi

ve/36_series/36.913/• Date: March 2009• Peak Data Rate

At l t IMT Ad d

– DL= 30bps/Hz– UL = 15 bps/Hz.

• “Assumption of antenna configuration is 8x8 or less for DL and 4x4 or less for UL”– At least IMT‐Advanced

– Research target• 100 Mbit/s for high

– 1 Gbit/s for low mobility • The system should target a downlink

4x4 or less for UL• Spectrum Flexibility

– 450−470 MHz band, – 698−862 MHz band, – 790−862 MHz bandThe system should target a downlink

peak data rate of 1 Gbps and an uplink peak data rate of 500 Mbps.

– The system should be able to support at least 300 active users without DRX in a 5 MH b d idth Th b f

790 862 MHz band, – 2.3−2.4 GHz band, – 3.4−4.2 GHz band, and– 4.4‐4.99 GHz band

• LatencyMHz bandwidth. The same number of RRC connections with DRX as in Release 8 E‐UTRA and E‐UTRAN (16k) is expected

• Coexistence– With other operators, with other flavors

Latency

p ,of 3GPP, legacy systems

LTE Comparison SummaryLTE Comparison Summary

http://www.atis.org/LTE/documents/IMT‐Advanced%20requirements.pdf

4G Takeaways4G Takeaways• 802.16m, LTE‐Advanced leading candidates

4G d ’t i th b th t h• 4G doesn’t raise the bar that much– Expect very similar to 802.16e, LTE– *Very* aggressive timelines– Then again, LTE = 3.9G

• High data rates ma not be achie ed• High data rates may not be achieved– Mostly due to bandwidth availability concerns (hard to find 100 MHz)– Channel bonding a necessity

• Explicit consideration of modes other than traditional vehicular / pedestrian modes (indoor microcellular)modes (indoor, microcellular)

• Continuing to increase emphasis on interoperability, services, openness• Capacity now measured in VoIP calls• Will the not‐so‐distant standardization impact the deployment of 802.16e / LTE?• Implication of large antenna arrays

– Too large for handhelds (8x8 @ 450 MHz)– Likely need femtocells (relays) to make it work







Wireless LANsWireless LANs

802.11

Jun 1997 802.11 2 Mbps ISM

802.11 Alphabet SoupSep 1999 802.11a 54 Mbps UNIISep 1999 802.11b 11 Mbps ISMOct 2001 802.11d global roamingJun 2003 802.11f interoperability

Past dates are standards approval dates. Future dates from 802.11 working group timelines

Jun 2003 802.11g 54 Mbps ISMOct 2003 802.11h spectrum managementJun 2004 802.11i securityOct 2004 802.11j Japanese spectrum

timelinesLetters are working group (WG) designations.

Letters assigned alphabetically as groups created

Sep 2005 802.11e real time QoSMay 2008 802.11k RRM measurementsMay 2008 802.11r fast roamingNov 2008 802.11y US 3.65 GHz

created.

No WG/ WG document

802.11c MAC Bridging work incorporated into 802.1d

Sep 2009 802.11w packet securityNov 2009 802.11n 100 MbpsJan 2010 802.11z Direct Link SetupJun 2010 802.11v network management

p802.11l “typologically unsound”802.11m doc maintenance802.11o “typologically unsound”802 11q too close to 802 1q

Jun 2010 802.11p vehicular (5.9)Sep 2010 802.11u external networksSep 2010 802.11s mesh networksJun 2011 802.11aa Video Transport Streams

802.11q too close to 802.1q802.11x generic 802.11 standard

802.11t (test) was canceled

51 /114

Dec 2012 802.11ac Very High Throughput < 6GHzDec 2012 802.11ad Very High Throughput 60GHz

http://grouper.ieee.org/groups/802/11/Reports/802.11_Timelines.htm

WiFi AllianceWiFi Alliance• Industrial consortium that promotes

802.11 Millions of WiFi Chipset Shipped802.11– www.wi‐fi.org

• Certifies interoperability between vendors’ products

• Certifies consistency with standards

Millions of WiFi Chipset Shipped

• Fills in the gap when 802.11 standards process is too slow (draft n)

• WiFi success owes significant debt to WiFiAlliance

• Line between 802 11 standards community• Line between 802.11 standards community and WiFi Alliance has gotten very blurry

• Certifications– 802.11a/b/g/n WiFi– 802.11e Wireless Multimedia80 . e e ess u t ed a– Draft 2.0 n– EAP (authentication)– (Optional) Wi‐Fi Protected Setup– (Optional) Wi‐Fi Multimedia

(O ti l) WMM P S

Wi‐Fi Alliance, Introducing Wi‐Fi Protected Setup™, January 3, 2007

– (Optional) WMM Power Save – (Optional / Mandatory) Wi‐Fi in handsets

Distributed Coordination Function (DCF)Distributed Coordination Function (DCF)

• Intended to combat “hidden nodes” d d k din an uncoordinated network and

generate fair access to channel•Basic components:

After aitin DIFS after last dete ted–After waiting DIFS after last detected transmission, source sends Request to Send (RTS)

–Destination replies with Clear to Send (if OK)

–Data is then transferred and ACKed–If an error occurs (e.g., collision), then station has to wait for DIFS +

• Network allocation vector– Acts as virtual carrier sense

/then station has to wait for DIFS + random backoff.

• Random backoff grows with # of collisions

– Duration given in RTS/CTS fields

• DIFS = DCF Interframe Space• SIFS = Short Interframe Space

802.11 overhead802.11 overhead

• Significant overhead involved inSignificant overhead involved in 802.11– RTS/CTS/ACK SIFS

– TCP, IP, MAC framing

– Real throughput is rarely come l PHYclose to PHY raw rate

wireless.ictp.trieste.it/school_2002/lectures/ermanno/System_Performance.ppt http://www.cs.tut.fi/kurssit/TLT‐6556/Slides/Lecture4.pdf

802.11n overview802.11n overview•Adds MIMO to WLAN OFDMO t i ith UNII ISM b d•Operate in either UNII or ISM bands

• Status:–In Ballot–But held up by IP battles

Streaming Home Multimedia (HDTV)

• http://arstechnica.com/news.ars/post/20070924‐dark‐australian‐patent‐cloud‐looms‐over‐802‐11n‐spec.html

• CSIRO (http://www.csiro.au/) holds some key IP, hadn’t signed letter of assurance, has history of WiFi lawsuits and sought injunctionshistory of WiFi lawsuits and sought injunctions

– Got bought off

• Last freely available draft–Enhanced Wireless Consortium (merger of TGnSync and WWiSE)

–http://www.enhancedwirelessconsortium.org/home/EWC_PHY_spec_V127.pdf (PHY)

–http://www.enhancedwirelessconsortium.org/home/EWC_MAC_spec_V124.pdf(MAC)

Source: http://www.tgnsync.org/products

(MAC)

802.11n PHY (in 1 slide)802.11n PHY (in 1 slide)•MIMO evolution of 802.11 OFDM PHY

–Up to 4 antennas per devicep p•20 and 40MHz channels

–Fully interoperable with legacy 802.11a/b/g–288 Mbps in 20MHz and 600 Mbps in 40MHz (64 QAM 4 spatial streams 1/2 guard(64 QAM, 4 spatial streams, 1/2 guard interval)

–Claim of 100 Mbps in real throughput

•Optional enhancementsT i b f i i h li ibl–Transmit beamforming with negligible overhead at the client

–Advanced channel coding techniques (RS)–Space Time Block Coding (Alamouti and others)others)

–1/2 guard interval (i.e., 400ns instead of 800 ns)

–7/8 rate coding

http://www.enhancedwirelessconsortium.org/home/EWC_PHY_spec_V127.pdf

802.11n MAC Features802.11n MAC Features• Supports 802.11e (QoS)• Frame aggregation

–Single and multiple destinations• Bi‐directional data flow• Link adaptation with explicit feedbackLink adaptation with explicit feedback and control of channel sounding packets

• Protection mechanisms–For seamless interoperability andcoexistence with legacy devicescoexistence with legacy devices

• Channel management –Including management of 20/40MHzoperating modes

–Channel estimation and feedback B d “802 11 N G i Wi l LAN–Channel estimation and feedback• Power management for MIMO receivers•Data aggregation

Broadcom, “802.11n: Next‐Generation Wireless LAN Technology,” White Paper, April 06

802.11n Certification802.11n Certification• Wi‐Fi Alliance

Certifying to Draft 2 0 while

Key Certification Features

– Certifying to Draft 2.0 while draft is approved

– Certify to Ratified Standard when available

• out of sponsor ballot to final wgballotballot

• Nov 2009– 22 August 2007 ‐ Almost 70

products certified for compliance with Draft 2.0 of the 802 11nthe 802.11n

• http://www.wi‐fiplanet.com/news/article.php/3578886

– Now >500 (lost count)h // f• http://certifications.wi‐fi.org/wbcs_certified_products.php?search=1&advanced=1&lang=en&filter_company_id=&filter_category id &filter subcategory

Wi‐Fi CERTIFIED™ 802.11n draft 2.0: Longer‐Range, Faster‐Throughput, Multimedia‐Grade Wi‐Fi® Networks

y_id=&filter_subcategory=&filter_cid=&date_from=&date_to=&x=30&y=18&selected_certifications%5B%5D=33

802.11p Operation802.11p Operation• “Dedicated Short Range Communications”

(DSRC)d– Started in IEEE 1609, spun into 802.11p

– Aka (WAVE) Wireless Access for Vehicular Environment

• IEEE 802.11a adjusted for low overhead joperations– 54 Mbps, <50 ms latency– 5.850 to 5.925GHz band

• Spectrum divided into 7 bands• Spectrum divided into 7 bands– 178 is control (safety)– 2 edge channels are reserved for future– The rest are service channels (not

l f )application specific)• Mix of roadside‐to‐vehicle and vehicle‐to‐

vehicle communications• Questions on business model

D. Jiang, V. Taliwal, A. Meier, W. Holfelder, R. Herrtwich, “Design of 5.9 ghz dsrc‐based vehicular safety communication,“ IEEE Wireless Comm, Oct 06, pp. 36‐43

Questions on business model– http://www.rita.dot.gov/press_room/press

_releases/index.html

802.11p Applications802.11p Applications

COLLISION COLLISIONCOLLISIONCOLLISION COLLISIONCOLLISION

• Emergency warning system for vehicles C ti Ad ti C i C t l IMMINENT

FRONT

Note 1: The OBU in the vehicle recognizing the threat transmits a WARNING and COLLISION PREPARATION

In-Vehicle Displays and Annunciations

~ ~

IMMINENT

LEFT

IMMINENT

FRONT

IMMINENT

FRONT

Note 1: The OBU in the vehicle recognizing the threat transmits a WARNING and COLLISION PREPARATION

In-Vehicle Displays and Annunciations

~ ~

IMMINENT

LEFT

IMMINENT

LEFT• Cooperative Adaptive Cruise Control • Cooperative Forward Collision Warning

• Intersection collision avoidance MESSAGE with the location address of the threat vehicle.

~ ~~ ~

Note 2: Only the OBU in the threatening vehicle processes the message because only it matches the threat address.

up to 100 m (328 ft)Note 3: COLLISION PREPARATION includes seat belt tightening, side air bag deployment, side bumper expansion etc

MESSAGE with the location address of the threat vehicle.

~ ~~ ~

Note 2: Only the OBU in the threatening vehicle processes the message because only it matches the threat address.

up to 100 m (328 ft)

up to 100 m (328 ft)Note 3: COLLISION PREPARATION includes seat belt tightening, side air bag deployment, side bumper expansion etc

• Approaching emergency vehicle warning (Blue Waves)

• Vehicle safety inspection • Transit or emergency vehicle signal

Traffic Signal

Traffic Signal

OBUs on Control Ch

Radar Threat Identification

expansion, etc.

Car NOT Stopping

Traffic Signal

Traffic Signal

OBUs on Control Ch

Radar Threat Identification

expansion, etc.

Car NOT Stopping

g y gpriority

• Electronic parking payments • Commercial vehicle clearance and safety inspections

From: IEEE 802.11- 04/ 0121r0 Available:http://www.npstc.org/meetings/Cash%20WAVE%20Information%20for%205.9%20GHz%20061404.pdf

y p• In‐vehicle signing • Rollover warning • Probe data collection • Highway rail intersection warning• Highway‐rail intersection warning

802.11r overview802.11r overview• Fast BSS Roaming/Transition within IEEE WLAN networks

– Preserve security with handovers <50msF BSS R i i ibl l i hi i ll d h bili d i• Fast BSS Roaming is possible only within a certain area called the mobility domain (MD), inter‐MD cases are not covered

– Mobility Domain (MD): Set of BSS grouped together with the same 48bit MD Identifier– FT functionality seeks to provide handover performance for RT servicesK I• Key Issues

– Resource Reservations– Security

• Collapsed 5 step process down to 3– Scanning – active or passive for other APs in the area– Authentication with a (one or more) target AP– Re‐association to establish connection at target AP

• Released 2008

http //www cs tut

61 /114

http://www.cs.tut.fi/kurssit/TLT‐6556/Slides/Lecture4.pdf

Reduction in Roaming TimeReduction in Roaming Time

S. Bangolae, C. Bell, E.Qi, “Performance study of fast BSS transition using IEEE 802.11r,” International Conference On Communications And Mobile Computing, 2006

http://www.networkcomputing.com/gallery/2007/0416/0416ttb jh l j i id 0CK4 K 20HC QQS L CKHSCJU 2JV

62 /114

b.jhtml;jsessionid=0CK4ZKR20HC5QQSNDLPCKHSCJUNN2JVN

802.11s• Modify 802.11 MAC to create

dynamic self‐configuring network of access points (AP) called and

802.11s

of access points (AP) called and Extended Service Set (ESS) Mesh

• Automatic topology learning, dynamic path selection

• Single administrator for 802 11i• Single administrator for 802.11i (authentication)

• Support up to 32 AP• Support higher layer connections

IP or Ethernetpp g y

• Allow alternate path selection metrics

• Extend network merely by introducing access point andintroducing access point and configuring SSID

1. http://standards.ieee.org/board/nes/projects/802-11s.pdf

802.11s802.11s• Key Technologies

–Topology Formation• Open 802.11s (Linux)

– http://www.open80211s.org/N WiFi h d– Internetworking

–Routing–Security

• Numerous WiFi mesh products– http://www.cs.wustl.edu/~jain/cse574‐06/ftp/j_jmesh/sld019.htm

Deployment Scenarios

64 /114http://ieee802.org/802_tutorials/nov06/802.11s_Tutorial_r5.pdf

J. Hauser, D. Shyy, M. Green, MCTSSA 802.11s Military Usage Case

Too much 2.4 GHz InterferenceToo much 2.4 GHz Interference• WiFi is very popular ‐> Very crowded

spectrum• Ofcom sponsored study• Ofcom sponsored study

(http://www.ofcom.org.uk/research/technology/research/exempt/wifi/wfiutilisation.pdf)– Generally more interference from other

impolite devices– Cities have more congestion802 11 Ci t d (802 11 08/1440 0)• 802.11a Cisco study (802.11‐08/1440r0)• Expect AP‐AP collisions in urban deployments

802.11‐08/1440r

Mass Consulting Ltd., “Estimating the Utilisation of Key Licence‐Exempt Spectrum Bands,” April 2009. Available online: http://www.ofcom.org.uk/research/technology/research/exempt/wifi/wfiutilisation.pdf

802.11aa802.11aa• Expectation of interference means degradation of performance

• 802.11‐08/0764r1“There are man a s in hich a diodegradation of performance

– Potentially quite bad for video, voice over WiFi

• 802.11aa PAR Goals (doc.: IEEE 802.11‐07/1972r14):

– “There are many ways in which audio video data is streamed over IP networks

– Some methods lend themselves to selective packet discarding, some do notP b bl h b i h i

/ )– “Graceful degradation of audio video streams when there is insufficient channel capacity, by enabling packet discarding without any requirement for deep packet inspection,

– Probably the best pragmatic approach is some simple signalling per UDP packet and let the application decide when it can provide discard hints.”

• 802.11‐08/0818r0p ,– Increased robustness in overlapping BSS environments, without the requirement for a centralised management entity,

– Intra‐Access Category prioritization of transport streams by modifying EDCA timing

– Managed contention access– Modify framing / coordination function

to reduce contention• Considering

b d / ltransport streams by modifying EDCA timing and parameter selection without any requirement for deep packet inspection,

– Improved link reliability and low jitter characteristics for multicast/broadcast audio video streams,

– broadcast / multicast– Distributed network management– Extending 802.11s mechanisms

,– Interworking with relevant 802.1AVB mechanisms” (Audio / Video Bridging)

802.11ac APAPAP

AP

AP

APAP

AP

AP

AP802.11ac AP

802.11ac• Very High throughput < 6 GHz• Target Application:

i ( d id i

APAPAP

APAP

802.11‐09/0630r1

– Streaming IPTV (and video in general)

– VoIP, smart phones• Requirements (IEEE 802.11‐

/ )

3 million IPTV subscribers in Japan in 3 yrs

08/1285r):– 15 bps/Hz– 500 Mbps / 40 MHz – single link– 1 Gbps / 40 MHz – multi‐station

802.11‐09/0630r1

p /• Key tech appears to be Spatial

Division Multiple Access– And A LOT of antennas

Up to 42.8 bps/Hz with 16 antennas

802.11‐09‐0532‐00‐00ac

802.11‐09/0303r1802.11‐09/0303r1

Move To 60 GHz UnlicensedMove To 60 GHz Unlicensed• 2001, FCC designated 57‐

64 GHz as unlicensed Strong O2 AbsorptionShorter Propagation

64 GHz as unlicensed– Similar around the globe

– Avoid the interference

More bandwidth => More throughput

http://www.gigabeam.com/technology.cfmhttp://www.gigabeam.com/technology.cfm

More antenna gain (same size)Rain Matters

http://wireless.fcc.gov/outreach/2004broadbandforum/comments/YDI benefits60GHz.pdf

802.11ad802.11ad

• Requirements • Known Issuesq– IEEE 802.11‐08/1285r0– > 1 Gbps @ 10 m

– Coexistence with 802.15.3c (60 GHz version)– Seamless handoffs

between 2.4/5 GHz and 60 GHz

version)

• PAR approved Jan 2009– Came from VHT Study yGroup

• 3 Gbps ~ uncompressed 1080p

Others in the marketOthers in the market

• WHDI (Amimon) • WiGig– variant of 802.11– 3Gbps (including

uncompressed 1080p) in a 40MHz @ 5 GHz

– Website: http://wirelessgigabitalliance.org/

– Big players– http://www.whdi.org/Technolo

gy/• WirelessHD (SiBeam)

V i t f 802 15 3

g p y• Atheros, Dell, LG, Samsung,

Microsoft, Nokia, Panasonic– Spec out end of 2010

• http://www.multichannel.com/a– Variant of 802.15.3c– 4 Gbps @ 60 GHz (4 channels)– OFDM + beamforming + (a

bunch of 802.11‐like services)

rticle/277089‐Wireless_HD_Delivery_Race_Heats_Up.php

)– Spec:

http://www.wirelesshd.org/pdfs/WirelessHD_Full_Overview_071009.pdf

http://www.hometheatermag.com/hookmeup/607hook/

• ~ 1 million units 2009, 13 million units 2013p 2013

• http://www.eetimes.com/news/semi/rss/showArticle.jhtml?articleID=217201265&cid=RSSfeed_eetimes_semiRSS

WiFi on the PlaneWiFi on the Plane• Idle hands are a good source of revenue

12 15% of c stomers pa for it on a plane

Aircell Coverage Area

– 12‐15% of customers pay for it on a plane– $7.95 smart phone, – $9.95 normal, < 3 hrs– $12.95 normal > 3 hrs– http://www broadbandreports com/shownews/1http://www.broadbandreports.com/shownews/1215‐Use‐InFlight‐WiFi‐102611

• Boeing / Connexion– Satellite based connection to plane– Launched in 2004, shut down in 2006 – citing no

market– Still provides services to US Gov planes

• http://www.boeing.com/news/releases/2007/q3/070814a_nr.html

– 800 lbs, 2 weeks to install, $1,000,000 / plane • Aircell (GoGo)

http://www.aircell.com/index.php?option=com_content&task=view&id=312&Itemid=1

, , $ , , / p• Row44

– Satellite based (leased from Hughes)– Planned for use by Southwest, Alaska, Norwegian Shuttle

( )– EVDO‐based ground network– 1,000 planes by end of 2009

• Delta, AirTrans. NorthWest, United, American, Air Canada,

• Virgin (28)– 2 nights to install (use in day), 150 pounds– http://www.row44.com/

Virgin (28)– $100,000 / plane, 1 day turnaround

WiFi on Smart PhonesWiFi on Smart Phones

• Statisticshttp://ipod.about.com/od/iphoneinterfacegallery/ig/iPhone‐Gallery‐‐Settings/iPhone‐WiFi‐Settings.htm

– 802.11‐09‐0453r00ac– 2008 ‐ 56 M (~44%)– 2011‐ 300M

2014 520M (90%)– 2014 – 520M (90%) • Cellular Providers Establishing WiFiRelationships for free service

– AT&T + Qwest (May 09)• http://blogs.wsj.com/digits/2009/05/06/qwe

st‐unveils‐wi‐fi‐deal‐with‐att/st unveils wi fi deal with att/– AT&T buys WayPort (Nov 08)

• http://online.wsj.com/article/SB122598801123705301.html

– Verizon with Boingo• http://wifinetnews com/archives/2009/05/v• http://wifinetnews.com/archives/2009/05/v

erizon_broadband_subscribers_get_free_wi‐fi.html

– T‐Mobile• https://content.hotspot.t‐

mobile.com/AssetProcess.asp?asset=com.default.main.001

– Sprint‐Nextel – odd man out??

802.11‐09‐0453r00ac

WiFi in … Bluetooth?WiFi in … Bluetooth?

• Bluetooth 3.0 Highspeed is 802.11nBluetooth 3.0 Highspeed is 802.11n– http://www.wi‐fiplanet.com/news/article.php/3816556

• Some Bluetooth advocates say that this is no yconcession as WiFi is not just 802.11







Interoperability StandardsInteroperability Standards

802.21, UMA/GAN 802.11u Industry Standards

802.21 (Media Independent Handoffs)802.21 (Media Independent Handoffs)• Key Services

– Triggers (state change, predictive, network initiated)– Network Information (services, maps, list of available networks)Network Information (services, maps, list of available networks)– Handover commands (client or network initiated, vertical handoffs)

• Jan 2009: Targeted publication• Follow on (mostly study groups)

b l d d ff ( l d d d– Mobile Broadcast Handoffs (e.g., Digital Video Broadcasting, MediaFLO, Digital Multimedia Broadcast)

– Inter‐network handoff Security– Multi‐radio power

l– Deployment Scenarios, – Emergency Services

http://www.ieee802.org/802_tutorials/july06/802 21‐IEEE‐Tutorial.pptV. Gupta, “IEEE 802.21 MEDIA INDEPENDENT HANDOVER,” IEEE 802.21 session #15 July 17, 2006

•UMA allows to access the mobile voice and data services of the cellular network over adata services of the cellular network over a Wireless LAN

• Subscribers are enabled to roam and handover between cellular networks and wireless networkswireless networks

•Mobile devices access the Core Network through Unlicensed Mobile Access Network (UMAN).

•UMAN has 3 major entities htt // t dUMAN has 3 major entities –Unlicensed wireless network –IP access network –UMA Network controller (UNC)

• UNC authorizes and authenticates the Mobile

http://www.umatoday.com

UNC authorizes and authenticates the Mobile devices for accessing the Core Network

• Part of 3GPP now–Generic Access Network (GAN)

• Products– http://www.umatoday.com/mobileHandsets.php

http://www.umatoday.com

VCCVCC• Addresses shortcomings in UMA’s Voice Call Continuity (VCC)• Status

– Initiated within 3GPP in June 20052006 through requirements has only fully completed the requirements stage– 2006 through requirements, has only fully completed the requirements stage

– 2008 being extended to IMS Service Continuity • Different technologies agree to virtual channel

– Joint control impossible• Some argue inferior to UMAg

– http://www.kineto.com/products/downloads/kineto_wp_UMA_VCC_2007.pdf• Some argue better than UMA

– http://www‐hk.huawei.com/publications/view.do?id=1480&cid=2622&pid=127

http://www.kineto.com/products/downloads/kineto_wp_UMA_VCC_2007.pdf

802.11u (Interworking with External k )Networks)

• Standard out in 2010Standard out in 2010

• Specifically addresses handoffs where user not preauthorized (generally because from another p (g ynetwork)

• Major Topicsj p– Network Selection

– Emergency Call support

– Authorization from Subscriber Network,

– Media Independent Handover Support S ti i f ti d t l f 802 21• Supporting information and control messages of 802.21

Other Cellular + WiFiOther Cellular + WiFi• Seamless Converged Communication Across

Network (SCCAN)– Motorola, Proxim, Avaya – Enterprise solutions– http://www.sccan.org/

• Mobile Integrated Go‐to‐Market Network IP Telephony Experience (Mobile IGNITE)

Mobile IGNITE

p y p ( )– Vendor driven solution– http://www.bridgeport‐

networks.com/partners/mobileignite.html• Wireless Wireline Convergence Working GroupWireless Wireline Convergence Working Group

– Alcatel, Cisco (International Packet Communications Consortium)

– http://www.packetcomm.org/index.shtml• Bottom line• Bottom line

– Even for standard convergence type activities there’s many different emerging standards.

– Need for special radios to navigate standards?

http://www.spectralink.com/products/images/NL‐01.gif

Interoperability SummaryInteroperability Summary• Popularity of WiFi is driving a lot of

integrationintegration– WiFi + Cellular– WiFi + Bluetooth

• Proliferation of standards + lack of silver bullet standard means that

Mobile voice trafficFixed voice traffic

Rapid growth ?

silver bullet standard means that optimal access technology will vary

• Supporting growth in use of VoIP will make networks less dependent on particular access technologies

Fixed voice trafficVoIP traffic

p g• Biggest issues are:

– Getting industries to agree (Herding cats)

– Managing security across

-03-02-01-00-99-98-97-96-95-94-93 -07 e

-92 -04 -05e

-06e

-08 e

Source: Nokia 13.6.2005Technology used to terminate the call

heterogeneous networks– Harmonizing handoff routines

• 802.21 is the standard that ties together the vertical handoff t d dstandards

Other Proprietary StandardsOther Proprietary Standards• Kleer

– http://www.kleer.com• TransferJet (Sony)

– http://www.transferjet.org/en/tj/tj_overviewh l – Proprietary low power RF for audio /

video• En‐Ocean

– http://www.enocean.com/en/

.html–Features

• http://www.sony.net/SonyInfo/News/Press/200801/08‐002E/index.html

• Electric induction field coupling• 3 cm range – Best known as energy scavengers

– Runs a proprietary wireless mesh protocol

• Insteon

• 3 cm range• 4.48 GHz center frequency• 560 Mbps (Max) 375 Mbps (effective)

–Adaptive modulation• “Touch & Get: transfer by touching devices together

– Mixes power line comm with RF comm

– Industry Alliance (15 manufacturers)• http://www.insteon.net/alliance‐

about.html

g• Some registration security

–Demonstrated shortly after announced• http://www.engadget.com/2008/01/06/video‐sonys‐transferjet‐gets‐demonstrated/

– Specification completed about.html• Wireless Valley is a member

– Open source http://www.efundies.com/

p p• http://www.toshiba.co.jp/about/press/2009_05/pr1901.htm?from=RSS_PRESS&uid=20090519‐569e

– 15 members in forum, mostly Japanese







Wireless Personal Area NetworksWireless Personal Area Networks

Industry and Open StandardsIndustry and Open Standards

• Proprietary / Industry802.15 Standards p y / y– Zigbee (on 802.15.4)

• Zigbee Pro

Bl t th ( i i ll )

802.15.1 April 2002 Bluetooth802.15.2 Oct 2003 Coexistence802.15.3 Jun 2003 High data rate802.15.3a UWB (high rate)802.15.3b Doc Maintenance

– Bluetooth (originally)– WiBree– WiMedia

802.15.3c Sep 2009 mm‐wave PHY802.15.4 May 2003 zigbee (PHY/MAC)802.15.4a March 2007 UWB (low rate)802.15.4b Sep 2006 Updates 802.15.4 document802.15.4c Jan 2009 Chinese WPAN PHY

– Z‐Wave– En‐Ocean

802.15.4d Mar 2009 950 MHz in Japan802.15.4e Sep 2010 MAC for 802.15.4c802.15.4f Mar 2011 Active RFID802.15.4g Jun 2011 Low data rate (40 kbps) smart meter802.15.5 Apr 2009 WPAN Mesh

– Insteon– Keer– TransferJet

p802.15.6 >2009 Body Area Networks802.15.7 >2010 Visible LightIGThz Terahertz interest group (300 GHz+)SGRFID RFID Study Group

TransferJet802.15.3a disbanded Jan 2006MBOA technologies became WiMediaHigh speed DS‐UWB basically dead after Freescale pulled out

Emerging 802.15 StandardsEmerging 802.15 Standards• 802.15.4c (China)

–779‐787 MHz band–779‐787 MHz band –Two PHY Modulations: MPSK PHY and O‐QPSK–Considering OFDM, beamforming

• 802.15.4d (Japan)• 802.15.5e MPNC

MPNC802.15.5e–Enhanced MAC for Industrial applications–Modified MAC for 802.15.4c changes

• 802.15.5–Mesh networking

PN 3

PN 2

MPNC

MPCN

MPNC

MPNC

Mesh

• Terahertz study group–300 GHz ‐> 3 THz –http://www.ieee802.org/15/pub/IGthz.html

• TG6 – Body Area Networkshtt // i 802 /15/ b/TG6 ht l

PN 1IEEE P802.15.5™/D0.01, July 2006

–http://www.ieee802.org/15/pub/TG6.html–Just starting

• 802.15.7 Visible Light Interest Group–http://www.ieee802.org/15/pub/IGvlc.html

ZigBeeZigBee

– the software

Standardhttp://www.zigbee.org/en/spec_download/download_request.asp

the software – Network, Security & Application layers– Brand management

• IEEE 802.15.4– “the hardware” – Physical & Media Access Control layers

Security32 / 64 / 128 bit ti

Application

API

ZigBee

Customer

Physical & Media Access Control layers• PHY

– 868MHz/915MHz, 2.4 GHz– Band specific modulations– 20‐250 kbps

• MACMAC

NetworkStar / Mesh / Cluster-Tree

32- / 64- / 128-bit encryption ZigBeeAlliance

• MAC– CSMA‐CA channel access– Support for ad‐hoc networks

• Zigbee Pro (Industrial grade)– Network Scalability

Fragmentation

PHY868MHz / 915MHz / 2.4GHz

MACIEEE

802.15.4

Silicon Stack App – Fragmentation– Frequency Agility– Automated Device Address Management– Group Addressing– Centralized Data Collection– Wireless Commissioning

• Open source implementations• Open‐ZB

h // b /

Source: http://www.zigbee.org/en/resources/

Silicon Stack App

– Wireless Commissioning– http://www.open‐zb.net/• Meshnetics Open‐MAC

– http://www.meshnetics.com/opensource/mac/• SmartGrid

– http://www.zigbeeexpo.com/

WiMediaWiMedia• Industry alliance from MBOA 802.15.3a• Standardized for US in Dec 2005 in ECMA‐

368 and 369

From: http://www.wimedia.org/en/events/documents/02WiMedia_Overview_CES2006.ppt

368 and 369– http://www.ecma‐

international.org/publications/standards/Ecma‐368.htm

– ECMA used specifically to avoid 802 standardization problems

&/or

PAL PAL WiNET

&/or &/or

PAL

IP&/or

PAL PAL WiNET

&/or &/or

PAL

IPIP

standardization problems• PHY

– Multiband OFDM QPSK– 53.3, 80, 106.7, 160, 200, 320, 400, 480

Mbps nominal data ratesR f 10 i d

WiMedia UWB Radio Platform

MAC & Policies

UWB PHY (MB-OFDM)

WiMedia UWB Radio Platform

MAC & Policies

UWB PHY (MB-OFDM)

WiFi802.11

ac/ad will be superior– Range of 10 m indoor

– Data can be interleaved across 3 bands, 7 defined patterns (channels)

– Mandatory support for band group 1 • MAC

• Wireless USB out in Dec 2006–Hub‐spoke model–Mandatory support for band group 1–Mandatory rates of 53.3, 106.7, 200 Mbps– Initial Belkin device didn’t live up to the hype

PAL: Protocol Adaptation LayerPAL: Protocol Adaptation Layerbe superior

– Peer to Peer, Ad‐hoc– AES 128– Support for Dynamic Channel Selection– Ranging via propagation delay

measurements

• http://www.eetimes.com/news/latest/showArticle.jhtml?articleID=196602148

• http://www.usb.org/developers/wusb/wusb_2007_0214.zip

• Bluetooth 3.0 devices in 2008– http://gizmodo.com/gadgets/wireless/nextge

n‐bluetooth‐30‐on‐the‐way‐179684.php• Wireless Firewire and IP also supported over

– Bluetooth‐like information discoveryWireless Firewire and IP also supported over WiMedia standard

From Fig 28:

UWB Now Dead (or on its Deathbed)UWB Now Dead (or on its Deathbed)The Wireless USB crowd is imploding,"

"Those still in existence are hunkering “By 2013, the only UWB solutions still on the market will be proprietary solutions ing

down, and I don't see a lot of traction. They have a lot of competition from 802.11n which is doing a lot of what the WiMedia Alliance promised,"

the market will be proprietary solutions in the industrial/medical segment. All other UWB will be gone by the end of the forecast period" in 2013, he added.”

p ,

Tan Rao, co‐founder of Radiospire

• WiMedia Disbanded

Tan Rao, co founder of Radiospirehttp://www.eetimes.com/news/semi/rss/showArticle.jhtml?articleID=217201265&cid=RSSfeed_eetimes_semiRSS

– March 2009– http://www.phonescoop.com/news/item.php?n=41133

– Sold off IP to Wireless USB, Bluetooth

• Public data: (from wibree.com (no more) and http://www theregister co uk/2006/10/06/wibree analysis/ )http://www.theregister.co.uk/2006/10/06/wibree_analysis/ )– 2.4 GHz ISM band– Range 10 meters– 1 Mbps data rate – Targets low power/low cost market– From

http://www.computerworld.com.au/index.php/id;992123146;fp;4;fpid;18• Up to 8 devices Master/Slave• Turns off frequency hopping• Expects different technology to serve as backbone between masters• Expects to share resources with full Bluetooth

• Many reports mentioned WiBree as a competitor to Bluetoothy p p– Brought into Bluetooth fold as low power alternative– “Bluetooth Low Energy”– https://www.bluetooth.org/About/bluetooth_sig.htm#Bluetooth%20Wirele

ss%20Technology

90 /114

ss%20Technology– Now a competitor to Zigbee

Z‐WaveZ Wave

• Originally Zensys proprietary– http://www.zen‐sys.com/

• Industry standard “Z‐wave”– http://www z‐wave com

Z‐Wave Alliance

http://www.z wave.com

• Low power alternative to Zigbee• PHY

– 9.6 kbps or 40 kbps– GFSK– 100 ft range

• 900 MHz ISM• http://www.z‐wavealliance.com/• http://en wikipedia org/wiki/Z Wave

http://www.z‐wave.com/modules/AboutZ‐Wave/?id=21&chk=4ed024468cb3d7f9095aa54227ea197a

• http://en.wikipedia.org/wiki/Z‐Wave

Other Proprietary StandardsOther Proprietary Standards• Kleer

– http://www.kleer.com• TransferJet (Sony)

– http://www.transferjet.org/en/tj/tj_overviewh l – Proprietary low power RF for audio /

video• En‐Ocean

– http://www.enocean.com/en/

.html–Features

• http://www.sony.net/SonyInfo/News/Press/200801/08‐002E/index.html

• Electric induction field coupling• 3 cm range – Best known as energy scavengers

– Runs a proprietary wireless mesh protocol

• Insteon

• 3 cm range• 4.48 GHz center frequency• 560 Mbps (Max) 375 Mbps (effective)

–Adaptive modulation• “Touch & Get: transfer by touching devices together

– Mixes power line comm with RF comm

– Industry Alliance (15 manufacturers)• http://www.insteon.net/alliance‐

about.html

g• Some registration security

–Demonstrated shortly after announced• http://www.engadget.com/2008/01/06/video‐sonys‐transferjet‐gets‐demonstrated/

– Specification completed about.html• Wireless Valley is a member

– Open source http://www.efundies.com/

p p• http://www.toshiba.co.jp/about/press/2009_05/pr1901.htm?from=RSS_PRESS&uid=20090519‐569e

– 15 members in forum, mostly Japanese







Emerging Cognitive StandardsEmerging Cognitive Standards

Cognitive RadioCognitive Radio

• An approach to wireless engineering pp g gwherein the radio, radio network, or wireless system is endowed with the capacities to:

i l if d i i f i– acquire, classify, and organize information (aware)

– retain information (aware)– apply logic and analysis to informationapply logic and analysis to information

(reason)– make and implement choices (agency)

about operational aspects of the radio, network or wireless system in a mannernetwork, or wireless system in a manner consistent with a purposeful goal (intelligent).

– “Cognitive Radio Definitions and Nomenclature,” Working Document SDRF‐06‐R‐0009‐V0 08Working Document SDRF‐06‐R‐0009‐V0.08

802.11h – Unintentionally Cognitive

• Dynamic Frequency Selection (DFS)

802.11h Unintentionally Cognitive

(DFS)– Avoid radars

• Listens and discontinues use of a channel if a radar is present

– Uniform channel utilization

• Transmit Power Control (TPC)Transmit Power Control (TPC)– Interference reduction

– Range control

P ti S i– Power consumption Savings

– Bounded by local regulatory conditions

802.11y• Ports 802.11a to 3.65 GHz – 3.7 GHz (US Only)

– FCC opened up band in July 2005

802.11y

– Ready 2008• Intended to provide rural broadband access• Incumbents

– Band previously reserved for fixed satellite service (FSS) and radar installations – including offshore

– Must protect 3650 MHz (radar)– Not permitted within 80km of inband government radar– Specialized requirements near Mexico/Canada and other incumbent users

• Leverages other amendments– Adds 5,10 MHz channelization (802.11j)– DFS for signaling for radar avoidance

(802.11h)W ki i h l• Working to improve channel announcement signaling

• Database of existing devices– Access nodes register at

http://wireless fcc gov/uls

97 /114

http://wireless.fcc.gov/uls– Must check for existing devices at same

site

Source: IEEE 802.11‐06/0YYYr0

802.16h802.16h• Draft1 to ballot Oct 06, 67%

approve, resolving pp , gcomments)

• Made it out of group, but not past sponsor ballot

– PAR extended to end of 2009– Going to Draft 9

• Improved Coexistence• Improved Coexistence Mechanisms for License‐Exempt Operation

• Explicitly, a cognitive radio standard

• Incorporates many of the hot d

p ytopics in cognitive radio

– Token based negotiation– Interference avoidance– Network collaboration– RRM databases

• Coexistence with non 802 16h• Coexistence with non 802.16h systems

– Regular quiet times for other systems to transmit

• Strongly suspect not going anywhere From: M. Goldhamer, “Main concepts of IEEE P802.16h / D1,” Document Number: IEEE

C802 16h 06/121 1 N b 13 16 2006y

– Meetings seem sparsely attended

– http://wirelessman.org/le/index.html#09_0004

C802.16h‐06/121r1, November 13‐16, 2006.

IEEE 1900 (SCC 41)IEEE 1900 (SCC 41)• Site: http://www.scc41.org/• 1900.1 – Terminology and Concepts1900.1 Terminology and Concepts• 1900.2 ‐ Recommended Practice for

Interference and Coexistence Analysis – http://crtwireless.com/blog/2008/04/02/1

9002‐approved/• 1900.3 – Conformance Evaluation for SDR

modulesmodules• 1900.4 – Architectural Building Blocks

– network resource managers – device resource managers– the information to be exchanged between

the building blocks sg‐whitespace‐09‐0057g• 1900.4a: Architecture and Interfaces for

Dynamic Spectrum Access Networks in White Space Frequency Bands

• 1900.4.1: Standard for Interfaces and Protocols Enabling Distributed Decision Making forEnabling Distributed Decision Making for Optimized Radio Resource Usage in Heterogeneous Wireless Networks

• 1900.5 – Policy Languages• 1900.6 – Information exchange between

sensors• Demographic trivia

– Majority Japanese– 1900.4 dominated by Europeans

sg‐whitespace‐09‐0057

White SpacesWhite Spaces• Concept:

–Allow unlicensed secondary users into i d TV b d i DSA

• High data rates will not be easy– Channels are only 6 MHzy

unoccupied TV bands via DSA• Initial problems with devices• Approved Nov 2008

– FCC 08‐260A• http://hraunfoss.fcc.gov/edocs_public/attachm

atch/FCC 08 260A1 pdf

Channels are only 6 MHz– High data rates will require channel bonding

• Issues in 802.11n when bonded 40 MHz 802.11n system coexists with 802.11b

– DTV is about 19 Mbps– Large scale MIMO is hard because of long

l hatch/FCC‐08‐260A1.pdf

•Requirement highlights (see Monisha’stalk for more detail)

– Two classes: fixed /portable– Daily geolocation database updates (DTV)

wavelength– “WiFi on Steroids” will mean range, not rate

• Lots of use cases being proposed– Home data (WiFi)– Video distributiony g p ( )

– =Sense for microphones• Licensed and unlicensed• ‐114 dBm• 60s

– Industry to define coexistence protocol

– Mobile DTV rebroadcast– Control apps

• Lots of potential coexisting protocols– 802.22– CogNeAy p

– Industry to define signal for transmitting to allow interferer identification

• Spectrum availability varies widely– Looks like greatest availability for rural

applications

g– Downbanded WiFi

• Easy if only database

applications• Lots of objections from TV broadcasters

and wireless microphone manufacturers / users

Selected White Space PlayersSelected White Space Players• White Space Alliance

– Pushed for opening White Spaces• White Space Database Group

– Goal to quickly turn around database thatPushed for opening White Spaces– Disbanded after decision

• CogNeA– www.cognea.org– Members

• Samsung Phillips HP Motorola Georgia Tech

Goal to quickly turn around database that satisfies FCC specs

– Members include Phillips, Comsearch, Dell, Google Inc., HP, Microsoft Corporation, Motorola Inc., MSTV, and NeuStar and broadcastersG l• Samsung, Phillips, HP, Motorola, Georgia Tech,

ETRI– Focused on low power personal/portable CR

devices, with focus on video• Simple design (OFDM PHY)• Compressed HD Video

C iti f t

– Google: • “We don’t plan to become a database

administrator ourselves, but do want to work with the FCC to make sure that a white spaces database gets up and running. We hope that this will unfold in a matter of months, not years.”

• 802 Executive Committee Study Group on– Cognitive features• Geolocation, sensing, DFS, TPC

– Publishing through ECMA (TG‐48 TG1)– Target end of 2009

• 802.19 White SpaceC i f 802

802 Executive Committee Study Group on White Spaces

– Assess the impact of the FCC White Space R&O on IEEE 802 activities

– Identify Use Cases of TV White Space Spectrum

– Identify what functions may be common– Coexistence for 802– Scope scaled back to use cases– Follow on from 802 ECSG on WS

• SDRF CRWG– Identifying use cases enabled by more

hi i d CR d b

Identify what functions may be common across 802 technologies

– Begin technical discussion on how to enable coexistence between various 802 technologies in the shared TV white space spectrum

– Prepare a Tutorial for March Plenarysophisticated CR database

• SDRF T&M• Testing procedures / issues for CR

– Make recommendations to 802 EC by March 2009 on next steps

– Transitioned coexistence work to 802.19 (Spring 2009)

802.22802.22• Wireless Regional Area Networks (WRAN)

– First explicit cognitive radio standardp g– Aimed at bringing broadband access in rural and remote areas– Takes advantage of better propagation characteristics at VHF and low‐UHF– Takes advantage of unused TV channels that exist in these sparsely

l t dpopulated areas– Still defining inter‐cell coexistence

• Likely published in 2009S d l hil FCC ttl– Some delay while FCC regs settle

– Many similarities with 802.16• Particularly at PHY

• 802 22 1802.22.1– Enhanced interference protection

• 802.22.2– Best practices for deployment

102 /114

– Best practices for deployment

Features of 802.22Features of 802.22• Data Rates 5 Mbps – 70 Mbps• Point‐to‐multipoint • TDD (mandatory), FDD (ok)

DFS TPC

• 802.16 MAC plus the following–Multiple channel support–Coexistence

I b t• DFS, TPC• Adaptive Modulation

– QPSK, 16, 64‐QAM, Spread QPSK• OFDMA on uplink and downlink• Use multiple contiguous TV channels

• Incumbents• BS synchronization (CBP)• Dynamic resource sharing

–Clustering support–Signal detection/classification routines• Use multiple contiguous TV channels

when available• Fractional channels (adapting around

microphones)• Space Time Block Codes

B F i

–Beacon support• Collaborative sensing

–Quiet periods scheduled by BS–Fall back channels

• Techniques in 802 22 will be extended• Beam Forming– No feedback for TDD (assumes channel

reciprocity)• 802.16‐like ranging

Techniques in 802.22 will be extended to other standards and to other bands around the world

• Security and spectrum sharing among base stations are larger research issues nowissues now

Incumbent or other CR user(except microphone user)

TV incumbent user Microphone userFractional useof TV channel

Other CR user or non-microphone incumbent (regulations permitting)

Incumbent or other CR user(except microphone user)

TV incumbent user Microphone userFractional useof TV channel

Other CR user or non-microphone incumbent (regulations permitting)

Channel BondingFrequency Switching

6 MHz Unused(6 MHz)6 MHz

f

GuardBand

6 MHz Unused(6 MHz)6 MHz

f

GuardBand

Cognitive Radio SummaryCognitive Radio Summary• White space has

i d h• Standards tend to be highly

focused on specific bandsenergized research• But CR is more than just DSA

focused on specific bands– But problems vary greatly by band– Blocks may have to be recombined

• Databases appear popular– Opportunistic spectrum utilization, collaborative radio, link reliability,

pp p p– 802.11y, White space

• DoD still trying to determine how it wants to play in CR y

advanced network structures

• Commercial implementations

p ystandards space– More involved in policy

language

starting to appear– 802.22, 802.11h,y, 802.16h– And may have been around for

a while (cordless phones witha while (cordless phones with DFS)







Summary and Future TrendsSummary and Future Trends

3GPP Will Win the Cellular Market3GPP Will Win the Cellular Market• Virtually all of the cellular incumbents

went with LTEwent with LTE• UMB was stillborn• 802.20 took too long to come out• WiMAX left with mostly greenfield

operators– ClearWire rollout seems slow– Adoption significantly underperforming

estimates• Didn’t say “LTE” will win• Didn t say LTE will win

– HSPA and LTE Advanced may squeeze out LTE

– But IPR issues may be better handled• WiMAX appears to have a future in

http://www.gsacom.com//downloads/charts/GSM_market_share_global.php4

WiMAX appears to have a future in “niche” data markets

– DoD, private radio, cable / DSL replacement

• I’m dubious of the success of 802.16m d d ll l d das an independent cellular standard

http://www.eng.chula.ac.th/files/AttachedFiles/Events/WirelessAcademicDays/3G_Broadband_and_Beyond_082108_Qualcomm.pdf

802.11 is Taking Over the World802.11 is Taking Over the World• It’s everywhere

h ff h b– Not just in the coffee shop, but in the air

– In virtually all smart phones very soon

– In Bluetooth!– The alphabet couldn’t contain it…

• I expect it to “win” everywhere it competesp– White Space (once it enters)– Home video distribution

• 802.11aa, ac, ad

• Seems like a necessity for mass• Seems like a necessity for mass‐market wireless devices

• Success is crowding the spectrum– Moving to other bands

Continuing Proliferation in Number of d dStandards

• Legacy systems not quickly fading and large # of new onesof new ones• AMPS is finally gone

– Dec 2008! ‐ Alltell– http://www.alltel.com/wps/wcm/connect/Per

sonal/home/p/customerservice/annoucements/announcements_analog_sunset_p.html/_ g_ _p

• Cellular may be finally winnowing down– Expect LTE / LTE Advanced

• Different techs solve different• Different techs solve different problems

– Fallback modes• Data over one, voice over another

– Bootstrap whitespace?Bootstrap whitespace?– Backhaul and access different

• Not new conceptually, but roles blur– EVDO for AirCell

Breeding Successful TechnologiesBreeding Successful Technologies

• Mobile WiMAX and 802.11n are MIMO standard, but so will LTEbut so will LTE– Transition of technologies can significantly extend

useful lifetime of deployments• Enhanced EDGE

WC A I O l L ’ k• WCDMA + MIMO may steal LTE’s market– 802.11n predates mobile WiMAX

• Also happening with CR applications– DFS, TPC, databases, ,– Contention Based Protocol?

• IMT‐Advanced solutions should be quite similar to LTE / 802.16e

• Standards can expect to continue to evolve even• Standards can expect to continue to evolve even post‐deployment– Need for SDR

• May make for smoother transition to 4G

110 /114

Erik Dahlman, Hannes Ekström, Anders Furuskär, Ylva Jading, Jonas Karlsson, Magnus Lundevall, Stefan Parkvall, “The 3G Long‐Term Evolution – Radio Interface Concepts and Performance Evaluation,” VTC 06

Ever Increasing Demand for SpectrumEver Increasing Demand for Spectrum

• Successful launch quickly leads to congestionE i i f l i di t ib t d• Emerging issues of overlapping distributed basestations

– Older for 802.11 access nodes– Also coexistence with other

• DSA enabled spectrum access• DSA enabled spectrum access– 802.11y, 802.11h, White Spaces– Spectrum trading (shorter time period)

• http://www.spectrumbridge.com/

• Difficult to find large contiguous bandwidthDifficult to find large contiguous bandwidth– Channel bonding / aggregation– Move up to 60 GHz?

• Flexible multiband devices to find spectrum as neededas needed

– E.g., TV, 2.4, 5, 60• Most high data rates assume large scale

MIMO– But 8x8 may be impractical

http://www.nab.org/AM/Template.cfm?Section=Small_Market_Television_Exchange1&CONTENTID=6489&TEMPLATE=/CM/ContentDisplay.cfm

But 8x8 may be impractical– Movement to higher frequencies

Take Aways (1/2)

• 4G very soon– But should be similar to LTE / 802.16eHi h d i i OFDM A A• High data rate systems migrating to OFDM + Antenna Array Processing PHY– OFDM – WiMedia, 802.11a,g, 802.16, 802.20, 802.22, UMB, LTE

OFDM + MIMO 802 11n 802 16e 802 20 UMB LTE– OFDM + MIMO – 802.11n, 802.16e, 802.20, UMB, LTE– And A LOT of antennas (8x8)

• More responsive/adaptive resource management (early cognitive radio)cognitive radio)– Multiple QoS levels – 802.11e; 802.16e; 802.20; UMB, LTE, EVDO, – Dynamic channel selection – WiMedia; 802.11h,y; 802.16h; 802.22 – Distributed sensing – 802 22Distributed sensing 802.22

• Coexistence given increasing interest– Vertical handoffs – 802.21, 802.11u– Legacy systems – 802 22 802 11h y 802 16hLegacy systems 802.22, 802.11h,y, 802.16h– Overlapping base stations

Take‐Always (2/2)• China pushing own standards

– 802 15 4c TD‐SCDMA TD‐SOFDMA802.15.4c, TD SCDMA, TD SOFDMA

• Emergence of Advanced Networking– 802.11s, 802.15.5, 802.16j, femto cells802.11s, 802.15.5, 802.16j, femto cells

• Convergence on AES for security– 802.11i, WiMedia, Mobile WiMAX, ,

• Convergence on all IP Backbone– Mobile WiMAX, UMB, LTE, ,

– Flatter access networks

– What does this mean for voice?

Cellular Data Rate ComparisonsCellular Data Rate Comparisons

http://www.rysavy.com/Articles/2007_09_Rysavy_3GAmericas.pdfestimated

Useful Websites (News, Promotional, d d )

WLANi fi

3GPP Familywww.gsmworld.com

Forums, Standards)

www.wi‐fi.orgwww.wi‐fiplanet.com/http://grouper.ieee.org/groups/802/11/

802.15bl h

www.gsmworld.comwww.umtsworld.comwww.gsacom.comwww.3gpp.orghttp://www.tdscdma‐forum.org/

3GPP2 Familywww.bluetooth.comhttps://www.bluetooth.org/www.wimedia.orghttp://www.zigbee.org/en/h // bf /

3GPP2 Familywww.cdg.orgwww.3gpp2.org

802.20http://grouper.ieee.org/groups/802/20/

802 21http://www.uwbforum.org/www.wibree.orghttp://www.multibandofdm.org/http://grouper.ieee.org/groups/802/15/

802.21http://www.ieee802.org/21/www.umatechnology.org

802.22http://grouper.ieee.org/groups/802/22/

E2R “Requirements and scenario definition ”802.16www.wimaxforum.orghttp://wimaxxed.comhttp://wimax.com

E2R Requirements and scenario definition, Available online: http://e2r.motlabs.com/Deliverables/E2R_WP4_D4.1_040725.pdf

http://grouper.ieee.org/groups/802/16/

Backup slidesBackup slides

Similar costs breakdowns across techsSimilar costs breakdowns across techs


Revenues in BillionsRevenues in Billions


P i i l f O th l FPrinciples of Orthogonal Frequency Division Multiplexing and Multiple Input

Multiple Output Communications Systems

Intersymbol InterferenceIntersymbol Interference

• Occurs when symbol period (Ts) is l h h l d l d

10-1BER Floor for various modulations

less than channel delay spread, τ• ISI introduces an error floor to BER

– Limits maximum throughput

+

x

10

BPSK QPSK OQPSK MSK

Modulation

Coherent Detection

• Solutions:– Equalization (high complexity)– Longer symbol periods (generally

l d t t ) ++x

xx

10-2

duci

ble

BER

means lower data rate)

+

+

+

x

x

10-3

Irred

+

10-210-4

10-1 100

QPSK limit

120 /114

10 10 10τT=delay spread

symbol period J. C.‐I. Chuang, "The Effects of Time Delay Spread on Portable Radio Communications Channels with Digital Modulation," IEEE JSAC, June 1987

Multicarrier communications: Longer i d dperiod, same data rate

•Concept:d l d–Divide original data stream at rate R into

L lower rate (R/L) streams on different carriers to increase symbol time

•Long history–KINEPLEX

• 1957 R.R. Mosier & R.G. Clabaugh –ANDEFT

• HF General Dynamics 60’sy–KATHRYN

• HF General Altronics 60’s•Effects

High receiver complexity B/L

( )H f


–High receiver complexity • separate receiver chain per carrier

–Bandwidth due to sidebands–Each subcarrier experiences flat fading

B/L

121 /114

Bc f

B

OFDMOFDM

• Much simpler to create multicarrier transmission using iFFTp g– Information carried in magnitude and phase of each bin

– Then can be recovered by using FFT at receiver

• Ideal inverse Fourier transform of multicarrier would be infinite Magnitude

duration sine waves– Cut at Symbol duration Ts

R t l i d i i

T0T0

– Rectangular windowing causes sinc spectrum in frequency domain with zeros at 1/TsOrthogonal s bcarriers

Frequency

122 /114

– Orthogonal subcarriers

Guard intervals and intersymbol i finterference

Guard interval Guard interval

OFDM Symbol OFDM Symbol OFDM Symbol

• If we space OFDM symbols by gaps at least as

Delay Spread Delay Spread

• If we space OFDM symbols by gaps at least as long as the delay spread, then there will be no intersymbol interferencey

• However, there will still be interference within the symbol (intrasymbol)

123 /114

y ( y )

Equalization and the DFTEqualization and the DFT

• While using longer symbol timing means OFDM can avoid i d ibl till h i t f i i b d firreducible errors, still have interfering energy in band from multipath– Received signal is the (linear) convolution of channel impulse response

with transmitted signalwith transmitted signal

• DFT Circular Convolution Theorem– Circular convolution of two discrete vectors in time domain

*y h x=

Circular convolution of two discrete vectors in time domain

– Is multiplication in the frequency domain

y x h= ∗p q y

• Implication: If we can make the system behave like a circular k k kY X H=

124 /114

p yconvolution, equalization is trivial– complex multiplication per FFT bin at the receiver

Cyclic PrefixCyclic Prefix

• Adding a cyclic prefix g y pat transmitter leads to circular convolution

• Note that misaligned timing still results in a circulara circular convolution, just time shifted

k f h– Makes for phase shifts in FFT bins

– Correct that in a t

125 /114

moment


Comments on Cyclic PrefixComments on Cyclic Prefix• We’re transmitting redundant bits

(no information transfer)J. Andrews, A. Ghosh, R. Muhamed, Fundamentals of WiMAX, Prentice Hall,

– Bandwidth penalty: L / (L + v)– Power penalty: L / (L + v)

• Penalty becomes negligible as Lb l (b h

, ,2007

becomes large (but there are tradeoffs! – more later)

• Power penalty generally more i t t i ti h

•Permits low complexity equalization for same data t important in practice where

systems are interference limited • Penalty can be avoided with zero

prefix

rates•Single carrier tap# approximately bandwidth delay product prefix

– Nothing transmitted in guard band (zero prefix)

– Receiver adds tail back to beginning

delay product–MAC

•OFDM, number subcarriers grows with bandwidth delay

126 /114

Receiver adds tail back to beginning of symbol

– Used in WiMedia

grows with bandwidth‐delay product, so

Frequency ErrorsFrequency Errors

• Primary sources ofPrimary sources of frequency errors– Doppler shift

– Clock mismatches

– Phase noise

• Effects– Reduction in amplitude ( i li i )

O. Edfors, M. Sandell, J. van de Beek D. Landström, F. Sjöberg, “An Introduction to Orthogonal Frequency Division Multiplexing,” Sep 98, Available online: (missampling sinc)

– Intercarrier interference

g q y p g, p ,http://epubl.luth.se/avslutade/0347‐0881/96‐16/esb96rc.pdf

127 /114

Effects of Frequency ErrorsEffects of Frequency Errors• Comments

– Impact greater for higher SNR l

Fading ChannelAWGN

signals– Note 5% estimation error can

lead to 5 dB effective degradation at 64‐QAM like g QSNRs

– Big frequency impact is why OFDM was originally for fixed deploymentsdeployments

• Techniques– Data aided– Non data aided

O. Edfors, M. Sandell, J. van de Beek D. Landström, F. Sjöberg, “An Introduction to Orthogonal

Frequency Division Multiplexing,” Sep 98 A il bl liNon data aided

– Cyclic prefix98, Available online:

http://epubl.luth.se/avslutade/0347‐0881/96‐16/esb96rc.pdf

128 /114

Channel EstimationChannel Estimation

• Channel assumed static for duration of symbol, though f / h b d d hfrequency/phase varying over bandwidth

• Solution, embed pilot symbols at regular intervals in the symbol– Used closest pilot– Interpolate

( )H f

129 /114

From IEEE Std 802.16-2004

Peak‐to‐Average Power RatioPeak to Average Power Ratio

• Sum of large number of ( h ) d d

P1dB,out 1dBP1dB,out 1dB

Output Power (dBm)

Output Power (dBm)

(somewhat) independent subcarriers leads to signal distribution that is somewhat Gaussian

11

11

N i FlN i Fl

Fundamen

tal

Fundamen

tal

Fundamen

tal

Gaussian• Implications

– long tails for amplitude distributionPossibly large ratios of peak to power 0

PAPR CDF for Varying # Subcarriers

MDSMDS P1dB,inP1dB,in

Noise FloorNoise Floor

Input Power (dBm)Input Power (dBm)BDRBDR

– Possibly large ratios of peak‐to‐power ratios

• Long tails can drive amplifiers into nonlinear region 10

-2

10-1

100

(a )N=16 (b) N=32 (c) N=64 (d) N=128 (e) N=256 (f) N=1024(a)

nonlinear region– Introduces harmonics and significant

out‐of‐band spectral energy

10-5

10-4

10-3

log(

CD

F)

(f)

(e)

(d) (c)

(b)

130 /1140 2 4 6 8 10 12 14 16

10-6

10

PAPR[dB]

http://www.ee.ccu.edu.tw/~wl/ofdm/class/class ppt/The PAPR problem.pps

Solution TechniquesSolution Techniques

• ClippingSpectral Effects of Windowing and Clipping

– Eliminate signals above a certain level or ratio

• Peak windowingFil k– Filter peaks

• Linear block code– Select only those codewords with

small PAPRsmall PAPR– Can also provide error correction

• Peak CancellationSubtract signals from high peaks

Peak Cancellation, Clipping, PAPR = 4dB

– Subtract signals from high peaks– Need to be similar bandwidth to

limit out‐of‐band interference• Symbol Scrambling

131 /114

Symbol Scrambling

http://www.ee.ccu.edu.tw/~wl/ofdm/class/class ppt/The PAPR problem.pps

Adaptive ModulationAdaptive Modulation

• Different subcarriers ( )H f

B/L

experience different flat fades

d ff

B/L

• Means different SINR• Adapting modulation scheme of each

Bc f

Bscheme of each subcarrier to its SINR allows the system to approach Shannon capacity

132 /114J. Andrews, A. Ghosh, R. Muhamed, Fundamentals of WiMAX, Prentice Hall, 2007

OFDMAOFDMA

• Multiple user access with • Allocation algorithmsOFDM

• Lots of flexibility possible when splitting up OFDM

– Maximum Sum Rate– Proportional fairness– Proportional rates subject to when splitting up OFDM

symbols and frames– Assign different subcarriers to

different users

p jconstraints

– Assign different time slots to different users

– Vary modulation and codingy g– Vary powers– More options available with

antenna arrays

133 /114J. Andrews, A. Ghosh, R. Muhamed, Fundamentals of WiMAX, Prentice Hall, 2007

OFDM SummaryOFDM Summary

• OFDM overcomes even severe intersymbol yinterference through the use of the IFFT and a cyclic prefix.

• Limiting factor is frequency offsetLimiting factor is frequency offset– Correctable via simple algorithm when preambles used

• Two key details of OFDM implementation are h i i d f h ksynchronization and management of the peak‐to‐

average ratio.• OFDMA provides a lot of flexibility to a system’sOFDMA provides a lot of flexibility to a system s resource allocation– Permits exploitation of multi‐user diversity

134 /114

Antenna Array Algorithms and MIMO

Antenna Array PrinciplesAntenna Array Principles•The use of multiple antennas provide two forms of diversity:

Di it i

10

B]–Diversity gain • Exploit multiple independent channels created by multipath diversity

•Works with uncorrelated antennasArray gain

0

5

g E

nvel

opes

[dB

–Array gain•Coherently combine energy from antennas•Works even with perfectly correlated antennas as received SNR increases linearly with the number of receive antennas

0 200 400 600 800 1000-10

-5

Samples

Fadi

ng

receive antennas

• Adding additional transceiver chains is expensive (SWAP and cost), but can provide tremendous (though competing) gainsp ( g p g) g– Increase the system reliability (decrease the bit or packet error rate)– Increase the achievable data rate and hence system capacity– Increase the coverage area

Decrease the required transmit power

136 /114

– Decrease the required transmit power

Receive DiversityReceive Diversity

• Oldest and simplest diversity Selection Diversity

technique• Receiver leverages independence of

fades on antennasfades on antennas– Selection Combining (SC)

• Choose antenna with maximum SINR• Lowest complexity 10

Average SNR Improvementsp y

– Equal Gain Combining (EGC)• Phase align and sum signals across antennas

6

7

8

9

, dB EGC

MRC

– Maximum Ratio Combining (MRC)• Weight signals by SINR• Best performance (system SINR is sum of

t SINR ) 2

3

4

5

6

10 lo

g (<

γ>/Γ

), d

SNR (dB)

SC

137 /114

antenna SINRs)

1 2 3 4 5 6 7 8 9 100

1

2

No. of branches, M

SCMRCEGC

Antennas

Open Loop Transmit Diversity (1/2)Open Loop Transmit Diversity (1/2)• Transmitter sends multiple signals

(possibly copies) TX RX

h1

(p y p )– These interfere at the receiver, but

if coded properly, the receiver can recover the signal

• Simplest implementation is • Receiver Alamouti Operation

TX Encoder

RX Decoderh2

S p es p e e a o sorthogonal space time block codes or Alamouti codes1

p

O SNR 2 1 Al i– Assumes flat constant channel over two symbol periods (may not be true for high mobility)

– Requires channel knowledge at i

• Output SNR 2x1 Alamouti

138 /114

receiver– No change in rate required

1. S. M. Alamouti, “A simple transmit diversity technique for wireless communications,” IEEE Journal on Selected Areas in Communications, vol 16 pp.1451–1458, Oct 1998

Open Loop Transmit Diversity (2/2)Open Loop Transmit Diversity (2/2)

• 2x2 STBC (same transmit A 4x2 Stacked Alamouti System

encoder) SINR

– Note number of h terms maximized when Nt = Nr for a

J. Andrews, A. Ghosh, R. Muhamed, Fundamentals of WiMAX, Prentice Hall, 2007maximized when Nt Nr for a

fixed number of antennas– Also full‐diversity, orthogonal

STBCs exist only for certain

Comparison of STBC and MRC

combinations of Nt and Nr.

• Can also use space‐time trellis codes for added 1‐2 dB, but

139 /114

those have exponential complexity order

Space‐Time Trellis CodingSpace Time Trellis Coding• Convolutional code applied to space and time domain• Each antenna output is mapped into modulation symbolEach antenna output is mapped into modulation symbol• Maximum likelihood sequence estimator ( Viterbi algorithm)

Example) Delay Diversity (by Wittneben [4])Encoder structure for two antennas Generator matrix form

[a1 a2 a3 a4]

QPSK mapping

140 /114

Closed Loop SystemsClosed Loop Systems•Transmit selection diversity

–Antenna(s) chosen which maximizes SINR–Equivalent to receiver selection diversity–Not as good as beamforming–Little bandwidth required–Makes most sense in in deployments withMakes most sense in in deployments with small bandwidths and small delay spreads (low range)

• Linear diversity precoding


y p g–Feedback channel state information to transmit encoder–Transmit encoder then attempts to find encoding matrix which maximizes SNR at the receiver

–Higher SNR than STBC–Higher SNR than STBC

141 /114

–Typically use some sort of codebook to reduce feedback bandwidth

Beamforming SystemsBeamforming SystemsNarrowband adaptive array or linear combiner

1 590

w1x1(t)

x (t)

0.5

1

1.5

30

6090

120

150

interfererw2x2(t)

y(t)

210 330

180 0

desiredsignal

wMxM(t)

• The weight vector is adjusted to improve the reception of some desired signal 210

240270

300

330 signalreception of some desired signal– Angle of arrival

• MUSIC, ESPRIT– Eigenbeamforming

• No physical interpretation, but useful in multipath environment

142 /114

• Minimize some cost function

• Useful for interference rejection, multipath fading mitigation, and increased antenna gain

Adaptive BeamformingAdaptive Beamforming•Narrowband beamforming is equivalent to spatial filtering

–By choosing appropriate sensor coefficients it is possibleBy choosing appropriate sensor coefficients, it is possible to steer the beam in the desired direction

–By varying the sensor coefficients (spatial filter taps) adaptively, the interference is reduced

•Wideband beamforming requires joint space‐time g q j pprocessing

–Phase shift at the antennas is frequency dependent

–Frequency‐dependent response is required (filter)•Common algorithms

–Maximum Signal to Interference and Noise Ratio (MSINR)g ( )–Minimum Mean Squared Error–Least Mean Squares–Minimum Variance Distortionless Response (MVDR)–Recursive Least Squares

143 /114

q–Similar to linear precoding, but may account for interferers

Performance ComparisonPerformance Comparison

• MRT refers to maximum ratio transmission – the choice of antenna weights that maximize 3 dBgreceived SNR

• With optimal eigenbeamformereigenbeamformer, canceling an interferer is equivalent to dropping an antenna elementan antenna element

Modified from: J. Andrews, A. Ghosh, R. Muhamed, Fundamentals of WiMAX, Prentice Hall, 2007

144 /114

Spatial MultiplexingSpatial Multiplexing• In rich scattering environments, independent data signals transmitted

from different antennas can be uniquely decoded to yield an increase infrom different antennas can be uniquely decoded to yield an increase in channel capacity

145 /114

Spatial Multiplexing TechniquesSpatial Multiplexing Techniques•Open loop (Unknown channel)

l k l h d•Closed loop (known channel)

l l–Maximum likelihood• Little additional gain, except at low SNR

–Zero‐forcing• Evaluates pseudo‐inverse of H

–Singular Value Decomposition• Computationally complex• Capacity (assuming waterfilling)

• Can dramatically increase noise power–MMSE

• Minimizes distortion• Like Zero‐forcing at high SNR, but without • For large SNR, capacity grows linearly

ith k f H i t l i {Ng g

the instability at low SNR–BLAST

• Layers & codes transmissions across antennas

with rank of H, approximately min{Nt, Nr}

–Approximations guided by• information capacity, • error probability• Effectively linear receiver with successive

interference cancellation• Receiver iterates through transmission streams using MMSE or ZF

• Works better in lab than real‐world due to

• error probability • detection MSE• received SNR

–Can tradeoff multiplexing for diversity

146 /114

Works better in lab than real world due to high SNR requirement

diversity

Relative Capacity as function of Antenna h iArray Technique

• 19 BS, 3 sectors, spaced 2.8 km, mix of users

• Proportional Fair scheduling

147 /114Source: WiMAX Forum

Correlation/Coupling EffectsCorrelation/Coupling Effects

• Spacing between antennas influence correlation p gand coupling

• Multipath components can act like interference for beamforming which reduces antenna gain

4x4, SNR = 20 dB, 30° AS Beamforming BER

148 /114

http://www.ngwnet.ac.uk/files/wspres/mimo2.thompson.pdf [Ref. D. Figueiredo, WPMC’04]

Diversity vs. BeamformingDiversity vs. BeamformingDiversity Combining

• Combine signals from differentAdaptive beamforming

• Focus the antenna’s gain in the Combine signals from different antenna elements using various algorithms

• Signal from each element is

gdirection of the desired signal– Achieved by manipulating the

weights associated with each elementprocessed separately

• Signals have to be uncorrelated for maximum performanceMiti t f di

element

• Antenna elements have to be separated by λ/2 to attain a certain phase difference in the signals

• Mitigates fading• Increases gain• Can improve polarization match

f

p g– Signals are correlated

• All advantages of diversity combining

• No interference rejection capabilities

• Has interference rejection capabilities– Typically > 20 dB

149 /114

MIMO SummaryMIMO Summary• Spatial diversity offers incredible improvements in reliability, comparable to increasing the transmit power by a factor of 10–100.

• These diversity gains can be attained with multiple receive antennas, multiple transmit antennas, or a combination of both.

• Beamforming techniques are an alternative to directly increase the desired signal energy whiledirectly increase the desired signal energy while suppressing, or nulling, interfering signals.

• In contrast to diversity and beamforming, spatial multiplexing allows multiple data streams to be simultaneously transmitted using sophisticated signal processingsignal processing.

• Since multiple‐antenna techniques require channel knowledge, the MIMO‐OFDM channel can be estimated, and this channel knowledge can be relayed to the transmitter for even larger igains.

• It is possible to switch between diversity and multiplexing modes to find a desirable reliability‐throughput operating point; multiuser MIMO strategies can be harnessed to transmit to

150 /114

multiple users simultaneously over parallel spatial channels.


Documents

Emerging Wireless Standards 09 - Cognitive Radio Technologies