4-wave talk globecom 2009

7/27/2019 4-WAVE Talk Globecom 2009

1/75

Research and Prototype Experience on WAVE Technology

Weidong Xiang, Ph.D.

Center for Vehicular Communications and Networks at University of Michigan, Dearborn

December 2009

1


2/75

1. Backgrounds and Technical Brief

2. Overview on Research Status

3. Vehicular Network Simulator

4. Key PHY/MAC Technologies

5. WAVE Prototype

6. Further Information

2


3/75

Our Research ObjectiveWe dedicate to researching, developing and commercializingWAVE technology and systems for intelligent transportationsystems (ITS) and Internet access by integrating ourresearch, algorithms, protocols and patents.

Our Sponsors include

3


4/75

Previously known as Vehicular Ad-hoc Networks (VANET)or Dedicated Short Range Communications (DSRC) ormore recently Wireless Access in Vehicular Environments(WAVE) systems are all refer to vehicular communicationsand networks. WAVE systems is based on the IEEE 802.11pstandard, which is expected to ratified in 2009 or later.

Key Technical Merits include adopting orthogonalfrequency-division multiplexing (OFDM) modulation scheme

to achieve a data rate of 6-27Mbits/s occupying a bandwidthof 10MHz over 5.850-5.925GHz band assigned by the FCC.

Main Applicationsare ITS, high-speed communications andInternet access, safety and security enhancements.

4


5/75

Road transportation is a fundamental element ofcontemporary society.

Road transportation in the US is presently in difficult

circumstances in terms of three aspects: safety, efficiency,and information access.

The National Highway Traffic Safety Administration(NHTSA) has reported that there are 6.2 million crashesannually resulting in more than 43,000bfatalities and a costto society of more than $230 billion [1]. Half of these

fatalities occur in cases of vehicles leaving a road andpassing through intersections. In addition, injuries anddamages from non-fatal accidents lead to significant costsin terms of health care and property. For example 2.7million people were injured in motor vehicle crashes in 2005[1].

5


6/75

Due to aging and increasingly usage of the road transportationinfrastructure in US, congestion has emerged as a critical issuethat negatively impacts on our lives in multiple ways: it creates

inefficiencies in roadway use; it wastes fuel; it causes widespreadpollution and noise; and it reduces personal quality time. Forexample, traffic congestion costs Chicago $7.3 billion per year [2].The average commute increased 14% in the last ten years, from22.4 minutes in 1990 to 25.5 minutes. In 2000 [3]. In many areas ofthe country, traffic congestion has become a major quality of life

issue that impacts decisions as fundamental as where to buy ahome or where to work [3]. We are experiencing increasingcongestion on our nation's highways, railways, airports andseaports. And we're robbing our nation of productivity and ourcitizens of quality time with their families." (The U.S. Secretary ofTransportation, Mary E. Peter. [4].)

6


7/757

Fig.1 The US traffic fatalities in recent years (upper) and percentage of the States with laws supporting various highway safety efforts (lower)Fig.1 The US traffic fatalities in recent years (upper) and percentage of the States with laws supporting various highway safety efforts (lower)Fig.1 The US traffic fatalities in recent years (upper) and percentage of the States with laws supporting various highway safety efforts (lower)

Fig.1 The US traffic fatalities in recent years (upper) and percentage of the States with laws

supporting various highway safety efforts (lower)


8/758

Safety systems widely adopted by automakers are typicallybased on individual vehicle implementations, such as air bagsand anti-lock brakes.

However, beyond some substantial initial gains, these single-vehicle systems have not been able to further alleviate fatalitiesor injuries significantly. Figure 1 shows that death number hasbeen gradually increased in recent years, even though most ofthe States have laws strongly supporting for highway safety.

Road accidents are often caused by driver carelessness orignorance, simple misconduct, or lack of experience. In someinstances, dangers due to severe weather and road conditionsand consequent obstruction of view are also responsible. It isextremely difficult to eliminate these human factors due to theinherent limits of human sensing and reaction speeds. All othermeasures applied failed to reduce the number of fatalities for thelast 15 years. More people in the US give their lives intransportation related accidents than any other single cause


9/759

Information access for vehicles is presently limited to voiceservices through the use of cellular phones (includingOnstar) and global positioning systems (GPS) . Presentlyused telematics and vehicle information systems weredesigned for unique (and sole) applications and these areonly utilized in certain geographic regions (such as thetolling systems installed and operated by a lot of Statesindividually) or on certain brands of vehicles (such as Onstarfor General Motor made vehicles).


10/7510

All of theseisolatedsystems taken together are inadequateto provide high-reliability and high-rate information servicethrough vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) wireless links for future intelligent

transportation systems (ITS), as advocated by theDepartment of Transportation (DOT).

Consumers are becoming accustomed to having wirelessaccess anywhere and anytime, and roadways are noexception. Particularly for passengers in vehicles, currentservices are inadequate for many comfort or entertainmentapplications such as web browsing, emailing and dataexchanging.

A reliable, ubiquitous, and fast wireless network within ITScould remedy this, and induce other benefits such asroadway electronic commerce


11/7511

Argument I : Is the WAVE technology the problem solver ?

The goals of reducing fatalities, increasing roadway efficiency, andenhancing roadway wireless services will require multiple solutions, with

significant contributions from several disparate communities. A trueenhancement of road transportation will demand systematic and synergisticcollaborations among government agencies, automotive manufacturers,Universities, public and private companies, and communication networkoperators.

The Administrator of the Research and Innovative Technology Administration

(RITA) of the US DOT, Mr. Paul Brubaker, have challenged the nation toreduce transportation accidents related death in the US by 90% over the nextdecade, through the use of better information technologies.

The technological challenge posed by inter-vehicle networking lies in theestablish of reliable, robust and real-time wireless links for both critical safetymessage and high-speed data exchange.


12/7512

Moreover, stretched over millions of miles, with thousands trafficsignal controlled intersections, with 250 Million vehicles runningover it and diverse human driver behaviors, traffic network itselfhas presented significant a challenge as well. With the aid of road

information infrastructure for delivering real-time local and globaltraffic messages and further assuming that most drivers will beacting rationally, the traffic network itself will become morecontrollable and predictable. However, all of these challenges arenot currently well studied and addressed, leading to significant

negligence

On the great potentials not only for multiple national societalbenefits but also to stimulate and advance the nation's researchand technology frontiers.


13/75

WAVE technology is a revolution solution for vehicle safetyenhancement by providing drivers with early warning, perceive andassistance. It is an extension of humans natural sensing and realizestelesensing of vehicles. Working as probes, vehicles report timely trafficand road condition information to transportation agencies, which is

thereafter shared by a large community. The WAVE system is a major ITS initiative that can enhance the

transportation environment in the aspects of safety, management and dataservices in a fast speed with a less cost when compared to other strategies,such as expensive road infrastructure expansion.

In the United States, DOT plans to equip every vehicle with a WAVE systemand install a large number of road side units (RSUs) in the main roads andhighways to make the WAVE service available.

Information for safety enhancement and ITS. Internet access for dataexchange and entertainment.

13


14/7514

The road information infrastructure is defined as thenetworks of many local road areas networks that are

connected each other through dedicated networking or theInternet in either wired or wireless links. While inter-vehiclenetworking specifies the V2V and V2I communications withina local road areas network. Figure 2 shows an example of theconfiguration of road information infrastructure where a cell

of the local road areas network consists of many onboardunits (OBUs) and roadside units (RSUs).


15/75

15

Fig. 2 An example of the configuration of road information infrastructure based on inter-vehicle networking and some applications

Fig. 2 An example of the configuration of road information infrastructure based on inter-

vehicle networking and some applications


16/75

16

The road information infrastructure are expected to be withthe features listed below:

1. Maintaining local vehicular awareness of surroundings in a real-timemanner, through accurate and fast sensing, surveillance, andinformation sharing among vehicles.

2. Extending perception from local and transient to global and long-

term using prediction and preemptive responses.

3. Translating of situational information to appropriate actions, anddeveloping multiple and collaborative automatic vehicle safetycontrol strategies


17/75

17

Fig.3 Lane change warning with the help of WAVE and GPS sensors


18/75

18

Fig.4 Intersection collision warning


19/75

19Fig.5 Braking and hazard warning


20/75

20Fig.6 Smart traffic light control


21/75

Vehicles have been operating on roadways on this planet formore than a century in an isolated way. We are at the righttime to connect these vehicles and bring our society into anew age.

Deployment of road information infrastructure willfundamentally smooth the progress of ITS by providingroadways with high performance physical platforms forgathering operational data.

This deployment will also turn driving and riding into acompletely new experience, safer and more pleasant thanever before.

21


22/75

N0 other solution has the promise to save so many lives asresolving this challenge. Furthermore, the potentials toimprove the nation's productivity by reducing road

congestion, to increase the nation's competitiveness byincreasing the efficiency of its road transport, to positivelyand significantly influence the environment in a greenerdirection by enabling reduction of its vehicle induced crudeoil usage, as well as to enhance the nation security byenabling the usage of vehicle recording and tracking, will beregarded as significant and far going

22


23/75

The realization of a nationwide road informationinfrastructure requires intensive research, development andmanufacturing activities.

This includes algorithm development for many layers of thecommunication protocol stack, functional definition ofvarious parts of the infrastructure, subsequent systemdesign, prototype assessment and massive production.

23


24/75

ITS has the potential to generate a fresh information technology industrybased upon roadway vehicles, and this will bring the US an opportunityto grow the high-tech sector of its economy and enhance itsinternational economic competitiveness.

The magnitude and breadth of the road information infrastructureimpacts on the economy of the US are substantial, multi-layered andprofound.

American automobile manufacturers are currently encounteringunprecedented difficulties and are continuously losing their market sharein North American.

The first adoption of inter-vehicle networking will help the Americanautomobile industry regain the lost market and a leading position in thecompetitiveness. The level of impact underlines the necessity to initiatethe development of road information infrastructure.

24


25/75

As the breadth and depth of road information infrastructure is explored,new technologies will need to be developed, and this may lead to a newindustry specializing in vehicular information access.

This would create jobs and increase the base of skilled high-technologyworkers.

It may incite new scientific discoveries that yield benefits in other areas.For example, wireless communication and networking techniques

developed for high-velocity heterogeneneous vehicular networks have avital application in military vehicles and wireless networks controlledsystems

25


26/75

The massive market of WAVE systems will sustain severaltier-one automobile suppliers.

Thousands of workshops will be needed to install WAVE

devices to existing billions of vehicles. WAVE systems will foster several WAVE services operators

with similar sizes to those of existing cellular mobilecommunication providers.

WAVE systems generate a fresh information industry based

on vehicles, of which the magnitude and degree of itsimpacts on our society are substantial, multi-layered andprofound.

26


27/75

There are approximately 250 Million passenger cars currently in

use in the U.S. and about 25 Million cars (10% of the total) are

manufactured every year. The market for inter-vehicle networking

is clearly huge and growing. For example, in the view of the

original equipment manufacturer (OEM) market, massive

production of inter-vehicle networking units will generate $100

Billion revenues if every vehicle in the US is equipped with an OBU

(assuming that an OBU costs $400). The equivalent revenues will

be generated if thousands of RSUs are installed nationwide.Moreover, the aftermarket will achieve and exceed the OEM

market after inter-vehicle networking devices have been widely

deployed. The aftermarket segment includes inter-vehicle

networking service and hardware maintenance.

27


28/75

Road information infrastructure is capable of capturing andrecording the detailed log of vehicle movements thatgenerate a new resourceful means for homeland security

and public safety related issues. On the other hand, certainlythe privacy protection concerns surface and related policiesare in need to guide associated records access and usage.Government should actively involve in making a suitablepolicy for the use of road information infrastructure.

28


29/75

Where is the fund ?

The current economic situation in the US, and in many countries worldwide, is one ofthe worst in decades.The US automotive industry has been particularly hard hit.

Government support is required simply to keep the US automotive companies and

hundreds of suppliers from going bankruptcy. Thus in these times automotivecompanies will not likely invest in major technology development programs withoutgovernment support. To sustain the large scale in road information infrastructureresearch, development and deployment, government support is a must.

National Science Foundation (NSF) program are far from sufficient since the relatedresearch is a high-risk and high-reward activity that requires innovative solutions andmechanisms, high-integration and cross-layer study, holist strategy and systematic

plan, intensive development and optimization design ranging from physical layertransmission technologies, networking algorithms and protocols, mobile computing,database management, complicated system operation, driver behavior analysisbased on partial cooperation and the incompletion of road information, integratedcircuits design and prototyping

29


30/75

What does the Government/Community do ?

RITA and Joint Program Office (JPO) of the US DOT have invested about $70M for aproof of concept effort for ITS deployment that ended late in 2008 with a conclusionthat the solution sought would not provide the desired outcomes and not at the

desired conditions.

It demonstrates that there is significant interest from US DOT for the deployment ofITS nationwide but not sufficient investment to meet the challenges.

US DOT restarted its efforts with a new formulation of national vehicularinfrastructure integration (VII) program into a so-called IntelliDriveSM. program that is

chartered to form a national open platform for ITS deployment [5]. The programfocuses on enabling activities such as specifications and interface definitions.Currently, no other source of funding is available specifically for the study andadvancement of this critical and high-impact societal challenge. The return oninvestment (ROI) is too far in the future with a great risk for the venture, commercialcapital or investment from the private section.

30


31/75

Road information infrastructure facilitates the national inter-operability of ITS,which is a goal established by the Congress in the Transportation Equity Act forthe 21st Century (TEA-21).

The ITS America, a Federal Advisory Committee of DOT, has actively worked ondefining optimal service and licensing related rules. The American Society for theTesting and Materials (ASTM) Workgroup E17.51 and the IEEE 802.11p Workgroupboth work on the wireless access for vehicular environments (WAVE) standardscovering the physical, media access control (MAC), network and applicationlayers specifications.

The ITS America estimates that the deployment of road informationinfrastructure will double the roadway capacity with an estimated cost of $206billions in 20 years. This expense is regarded as very efficient when compared toother solutions.

31


32/75

What are the research topics ?

The fast varying and harsh vehicular environment as well as thecomplicated road information infrastructure bring in many challenging

research areas, which include advanced physical layer technologies,novel network configuration (mesh network and delay tolerancenetwork, DTN), effective media access control (MAC) protocols, robustnetwork algorithms and schemes, mobile computing, multimedia,database management and data mining, security, prototyping, chip

design, market and policy. Some research work has been conductedworldwide to address the above technical challenges. However, theseefforts are still in the early stage of study and development leaving a lotof technical challenges unsolved or untouched. More intensive researchand implementation activities are highly desired to remove majortechnical barriers and presenting theoretical guidance and perspective.

32


33/75

Argument II: Which is the better ?

Besides the IEEE 802.11p, there are two related IEEE 802.16e and IEEE802.20 work groups, aiming at the mobile air interface for high speed

wireless access on roads. The former uses 2-6GHz frequency rangedesigned for the mobility objectives in a low mobility. The latter adopts aband less than 3.5GHz applicable for the high-speed mobility situationswith large coverage, like high-speed trains.

In addition, the third generation cellular system (3G) and long-termevolution (LTE) also have a potential application to a vehicularenvironment

33


34/75





5. WAVE Prototype


34


35/75

We have 3+ years experience in WAVE research and leading in the

prototype development in both the research and industry,

including

Applied a patent for Doppler shift compensation for vehicle environment.

Demonstrated in various international conference intensively recently(IEEE Tridentcom, IEEE VTC, IEEE Globecom)

Edited a special issue on WAVE technology (EURASIP on Wireless

Communications and Networks Special Issue on WAVE Technology )

Hold the first international conference on WAVE 2008 in December, 2008.

Developed of a vehicular network simulator for system design, protocolevaluation and algorithm optimization based on GIS information and

measured WAVE channel model.

Completed a functional WAVE prototype and are working towards WAVEdevelop kit and WAVE chip design.

35


36/75





5. WAVE Prototype


36


37/75

A vehicular networks simulator is in need to evaluate howthe traffic and safety are enhanced by using WAVEtechnology.

The presented MAC and network protocols need to beverified based on real channel models and physical signalsimulator by considering the GIS information.

Researchers, engineers and operators are all in need for sucha vehicular network simulator

37


38/75


39/75

39

1

n

n ii

D d=

=

1

( )( 1)!

n

n nDn

n

DP D e

n

=

( 1)

2

( )( , ) 1( 1)!

inR R

s

i

RP n D e ei

=

=


40/75

40

In this task, the PI will develop a WAVE simulator for research andengineering purposes. The unique of the simulator lies on its cross-layer and systematic strategy built upon WAVE channel modelsthat will be proposed by the PI based on extensive experiments.The simulator will integrate signal format, vehicle type,

transceiver settings, TIGER database (shapefiles), terrains, roadsituations, vehicle distribution models and rational driverbehaviors. The simulator is capable of offering physical layerspecifications including signal coverage, channel fading, BER,PER, and packet latency. It also has interface to high layermodules and protocols. The developed simulator provides a

confident and truthful evaluation for high layers' algorithms,protocols and performances. A snapshot of the simulator is shownin Figure 3.


41/75

41

Figure 7 The snapshot of the WAVE simulator


42/75





5. WAVE Prototype


42


43/75

There is still in lack of a generic WAVE channel modeling inboth academia and industry.

We are working on theoretical study. We have done some field testing in summer, 2009.

43


44/75

44


45/75

45


46/75

46

d0

,t tv

,r rv

r1,1

r1,2

Tx

Rx

r1,1

r1,2

0

dt

0

1

tv t

rv t

1,a

1,b

1 1 1 1,( ) 0, ( , )a tg =

1 1 1 1,( ) 0, ( , )b tg =

1 1 1 1, ,( ) , ( , )a bg =


47/75

47

0 02 cos( ) 2 cos( )0 0 0 0( , ) ( )

t t r r j f t j f th t g t e e =

(1 )1

22 ( )

1 1 1 1 1 1 1 1

0

( , ) ( ) ( ) ( ( )) dj f t

h t g t e d

=

( 2 )2 2

2 22 ( , )

2 2 2 2 2 2 2 2 2 2 2 2

0 0

( , ) ( , ) ( , ) ( ( , )) dj f t

h t g t e d d

=

( 3)3 3

2 22 ( , )

3 3 3 3 3 3 3 3 3 3 3 3

0 0

( , ) ( , ) ( , ) ( ( , )) dj f t

h t g t e d d

=

3

0

( , ) ( , )i

i

h t h t =

=


48/75

48

1 1 01

( )10( )

ln(10)i

p

=

1 010 ( , )( , )ln(10)

i ii i i

p

=

0 20 40 60 80 100 120 140 160 180-600

-400

-200

0

200

400

600

DOD (degree)

Fd

(H

z)

Betat0=90 (degree)

Beta0=45 (degree)

Beta0=135 (dgree)


49/75

49

2

1

( ) 1 ( / )

0

d

J d d

f FS f F f F

others


50/75

50

0 200 400 600 800 1000 1200-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

Number of OFDM Symbols

NormalizedAutocorrelation

Coefficients

Previous Results

d0=1m

d0=10m

d0=100m


51/75

51

-200 0 200 400 600 800 1000 1200 1400 1600 1800

25

30

35

40

45

50

55

60

Doppler Frequency (Hz)

DopplerSpectrum(

dB)

Previous Results

d0=1m

d0=10m

d0=100m


52/75

52

1

1

1

(1 )(1 ( ))ICI sP x R T x dx

=

The power of ICI caused by Doppler frequency can be given by,


53/75





5. WAVE Prototype


53


54/75

We have completed a full function WAVE prototype based oncomputer (Matlab codes).

We are working on rebuilding the functions using FPGA We are look for collaboration on WAVE chip design

54


55/75

55

Fig.5 WAVE Transceiver Diagram


56/75

56

Fig.6 Sundance SMT8036 DSP Board


57/75

The DSP board consists of a TI C320C6416 (600MHz) based DSP module(SMT365) and a dual high speed ADC/DAC module (SMT370).

SMT 365 has 6 20Mb/s communication ports, 4MB SRAM at 133MHz and

8M ash ROM (boot code).

SMT 370 is dual channel high-speed ADC/DAC module. The modulecontains 2 14-bit ADCs sampling at up to 105MHz and dual 16-bit DACssampling at up to 400 MHz. The core of SMT 370 is Xilinx Virtex FPGAintegrating the main functions of the module.

The SMT 310Q module serves as carrier board for hosting of module SMT365 and SMT 370 in the standard PCI interface.

57


58/75

58

Fig.7 WAVE RF front ends


59/75

59

Fig. 8 Garmin Mobile 10x GPS sensor with Bluetooth interface. Theinterface protocol is NMEA 0183.


60/75

60


61/75

61


62/75

62


63/75

63


64/75

64

http://www.vehi-com.com/

WAVE 2008.WMV


65/75

65


66/75

66


67/75

67

Invention 2007: Selected to be one of eight most significant inventionsin 2007, the University of Michigan, Ann Arbor, October, 2007


68/75


69/75


70/75





5. WAVE Prototype


70


71/75

71

Digital Map

Internet Access

Traffice

Message

DSRC(WAVE)

Wireless

Interfaces

CellularUWB Wi-Fi

Wi-MaxBluetooth

Zigbee

GPS

Multimedia

Commerce

RFID

Wired

Interfaces

USBEthernet RS232AV

Road

Awareness

ETC

Broadcasting

Receiver

Safety

Messages

AM/FM/XM

DAB

DVB

Voice

Data Exchange

Public Safety

Notebook

Computer

Cellular Phone

PDA, DVD/CD,MP3 player

Wried

Interfaces

Wiireless

Interfaces

Portable Devices

Wi-Fi

Wi-MaxIn Fields

infrastructures at

home, offices and

hotspots

On board

Access Point

On BoardUsers

On BoardUser

On BoardUser

VII System

Instrument Panel

Camera


72/75

72

We coined a popular name of Wi-Vi to represent the intra-

vehicle infrastructure adopting ultra-wide band (UWB) radio,

which is proposed to provide onboard passengers with high-

speed wireless access to the Internet and Entertainment andthus turns the riding into a completely new experience.


73/75

73

The requirements of intra-vehicles communications: High-speed, >1Gbits/s, for multimedia and Internet accesses Duplex communications networks Support real-time safety and traffic message and multimedia

Product specifications Low cost Massive productivity Easy to installation, maintenance, and update. Flexibility of reconfiguration Scalable for various vehicles and environments (temperature, high

electromagnetic noise and interferences, and vibrations)

Extended Applications Airplane and ships Military vehicles


74/75

74

Vehicular UWB channel modeling

Fig.3 The illustration of the UWB channel measurementexperiment within commercial vehicles

0 200 400 600 800 1000 12000

0.05

0.1

PDF

UWB RMS Delay Distributions for Scenarios 1 to 5 (from top to bottom)

200 220 240 260 280 300 3200

0.1

0.2

PDF

100 120 140 160 180 200 2200

0.05

0.1

PDF

100 120 140 160 180 200 2200

0.05

0.1

P

DF

0 100 200 300 400 500 600 700 800 9000

0.1

0.2

Time (ns)

PDF

-5 0 5 10 15 2010

-5

10-4

10-3

10-2

10-1

100

SNR dB

BER

BER vs SNR

AWGN

UWB Scenario 1

UWB Scenario 2

UWB Scenario 3

UWB Scenario 4

UWB Scenario 5


75/75

Acquisition of Instruments for the Research of Applying UltraWide Band Based Wireless Networks to Vehicles for

Communications and Controls, NSF MRI , 2008-2011.

Published the first in-vehicle channel modeling paper (IEEEJournal of Selected Areas in Communications )

Redesign multiple-band (MB) OFDM Alliance (MBOA) signalformat to realize greater than 1Gb/s data rate by adoptingmultiple input and multiple out (MIMO) technology.

4-wave talk globecom 2009

Documents