emerging trends in vehicular communications -...
TRANSCRIPT
1
Emerging Trends in
Vehicular Communications
Rajeev Shorey (Ph.D)
Fellow Indian National Academy of EngineeringSenior Member IEEE
President, NIIT University, India
www.niituniversity.in
(Formerly GM Research Labs)
IEEE New York(With collaboration with IEEE Delhi, India)
June 8, 2011
Forward radar
Computing platform
Event data recorder (EDR)
Positioning system
Rear radar
Communication facility
Display
Acknowledgement
• Chair, IEEE Delhi Section
– Prof. S. K. Koul, Indian Institute of Technology, Delhi, India
• IEEE New York Section
– Dr. Amitava Dutta-Roy
2
3
Structure of the Talk
• Introduction
– Convergence in the Automotive Sector
• A peek at OnStar by GM
• Vehicular Ad Hoc Networks (VANETs)
• Standardization Efforts
– Emergence of DSRC
• Dedicated Short Range Communications
• Emerging Applications and Services
• Technical & Research Challenges in VANETs
• Conclusion
Welcome to theWorld of
‘Smart Vehicles’
4
5
Introduction & Motivation
• Vehicles are becoming smarter by the day
– Electronics, Controls, Software (ECS) is now the dominant component in vehicles !
– Advent of smart Computing & Communications
• Vehicular Ad Hoc Networks (VANETs)
– Safety Applications are the Key enablers for VANETs
– There are a plethora of challenges in VANETs
– Several OEMs need to collaborative to succeed in the highly competitive market
6
Electronics, Controls & Software in Automotive Sector
• Increasing role of Electronics and Software in the automotive sector– From 15% in 1990s to 37% in the current decade, an
exponential increase of 146%
• Automotive electronics and control systems– Key properties
• High-integrity
• Real-time
• Distributed
• Hybrid systems
– Requiring development processes with robust verification and validation
– The activities in this thrust area are centered on formal methods based design and verification of control software
7
Electronics, Controls & Software
8
Automotive Software
9
On-Board Systems (Smart Car)
10
Convergence in the Automotive Sector
Applications
Emerging Services
NewBusinessModels
&Demands
Heterogeneous
Technologies(Hardware, Software,
Middleware)
• Emergency call
• Breakdown call• Vehicle diagnostics
• Stolen vehicle tracking• Remote immobilization• Remote lock/unlock• Online services
• Safety• Infotainment
• 3G • WLANs
• ZigBee• RFID • Sensors • GPS
• XM Radio, …
Automotive
Manufacturers
11
Convergence of Technologies
Next Generation of Real Time Control, Communication and Computation for
Wireless Systems
Computation Communication
Control
Internet
Sensors and Actuators
Added Dimension
RFID Technology
13
A Peek at OnStar
Vehicle to Infrastructure Communications
14
OnStar System
V2V Communications
Enterprise
BackendOnStar Channel
Telematics
Platform
DSRC Communication
(IEEE 802.11p Standard)
Cellular Communication
15
What is OnStar (by GM)?
• Provides multiple “Telematics” related services
• Leverages Cellular Channel
• Supports– Data
– Audio
• Customers subscribe to a set of services
• Cost depends upon the number of subscribed services
16
Select OnStar Services
• Automatic Crash Response
• Automatic Air Bag Deployment Response
• Emergency Services
Vehicular Communications
V2V or C2C or VANETs
17
18
Vehicular CommunicationsApproaches
• Vehicle to Infrastructure
• Roadside Units
• WLAN technologies
• Base stations
• Cellular technology
• Vehicle to Vehicle
• DSRC standard
• In Vehicle
• ZigBee
Enabling Technologies
19
Vehicular Positioning
• Accurate autonomous geo-spatial positioning finds itself at the core of most VANET applications
– All Safety applications
• GPS: 10 – 15 m accuracy
• DGPS: approx 1 m positional accuracy
20
On-Board Computation Platforms
• Capabilities of Computation Platforms
– Processing large amounts of sensor data
– High-bandwidth communications
– Highly integrated sensor fusion filters
– Complicated path prediction and application logic
• Computation platforms in the Automotive domain pose a tradeoff
– Cost & Performance
21
22
A Modern Vehicle is a Computer on Wheels
Forward radar
Computing platform
Event data recorder (EDR)
Positioning system
Rear radar
Communication facility
Display
• Processing power: comparable with a Personal Computer + a few dozens of
specialized processors
• Communication: typically over a dedicated channel:
Dedicated Short Range Communications (DSRC)
• In the US, 75 MHz at 5.9 GHz;
• In Europe, 20 MHz requested but not yet allocated)
• Envisioned protocol: IEEE 802.11p
• Penetration will be progressive (over 2 decades or so)
(GPS)
- Human-Machine Interface
- Navigation system
23
Sensor Networks for Automotive Applications
24
Traditional Sensors in a Vehicle
• Radar
• Ultrasonic systems
• Vision and LIDAR systems
Traditional sensors have their natural limits
• They only sense the immediate vehicle environment
(short-haul)
• Mostly passive (radar has limited data capabilities)
• Relatively expensive and typically not versatile
25
Vehicular Sensors
LongRange
Sensors
Short Range
Sensors
Short-RangeBlind-Spot Sensors
Rear Vision System• Object detection• Far IR capability
EnhancedDigital Map
System
Short-RangeSensors
Long-RangeScanning
Sensor
Forward Vision System• Lane tracking• Object detection• Far IR capability
Sensor Strategy
VANETs
26
27
Vehicular Communications
Approaches
• Vehicle to Infrastructure
• Roadside Units
• WLAN technologies
• Base stations
• Cellular technology
• Vehicle to Vehicle
• DSRC standard
• In Vehicle
• ZigBee
28
VANET: Freeway Topology
29
Unique Characteristics of V2V Networks
• V2V is a special case of ad hoc network• Predictable, high mobility that can be exploited for system
optimization• Dynamic, rapidly changing topology
– Due to high mobility• Constrained
– Largely one-dimensional movement due to static roadway geometry
• Potentially large-scale• No significant power constraints
– Unlike sensor and other types of mobile networks• Limited battery life is a major concern
• Broadcasting takes precedence over Unicast routing– V2V networks are All Broadcast Networks
V2X Communications
Active Safety Applications
30
Categories of Applications
• Active Safety
– Early Applications
– Later Applications
• Congestion Notification
• Infotainment
31
VANET Applications Use Cases
• VANET communications (V2V and V2I) can be used for dozens of potential applications with highly diverse requirements
32
Most Representative VANET Applications
Assist driver with signage
• Traffic Signal/Stop Sign/Rail Crossing Violation Warning
Assist Driver at Intersections
• Left Turn Assistance
• Intersection Collision Warning
Assist Driver on Special Road Conditions
• Work Zone Warning
• Rollover Warning
• Road Condition Warning (vehicle sensor based e.g. obstacles,
unpaved road, black ice, etc.)
• Road Condition Warning (infrastructure based)
33
VANET Applications
Assist Driver in Potentially Dangerous Situations
• Forward Collision Warning
• Emergency Brake Lights
• Blind Spot Warning
• Lane Change Warning
• Wrong Way Driver Warning
• Rail Collision Warning
Assist Driver in Normal Situations
• Highway Merge Assistance
• Visibility Enhancer (through obtaining data from other cars)
• High Beam Turnoff request
• Assist Driver in Accident Situations
• Crash/breakdown Warning
• Pre-crash sensing (imminent or unavoidable collisions)
• Event Data Recording 34
35
V2X Active Safety Applications• Event reporting applications
– Generate messages only for the duration of the event
– Report events based only on information present at sending vehicle
– Examples: EEBL (Emergency Electronic Brake Lights), RCHA (Road Condition Hazard Ahead)
• Persistent applications
– Require repeated exchange of vehicle kinematics in a local neighborhood
– Predict and report events by processing exchanged information
– Examples: CCW (Cooperative Collision Warning), BSW (Blind Spot Warning)
`̀
`̀
Driver InteractionDriver InteractionDriver InteractionDriver Interaction: Applications raise advisories or warnings
to help the driver avoid accidents
36
Vehicle Safety Scenarios
Avoiding rear-end
collision
Avoiding lane
change collision
Vehicle brakes
hard
Collision mitigation
Traffic signal
Avoiding
intersection
collision
37
Vehicle Safety Scenarios
Avoiding rear-end
collision
Vehicle brakes
hard
Collision mitigation
Traffic signal
Avoiding
intersection
collision
Solution : Vehicle to vehicle/ Infrastructure / Roadside communication of information
V2V Messages
� Very Adhoc ( > 40 MPH speeds)
� Low latency � High reliability (low PER) � Authenticated & Secure� Multihop
V2X Communications
Key Challenges
38
Challenges
• Design and Development of VANET is a technically and economically challenging endeavour
• What are the Key Technical challenges?
39
Key Technical Challenges
• Inherent characteristics of the Radio channel
– VANET presents scenarios with unfavorable characteristics for developing wireless communications
• Multipath
• Fading effects
• Very high speed
40
Key Technical Challenges
• Lack of an online centralized management and coordination entity
– Totally decentralized and self-organizing network
• Fair and Efficient use of the available BW of the Wireless channel is a hard task
• Lack of an entity that is able to synchronize and manage the transmission events of different nodes
41
Key Technical Challenges
• High mobility, scalability requirements and wide variety of environmental conditions
– High mobility presents a challenge to most iterative optimization algorithms aimed at making better use of the channel bandwidth
42
Key Technical Challenges
• Security and Privacy needs and concerns
– Challenge in balancing Security and Privacy needs
– Rx want to make sure that they can trust the source of information
– The availability of such trust might contradict the privacy requirements of the sender !
43
Key Technical Challenges
• Standardization versus Flexibility
– There is a need for standardizing communications to allow VANET to work across various makes and brands of OEM
– OEMs would want to create product differentiation with their VANET IP
– These goals are somewhat in tension !
44
Key Challenges from an Application and Socio-Economic Perspective
• Analyzing and Quantifying the benefit of VANET for traffic safety and transport efficiency
• Analyzing and Quantifying the cost-benefit relationship of VANET
• Designing deployment strategies for VANET that are not based on a single infrastructure and/or service provider
• Embedding VANET in ITS architectures
– Truly cooperative systems need to be developed
45
46
PKI Design for Secure V2X Communications for Safety
47
Security Threats in V2V &V2I
Figure Source : http://ivc.epfl.ch/
48
Example Attack: Generate “intelligent collisions”
SLOW
DOWN
The way
is clear
49
Security Attributes for V2X Safety Apps
• Message Integrity and Entity Authentication
– Message has been transmitted by a genuine vehicle, and has not been tampered with in transit
• Non-repudiation
– The receiver of a message is able to prove afterwards that the sender in fact did transmit this message
• Privacy: Multiple notions of privacy
– Anonymity: Not possible to determine the identity of the vehicle from a message transmitted by the vehicle
– Unlinkability: Not possible to deduce that multiple transmissions were from the same vehicle.
• Correctness based on non-cryptographic techniques
– For detecting compromised/malfunctioning units
Design Objective: Satisfy above attributes without affecting performance
of V2X Safety Apps
• A successful authentication mechanism should fulfill several properties
– Secure Authentication
– Non-repudiation
– Denial of Service (DoS) resilience
– Support for multi-hop communication
50
Authentication
• Authenticated data ensures receivers can verify that the message received was sent by the appropriate entity and that it has not been modified in transit
• If an attacker can pose as another entity or modify another entity’s packets without being detected, the mechanism fails to provide secure authentication
51
Non-Repudiation
• Non-repudiation allows a receiver to prove to a third party that the sender is accountable for generating a message
• What happens if the broadcast mechanism lacks non-repudiation?
– A malicious party can claim another party generated the message
52
Denial of Service (Dos) Resistant
• A mechanism should require little computational or memory resources such that other OBU operations may proceed unimpaired
• Given the relatively expensive nature of digital signature verification (7 ms for ECDSA), an attacker can launch a computational DoS by flooding a receiver with invalid signatures such that the receiver wastes processing power to verify the signatures
53
Multi-Hop Authentication
• There should be a provision for Multi-Hop Authentication
• Inherent Challenges
54
55
Reference Solution:Public Key Infrastructure (PKI)
• How PKI enables nodes to talk to one another:– Asymmetric Key Cryptography: A message is signed using the
Private key of the sender and verified using the Public key of the sender.
– Certificate: A message signed by a trusted entity called the Certificate Authority (CA) that binds a principal and its public key
• How PKI evicts compromised/malfunctioning nodes from system:– Certificate Revocation List (CRL): A message signed by the CA
that lists all the revoked principals
Message Structure
Digital certificate
Message payload (m)
Digital signature on ‘m’
CACA
NodeNode
NodeNode NodeNode
NodeNode
PKI High-level Architecture
56
Design drivers for a PKI for V2X Communications for Active Safety
• Resource-constrained Platform– Participants have limited computational prowess– Limited memory and storage
• System-wide Scalability Issues– Large number of participants– Interactions are expected to be spatially localized
• Communication Aspects– Connection to Infrastructure is expected to be either
intermittent or costly
– Message transmissions are likely to be lossy and unreliable
• Interoperability– Security Architecture needs to be extensible
57
Efforts in Standardization
WAVEWireless Access in a Vehicle Environment
Peek at Various Wireless Standards
IEEE 802.15.3 UWB, Bluetooth
Wi-Media, BTSIG, MBOA
WAN
MAN
LAN
PAN ETSI HiperPAN
IEEE 802.11 Wi-Fi Alliance
ETSI-BRAN HiperLAN2
IEEE 802.16d WiMAX
ETSI HiperMAN & HIPERACCESS
IEEE 802.20IEEE 802.16e
3GPP (GPRS/UMTS)3GPP2 (1X--/CDMA2000)
GSMA, OMA
SensorsIEEE 802.15.4(Zigbee Alliance)
RFID(AutoID Center)
RANIEEE 802.22
59
Standardization Efforts
• Vehicular Infrastructure Integration (VII)
• Vehicle Safety Communications (VSC)
– Backed up by
• Crash Avoidance Metrics Partnership (CAMP)
• US Federal Highway Administration (FHWA)
• US National Highway Traffic Safety Administration (NHTSA)
60
WAVE Communications Architecture & Standards
IEEE P1609 committee for DSRC standardization� P1609.1 -- Resource Manager � P1609.2 -- Security Services for Applications and Management Messages � P1609.3 -- Network Services - Intermediate Layers � P1609.4 -- Medium Access Control (MAC) Extension Services � 802.11p -- WAVE physical and lower MAC layers
PHY (IEEE 802.11p)
MAC (IEEE 1609.4)
LLC (IEEE 802.2)
WSMP(IEEE 1609.3)
IP
UDP
General AppsSafety Apps
Other Apps
PLME
MLME
IPLME
Data Plane
WME
LLCME
WSMLME
Management Plane
1609.2
SecurityStandard
61
802.11p PHY as extension of 802.11a
Frequency (GHz)
5.8
55
5.8
60
5.8
65
5.8
70
5.8
75
5.8
80
5.8
85
5.8
90
5.8
95
5.9
00
5.9
05
5.9
10
5.9
15
5.9
20
5.9
25
Uplink
Downlink
IEEE 802.11a/RA WB - 52 carrier OFDM /w 48 data carriers, 10 MHz channels
Ch 172 Ch 174 Ch 176 Ch 178 Ch 180 Ch 184Ch 182
Optional 20 MHzOptional 20 MHz
Control
Channel
5.850 – 5.925 GHz for WAVE in NA. Licensed ITS radio service bands
OFDM with BPSK, n-QPSK & n-QAM and varied datarates
Frequency Bands
62
63
CAMP Consortium
Research & Development North America, Inc.
A Daimler Company
Vehicle Safety Communications 2
Intelligent Transportation Systems
CAMP
Research & Development North America, Inc.
A Daimler Company
Vehicle Safety Communications 2
Intelligent Transportation Systems
CAMP
64
Conclusion
• Vehicular Communications is a highly challenging area
• Slow penetration makes connectivity more difficult
• Security leads to a substantial overheads
– Must be taken into account from the beginning of the design process
• The field offers plenty of novel technical challenges
– Enabling PKI
– Scalability
– VANET Performance with multiple simultaneous applications
– Interoperability
– Infrastructure related issues
– Need for Multi-hop communications (?)
65
Business Challenges
• Telematics Platform– Low cost– Light weight– Capable of supporting heterogeneous
applications/services with “low” footprint
• Key Question– What should be the most appropriate Architecture for
the Telematics Platform?• Interfaces• Technologies?
– WiMax, WiFi, ZigBee, 4G, LTE, …
66
Conclusion: Technology
• Emergence of multi-modal distributed sensors for automotive applications
• Important Trends– Combination of Data/Audio/Video– 3D Machine Vision/Video Imaging Technologies
• Key Challenges– Low cost– Low complexity
• Management, Maintenance, Overheads
– Security
67
Concluding Remarks
• OEMs will need more and more “flexibility”
– Ever changing technologies
– Newly emerging solutions/services
• Future vehicles are likely to be “plug and play”
– At least as far as ECS is concerned
• The sector is highly sensitive to ‘cost’
– Even a $1 addition is a huge challenge in the highly competitive market !
68
Thank you