IEEE INTERNET OF THINGS JOURNAL, VOL. 1, NO.

4, AUGUST 2014 289

Connected Vehicles: Solutions and Challenges

Ning Lu, Student Member, IEEE, Nan Cheng, Student Member, IEEE, Ning Zhang, Student Member, IEEE,
Xuemin Shen, Fellow, IEEE, and Jon W. Mark, Life Fellow, IEEE

Abstract—Providing various wireless connectivities for vehicles vehicles in large cities, which is responsible for traffic congestion
enables the communication between vehicles and their internal and and the consequences in terms of huge economic cost and
external environments. Such a connected vehicle solution is ex- environmental problems. It is reported that the cost of extra
pected to be the next frontier for automotive revolution and the key
to the evolution to next generation intelligent transportation systems travel time and fuel due to congestion in 498 U.S. urban areas
(ITSs). Moreover, connected vehicles are also the building blocks of was already USD 121 billion in 2011, and produced
emerging Internet of Vehicles (IoV). Extensive research activities during congestion was 56 billion pounds, compared to USD
and numerous industrial initiatives have paved the way for the 24 billion and 10 billion pounds in 1982, respectively [2].
coming era of connected vehicles. In this paper, we focus on wireless Connected vehicle solutions are very promising to alleviate
technologies and potential challenges to provide vehicle-to-x con-
nectivity. In particular, we discuss the challenges and review the traffic congestions via intelligent traffic control and management
state-of-the-art wireless solutions for vehicle-to-sensor, vehicle- [3], as well as to improve the road safety via on-board advanced
to-vehicle, vehicle-to-Internet, and vehicle-to-road infrastructure warning and driving assistance systems [4]. The second one is the
connectivities. We also identify future research issues for building ever-increasing mobile data demand of users on road. In recent
connected vehicles. years, the demand for high-speed mobile Internet services has
Index Terms—Connected vehicles, intelligent transportation increased dramatically. People in their own cars expect to have
systems (ITSs), Internet of Vehicles (IoV), intra-vehicle wireless the same connectivity as they have at home and at work.
sensor networks, vehicular networks. Connecting vehicles to the Internet can be envisioned not only
to meet the mobile data demand [5], but also enrich safety-related
I. INTRODUCTION applications, such as online diagnosis [6] and intelligent anti-
theft and tracking [7], in which the servers can be on the Internet
A S AN indispensable part of modern life, motor vehicles
have continued to evolve since they were invented in the
Second Industrial Revolution. Nowadays, people expect more
cloud. Internet-integrated vehicles have hit the road, and it is
predicted that the percentage of Internet-integrated vehicle ser-
than vehicle quality and reliability. With the rapid development vices will jump from 10% today to 90% by 2020 [8]. In addition,
of information and communication technologies (ICT), equip- government mandate has put the connected vehicle revolution on
ping automobiles with wireless communication capabilities is the fast track. The European Commission proposed to implement
expected to be the next frontier for automotive revolution. a mandatory “eCall” system in cars from 2015, by which cars
Connected vehicles on the go are proactive, cooperative, well- can automatically establish a telephone link for emergency
informed, and coordinated, and will pave the way for supporting services in case of a collision [9]. Not surprisingly, the U.S.
various applications for road safety (e.g., collision detection, lane Department of Transportation’s (DOT) National Highway Traf-
change warning, and cooperative merging), smart and green fic Safety Administration (NHTSA) recently announced that it
transportation (e.g., traffic signal control, intelligent traffic will start taking steps to enable communications between light
scheduling, and fleet management), location-dependent services vehicles [10].
(e.g., point of interest and route optimization), and in-vehicle Connected vehicles refer to the wireless connectivity-enabled
Internet access. The market of connected vehicles is booming, vehicles that can communicate with their internal and external
and according to a recent business report, the global market is environments, i.e., supporting the interactions of vehicle-to-
expected to reach USD 131.9 billion by 2019 [1]. Academia and sensor on-board (V2S), vehicle-to-vehicle (V2V), vehicle-to-
the automotive industry are responding promptly by exploring road infrastructure (V2R), and vehicle-to-Internet (V2I), as
reliable and efficient connectivity solutions. shown in Fig. 1. These interactions, establishing a multiple
There are two immediate driving forces of bringing wireless levels of data pipeline to in-vehicle information systems, en-
connectivity to vehicles. The first one is the urgent need to hance the situational awareness of vehicles and provide motorist/
improve efficiency and safety of road transportation systems. passengers with an information-rich travel environment. Further,
Growing urbanization yields an increasing population of connected vehicles are considered as the building blocks of the
emerging Internet of Vehicles (IoV), a dynamic mobile commu-
Manuscript received April 06, 2014; accepted May 16, 2014. Date of nication system that features gathering, sharing, processing,
publication May 30, 2014; date of current version August 01, 2014. computing, and secure release of information and enables the
The authors are with the Department of Electrical and Computer Engineering, evolution to next generation intelligent transportation systems
University of Waterloo, Waterloo, ON N2L 3G1, Canada (e-mail: n7lu@
uwaterloo.ca; [email protected]; [email protected]; sshen@
(ITSs) [11]. The development and deployment of fully connected
uwaterloo.ca; [email protected]). vehicles requires a combination of various off-the-shelf and
Color versions of one or more of the figures in this paper are available online at emerging technologies, and great uncertainty remains as to the
http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/JIOT.2014.2327587 feasibility of each technology. In this paper, we focus on the

2327-4662 © 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
290 IEEE INTERNET OF THINGS JOURNAL, VOL. 1, NO. 4, AUGUST 2014

Fig. 1. Overview of connected vehicles.

wireless technologies and present an overview of industrial the vehicle mass [12]. Moreover, the installation and mainte-
and academic advances for establishing vehicle-to-x (V2X) nance of aftermarket sensors (providing add-on functions) are
connectivity. The pros and cons of each option are presented inconvenient using cable connection. Recent advances in wire-
to demonstrate its feasibility. We also discuss the potential less sensor communication and networking technologies have
challenges and identify research issues in building vehicular paved the way for an intriguing alternative, where ECU and
connectivity. Our goal is to further promote the importance and sensors are composed of an intra-vehicle wireless sensor net-
research activities in the field of connected vehicles. The litera- work, leading to a significant reduction of deployment cost and
ture [12] and [13] serve as similar goal, but with different focus. complexity. There exist multiple candidate wireless technologies
The remainder of the paper is organized as follows. Section II to build intra-vehicle wireless sensor networks, and the feasibili-
discusses the challenges and existing/potential solutions to intra- ty of different wireless options to in-vehicle environments has
vehicle connectivity. Section III focuses on inter-vehicle con- been a research focus.
nectivity. The V2I and V2R connectivity solutions are presented
in Sections IV and V, respectively. Section VI discusses further A. Characteristics and Challenges
challenges and provides concluding remarks. Different from generic wireless sensor networks, intra-vehicle
wireless sensor networks show unique characteristics that pro-
vide the space for optimization.
II. INTRA-VEHICLE CONNECTIVITY
1) Sensors are stationary so that the network topology does
With increasing intelligence, modern vehicles are equipped not change over time.
with more and more sensors, such as sensors for detecting road 2) Sensors are typically connected to ECU through one hop,
conditions and driver’s fatigue, sensors for monitoring tire which yields a simple star-topology.
pressure and water temperature in the cooling system, and 3) There is no energy constraint for sensors having wired
advanced sensors for autonomous control. The number of sen- connection to the vehicle power system.
sors is forecasted to reach as many as 200 per vehicle by 2020 In spite of favorable factors, the design and deployment of
[14]. Such a big quantity of sensing elements are required to intra-vehicle wireless sensor networks are still challenging.
report event-driven or time-driven messages to the electrical 1) The intra-vehicle communication environment is harsh due
control units (ECU) [15] and receive feedback if necessary. To to severe scattering in a very limited space and often none-line-
do so, an intra-vehicle communication network should be care- of-sight. This is the major reason for extensive effort to charac-
fully designed. Wired solutions [16], [17], such as controller area terize the intra-vehicle wireless channels [18], [19]. 2) Data
network (CAN) protocol, FlexRay, and TTEthernet, require transmissions require low latency and high reliability to satisfy
cable connections between ECU and sensors. Cables and other the stringent requirement of real-time intra-vehicle control
accessories nowadays can add significant weight (up to 50 kg) to system. 3) Interference from neighboring vehicles in a highly
LU et al.: CONNECTED VEHICLES: SOLUTIONS AND CHALLENGES 291

densed urban scenario may not be negligible. 4) Security is cost and no power supply to RFID tags. Moreover, the
critical to protect the in-vehicle network and control system from experiments show promising results in terms of the coherence
malicious attacks [20]. In face of these challenges, the intra- bandwidth and the transmission reliability. These studies also
vehicle wireless sensor network has become a research focus in identify two major challenges: 1) connection outage due to large
the area of intelligent vehicle systems, and the following wireless power loss at some locations and 2) difficulty to guarantee the
technologies have been investigated extensively in the literature. critical data transmission due to collisions among simultaneous
transmissions. The suggested solution to address challenge 1)is
to use advanced antennas or the active RFID technology [33]
B. State-of-the-Art Alternatives
(additional wires to vehicle battery are required); for the second
1) Bluetooth: Bluetooth is a short-range wireless technology challenge, efficient and reliable MAC protocols [34] should be
based on the IEEE 802.15.1 standard and operating in the in- developed for the RFID sensor network.
dustrial, scientific, and medical (ISM) frequency band (2.4 GHz). 4) Ultra-Wideband: Ultra-wideband (UWB) refers to radio
It allows the communication between portable devices at a data technology that operates in the 3.1–10.6 GHz frequency band
rate up to 3 Mb/s, and is highly commercialized for consumer (an astonishing bandwidth of 7.5 GHz) and can support short
electronics [21]. The Bluetooth devices are common in current range communications at a data rate up to 480 Mb/s and at a very
automobiles, such as the Bluetooth headset and rearview mirror. low energy level [12]. UWB systems have a number of unique
However, the Bluetooth transmission requires a high power level advantages, such as resistance to severe wireless channel fading
so that it might not be viable for battery-driven sensors in and shadowing, high time domain resolution suitable for
vehicles [12]. Moreover, due to the poor scalability, a localization and tracking applications, low cost, and low
Bluetooth network can only support eight active devices processing complexity [35]. UWB has been adopted as a key
(seven slave devices and one master device) [22]. physical (PHY) layer technology in the ECMA-368 standard
2) ZigBee: One option for enabling the V2S connectivity is specified by the WiMedia Alliance [26]. For intra-vehicle
through the use of ZigBee technology, which is based on the scenarios, extensive research [36]–[44] has demonstrated the
IEEE 802.15.4 standard and operates on the ISM radio spectrum feasibility of UWB technology for satisfying the stringent
(868 MHz, 915 MHz, and 2.4 GHz) [28]. As the first attempt to reliability and energy requirement of on-board sensor
evaluate ZigBee performance in an in-vehicle environment, networks. The design and implementation of an intra-vehicle
research in [29] reports the results of packet transmission UWB communication testbed is reported in [36]. The testbed is
experiments using ZigBee sensor nodes within a car under built to transmit automotive speed data from four wheel speed
various scenarios (considering different sensor locations and sensors to the ECU. The measurement result shows a high data
ON/OFF status of the vehicle engine). This study demonstrates that delivery reliability. Most of the existing works [37]–[43] aim at
ZigBee is a viable and promising solution for implementing an proposing appropriate channel models which can capture the
intra-vehicle wireless sensor network. ZigBee is low-cost and propagation characteristics of in-vehicle environments. These
can provide an acceptable data rate (250 kb/s in 2.4 GHz studies focus on different parts of the vehicle, including
frequency band). As stated in [29], however, the challenge of passenger compartment [37], [41], engine compartment [38],
implementing ZigBee sensors is to combat the engine noise and [40], trunk [39], and locations beneath the chassis [40], [42],
interference from Bluetooth devices. In [30], the performance of [43], as the channel statistics are quite different from location to
ZigBee intra-vehicle sensor networks is thoroughly studied in the location. Recent measurement study [43] also models small-
presence of Bluetooth interference, based on a realistic channel scale fading within the vehicle which has not been considered
model. Data latency of in-vehicle sensor applications is an before. Given an underlying wireless model, how to define the
important network design consideration. To meet the hard most suitable transmission techniques at PHY layer is a critical
latency requirements, [31] derives necessary design para- issue for UWB-based intra-vehicle sensor networks. [44] is the
meters of medium access control (MAC) protocol based on a first attempt to investigate optimal power control, rate adaptation,
star topology. A thorough analysis of transmission latency of and scheduling in this regard.
using ZigBee is conducted in [15]. In addition, engineering 5) 60 GHz Millimeter Wave: Communications at 60 GHz
issues, such as the interactions with the existing CAN band, often referred to as millimeter-wave (mmWave) com-
backbone, are also discussed in [15]. munications, pave another way for building intra-vehicle
3) Radio-Frequency Identification: The feasibility of using the connectivity. Operating in the frequency band between 57 and
radio-frequency identification (RFID) technology for building 64 GHz mmWave communications can support multi-Gb/s
intra-vehicle sensor networks is studied in [25] and [32]. The wireless connections in a short range for bandwidth-intensive
rationale of the considered passive RFID solution is that each multimedia applications [45]. mmWave-based PHY layer has
sensor is equipped with an RFID tag and a reader connected to the been specified in the IEEE 802.15.3c [46] and the IEEE 802.11ad
ECU, which periodically retrieves the sensed data by sending an [47]. There has been increasing research interest in applying
energizing pulse to each tag. Extensive experiments have been mmWave communication technology to multimedia transmis-
conducted in these two studies for understanding the capabilities sion in the intra-vehicle environment, such as high definition
and limitations of RFID technology, including the wireless video transmission for seat-monitor in the vehicle. As the
channel characteristics between the reader and RFID tags at propagation loss becomes more serious in 60 GHz frequency
different locations, packet reception rate, and maximum packet band, the first and foremost is to gain a better understanding of
delay. The passive RFID solution has obvious advantages: low propagation characteristics inside the vehicle. Propagation
292 IEEE INTERNET OF THINGS JOURNAL, VOL. 1, NO. 4, AUGUST 2014

TABLE I
SUMMARY OF FEATURES OF EXISTING ALTERNATIVES

measurement campaigns have been performed in [48]–[51], to unified channel model that can be applied for all scenarios (e.g.,
investigate the small-scale parameters [48]–[50], the large-scale urban, rural, and highway), and the existing channel models, only
parameters [50], and the impact of passenger and antenna for a specific scenario, have their own merits and deficiencies.
locations [49], [51]. The results show that mmWave is a Reference [56] also provides suggestions for V2V communica-
promising wireless solution for intra-vehicle multimedia trans- tion systems based on the channel characterization. For example,
missions. A summary of features of existing V2S solutions is the adoption of multiple antennas would enhance the communi-
shown in Table I. cation reliability.
From a network perspective, compared to typical low-velocity
nomadic mobile communication systems, VANETs also present
III. INTER-VEHICLE CONNECTIVITY unique characteristics that have a significant impact on inter-
It is widely believed that the advances of inter-vehicle vehicle connectivity.
communications will reshape the future of road transportation 1) The network topology changes frequently and very fast
systems, where inter-connected vehicles are no longer information- due to high vehicle mobility and different movement
isolated islands. By means of inter-vehicle communications or trajectory of each vehicle.
V2V communications, information generated by the vehicle- 2) Due to the high dynamics of network topology and limited
borne computer, control system, on-board sensors, or passengers range of V2V communication, frequent network partition-
can be effectively disseminated among vehicles in proximity, or ing can occur, resulting in data flow disconnections.
to vehicles multiple hops away in a vehicular ad hoc network 3) Surrounding obstacles (e.g., buildings and trucks) can lead
(VANET). Without the assistance of any built infrastructure, a to an intermittent link to a mobile vehicle.
variety of active road safety applications (e.g., collision detec- In addition to the technical challenges, the following features
tion, lane changing warning, and cooperative merging) [52] and can benefit inter-vehicle communications: 1) the vehicle mobili-
infotainment applications (e.g., interactive gaming, and file ty is map-restricted and can be predicted in a certain time interval
and other valuable information sharing) [53] are enabled by to a certain degree; 2) there is no power constraint on commu-
inter-vehicle wireless links. nications and each vehicle can have relatively powerful proces-
sing capability; and 3) with the aid of global positioning system
A. Characteristics and Challenges (GPS), vehicles can locate themselves with an error up to a
few meters.
VANETs have attracted extensive research attentions for
The academia, industry, and government institutions have
many years, and how to establish efficient and reliable wireless
initiated numerous activities in the area of VANETs. An over-
links between vehicles is a major research focus.
view of the current and past major related projects in the USA,
The most cumbersome challenge is to combat the harsh
Japan, and Europe can be seen in [57]. The standards and
communication environment. In urban scenarios, the line-of-
standardization process of VANETs are given in [58] and
sight (LOS) path of V2V communication is often blocked by
[59]. There have been significant research efforts in the past
buildings at intersections. While on a highway, the trucks on a
decade for the development of VANETs, including comprehen-
communication path may introduce significant signal attenuation
sive surveys (e.g., [57] and [60]). Within the scope of this paper,
and packet loss [54]. Field tests in [55] demonstrate that multi-
we only review the options of wireless technology for enabling
path fading, shadowing, and Doppler effects due to high vehicle
inter-vehicle communications.
mobility and the complex urban environment will lead to severe
wireless loss, and with a large scale of vehicles transmitting
B. DSRC/WAVE
simultaneously, the mutual interference plays an important role
as well. Accurate modeling of the propagation environment is Dedicated short-range communications (DSRC) is a key
premier to design reliable V2V communication systems. Refer- enabling wireless technology for both V2V and V2R commu-
ence [56] presents an overview of the state-of-the-art vehicular nications. The U.S. Federal Communications Commission
channel measurements. It is noteworthy that there is a lack of (FCC) has allocated 75 MHz bandwidth at 5.9 GHz spectrum
LU et al.: CONNECTED VEHICLES: SOLUTIONS AND CHALLENGES 293

high collision probability of the broadcasted packets. For unicast

communications using DSRC MAC, the collision probability of
two or more transmissions is reduced by the adoption of a two-
way handshaking mechanism, i.e., request-to-send/clear-to-send
(RTS/CTS), before the actual data is transmitted. However, the
RTS/CTS mechanism and the acknowledgment from data
recipients (ACK) are not implemented for broadcast services.
As time division multiple access (TDMA) is capable of
controlling channel access more precisely, by which the
vehicle only needs to listen and broadcast during the acquired
time slot, many alternatives have been proposed to guarantee the
Fig. 2. Example of safety applications based on V2V communications.
quality of service (QoS) of safety and other real-time applications
in highly densed vehicular scenarios based on TDMA, such as
band for DSRC. The dedicated bandwidth is further divided into
the MAC protocols proposed in [66]–[69].
seven channels to support safety and nonsafety services simul-
taneously. The specifications of DSRC are in the IEEE Standard
C. Dynamic Spectrum Access
for Wireless Access in Vehicular Environments (WAVE), in-
cluding the IEEE 802.11p for PHY and MAC layers and the In spite of the DSRC spectrum, V2V communications still face
IEEE 1609 family for upper layers. Many automotive and ICT the problem of spectrum scarcity due to the following reasons:
manufacturers, academia, and governments have responded 1) the ever-increasing infotainment applications, such as high-
positively and are actively working in collaboration to bring quality video streaming, require a large amount of spectrum
this promising technology to fruition. There have been extensive resource, and thereby the QoS is difficult to satisfy merely by the
research efforts from academia to characterize communication dedicated bandwidth and 2) in urban environments, the spectrum
properties of DSRC [55], [61]–[65], and to enhance DSRC scarcity is more severe due to high vehicle density, especially in
performance both in the PHY and MAC layers [66]–[69]. some places where the vehicle density is much higher than
1) PHY Layer: DSRC PHY layer adopts almost the same normal [71], [72]. Numerical study [73] has reported the limita-
orthogonal frequency division multiplexing (OFDM) modu- tion of the dedicated spectrum in supporting the increasing
lation as the IEEE 802.11a/g standard and is able to support a demand of V2V applications.
data rate of 3–27 Mb/s on a 10 MHz channel [58]. Through Due to recent advances in cognitive radio, dynamic spectrum
simulation [61], empirical study [62], and measurement access (DSA) is becoming a possible complementary technology
campaigns [55], [63], the performance of DSRC PHY layer for DSRC. DSA allows vehicles to communicate opportunisti-
has been well understood. Although several research works have cally on spatially and/or temporally vacant licensed spectrum for
demonstrated that DSRC PHY layer is adequate to support safety other communication systems [74]. The feasibility of using TV
message delivery, many challenges remain, such as: 1) reliable white space for V2V communications has been demonstrated in
communication is not guaranteed especially when the LOS path [70], [75], and [76] recently. TV white space refers to unused
is obstructed or the delay spread of wireless channel is too large television spectrum between 54 and 698 MHz, and provides
[64]; 2) cross-channel interference introduces performance superior propagation and building penetration compared to the
penalty when two adjacent channels are operated simultaneously DSRC frequency band [77]. It is noteworthy that the IEEE has
[58]; and 3) a gray-zone phenomenon is particularly observed in standardized the IEEE 802.11af [78] and the IEEE 802.22
[63], i.e., the behavior of intermittent loss rate during the [79] based on DSA on TV white space for wireless local area
transmission. To well fit the vehicular environment, DSRC network (WLAN) and wireless regional area network (WRAN),
PHY layer is required to keep evolving. More challenges in respectively.
this evolution are discussed in [64], such as the difficulty in The first measurement campaign of V2V cognitive commu-
estimating the channel condition accurately. Some guidelines for nication between two moving vehicles over TV white space is
OFDM system design in the DSRC PHY layer are also given in reported in [75], followed by [76] and [70] where multihop inter-
[56], such as a suggestion of a modified pilot pattern to reduce vehicle communications are considered. The implemented DSA
receiver complexity. system consists of three subsystems: 1) the two-layer control
2) MAC Layer: Dissemination of safety messages, either time- channel subsystem; 2) the multihop data communication sub-
driven (periodic) or event-driven (as shown in Fig. 2), is mostly system; and 3) the spectrum sensing and channel switching
based on one-hop broadcast, i.e., distributing the same safety subsystem. The system parameters compared to DSRC system
message to all the nodes within the communication range, and are given in Table II. In terms of the availability of TV spectrum,
requires low latency and high reliability, e.g., the dissemination a general geo-location database approach is proposed in [80] to
of emergency braking message. However, based on the legacy create a spectral map of available channels in a given geographi-
IEEE 802.11 distributed coordination function (DCF), the cal area. Preliminary results from [70], [75], [76] and [80] have
current version of DSRC MAC is contention-based and demonstrated a great potential of DSA over TV white space to
thereby does not support efficient and reliable broadcast solve the expected spectrum shortage, e.g., to meet the commu-
services. Specifically, the poor performance of the DSRC nication need for a platoon of vehicles on a highway. This is the
MAC in supporting safety applications is mainly due to the first step toward a heterogeneous spectrum for inter-vehicle
294 IEEE INTERNET OF THINGS JOURNAL, VOL. 1, NO. 4, AUGUST 2014

TABLE II smart phones. MirrorLink-enabled vehicle infotainment systems

COMPARISON OF DSRC AND DSA [70] SYSTEM PARAMETERS
are already in the market, such as Toyota Touch 2 [86]. For
iPhone users, especially, Apple recently released CarPlay [87] as
a standard of tethering iPhone to cars. In addition to smart
phones, Internet connectivity can also be brought in the
vehicle by aftermarket devices, such as GM OnStar [88].
OnStar uses a built-in cell phone to communicate with the
outside world, and provides subscription-based services, such
as voice calls, emergency services, and Internet access. The
advantage of aftermarket devices is that the vehicle does not need
to have a pre-embedded infotainment system.
2) Built-in Connectivity: Built-in option integrates cellular
communications. Technical challenges remain, such as the service in the on-board infotainment system. The Internet
design of efficient MAC protocol with QoS provisioning for connection relies on the built-in cellular module, rather than
both safety and infotainment applications. smart phones of motorist/passengers. For example, through
built-in cellular communications, BMW ConnectedDrive [89]
combines various elements from online applications, driver
IV. V2I CONNECTIVITY assistance, call center services, and solutions to providing
Internet connectivity is becoming a must-have feature of Internet connection for in-vehicle mobile devices. Audi connect
modern vehicles. The industry has responded promptly using [90] is another example of built-in solution. Especially, using the
off-the-shelf technologies, aiming to build a huge mass market of state-of-the-art cellular technology, the concept of LTE connected
Internet-connected cars, whereas the academia focuses on the car is emerging, which is to connect the cars to the Internet through
development of optimal solutions to enable connections between the 4G-LTE cellular network. Compared to the 3G cellular service,
vehicles and the Internet. Wireless access technologies play a LTE can offer ultra-high speed, high-bandwidth connectivity, and
vital role in delivering the Internet services to vehicle users. lower cost. NG connected program is conceived and founded
Cellular and WiFi are two promising candidates. The cellular by Alcatel-Lucent to lead the research and development of
networks, such as 3G and 4G-LTE, can provide reliable and LTE-connected cars [91]. Verizon Wireless recently has also
ubiquitous access services. The feasibility of using low-cost aggressively pursued its LTE-connected-car strategy to power
roadside WiFi access point (AP) for outdoor Internet access at future automotive telematics applications. The best way to enable
vehicular mobility has also been demonstrated in [81]. We first Internet connectivity in vehicles is still in debate. Built-in options
review the up-to-date status of industrial progress, and then could provide motorist/passengers with stronger connections and
discuss research issues toward efficient and robust V2I customized services compared to brought-in options. The
connectivity. limitation is that the cellular connectivity cannot evolve once it
is embedded. A summary of a few cellular-based industrial
A. Industrial Solutions solutions is shown in Table III.
Earliest concepts of Internet-enabled vehicles were proposed
B. Drive-Thru Internet
in 1990s in the literature, such as “the Internet multimedia on
wheels” [82], “web on wheels” [83], and “the network vehicle” As a popular wireless broadband access technology, WiFi,
[84]. Nowadays, Internet-integrated vehicle is no longer operating on the unlicensed spectrum, offers the “last-hundred-
conceptual due to numerous initiatives in the automotive, tele- meter” backhaul connectivity to private or public Internet users.
communications, and consumer electronics industry. Existing With millions of hotspots deployed all over the world, WiFi can
solutions connect vehicles to the Internet through widely be a complementary solution to vehicular Internet access with
deployed cellular network infrastructure, and can be divided low cost. Recent research has demonstrated the feasibility of
into two categories, i.e., brought-in and built-in, advocated by WiFi for outdoor Internet access at vehicular mobility [81]. The
different automobile manufacturers. built-in WiFi radio or WiFi-enabled mobile devices in the vehicle
1) Brought-in Connectivity: The brought-in option caters to can access the Internet when the vehicle is moving in the
3G/4G mobile users who prefer tethering their own smart phone coverage of WiFi hotspots, which is often referred to as the
to the car. The most popular tethering technology, namely drive-thru Internet [92], as shown in Fig. 3. This kind of WiFi
MirrorLink, is powered by car connectivity consortium (CCC) access is deployable to offer a low-cost data pipe for vehicle
[85], an organization calling leading automobile (e.g., users, and recent advances in Passpoint/Hotspot 2.0 make WiFi
Volkswagen and Toyota) and ICT manufacturers (e.g., Sony more competitive to provide secure connectivity and seamless
and Nokia) together to create a phone-centric car connectivity roaming [97]. Moreover, with the increasing deployment of the
solution. Using MirrorLink, a device interoperability standard in urban-scale WiFi network (e.g., Google WiFi in the city of
essence, the motorist/passengers in a vehicle can connect the Mountain View), there would be a rapid growth in drive-thru
phone to the vehicle infotainment system via wires (e.g., USB) or Internet connectivity.
wirelessly (e.g., WiFi or Bluetooth), so that the vehicle gains 1) Characteristics and Challenges: Unlike an indoor WiFi
immediate access to the Internet and some duplicate functions of network, which only serves stationary or slow-moving users,
LU et al.: CONNECTED VEHICLES: SOLUTIONS AND CHALLENGES 295

TABLE III
SUMMARY OF INDUSTRIAL SOLUTIONS

thoroughly discussed in [95]. Experiments in [81] and [96]

are conducted for large-scale urban scenarios with multiple
vehicles. The data sets used in the experiments are collected
from the city of Boston with in situ open WiFi APs. It is shown
that in [81] with a fixed 1 Mb/s data rate, vehicles can gain a
median upload throughput of 240 kb/s and a median one-drive-
thru uploaded data volume of 216 kB. In addition, the average
Fig. 3. Illustration of drive-thru Internet.
connection and inter-connection time are shown to be 13 and
75 s, respectively. The experiment in [96] shows a long-term
unique characteristics of drive-thru Internet impose many throughput of 86 kb/s averaged over both connection and inter-
challenges on reliable and robust Internet access. First of all, connection periods. A summary of real-world measurement
high vehicle mobility yields a very short connection time to the results is shown in Table IV.
WiFi AP, e.g., only several tens of seconds, which greatly limits 3) Improvement Strategy: Due to high vehicle speeds,
the amount of data transferred in one connection. For example, intermittent links, and potential severe channel contentions,
the overall connectivity range to a roadside AP is around the throughput of per drive-thru is limited as observed in real-
500–600 m, corresponding to a connection time of 18–21 s world tests. It fundamentally restricts the QoS of data appli-
for a vehicle moving at 120 km/h [92]. Moreover, time spent in cations. To improve the performance of drive-thru Internet,
WiFi association, authentication, and IP configuration before solutions in the literature include: 1) reducing connection
actual data transfer is not negligible. Second, V2I communi- establishment time [96]; 2) improving transport protocols to
cations also suffer from high wireless loss due to the channel deal with the intermittent connectivity and wireless transmission
fading and shadowing [56]. Third, the WiFi protocol stack is not losses [96]; 3) enhancing MAC protocols for high mobility [95],
a specific design for a high mobility environment. [96], [98]; 4) designing efficient handoff schemes to deal with
2) Real-World Measurement: There have been a number of the frequent disruptions [5]; 5) using the multihop V2V
real-world measurements in the literature based on diverse communication capability for relaying data traffic [99], as
testbed experiments to characterize and evaluate the shown in Fig. 3; 6) exploiting cooperation among vehicles
performance of the drive-thru Internet. In [92] and [93], the [100], [101]; and 7) optimal deployment of WiFi APs [102].
drive-thru Internet based on the IEEE 802.11b and 802.11 g is
C. Toward Cost-Effective Solutions
evaluated, respectively, in a planned highway scenario where
two APs are closely deployed. Different vehicle speeds (80, 120, Cellular-based access technologies play a vital role in provid-
and 180 km/h) and different transport layer protocols (UDP and ing reliable and ubiquitous Internet access to vehicles, however,
TCP) are considered. A very important observation is that the with relatively high cost both in terms of capital expenditures
drive-thru Internet has a three-phase (entry, production, and exit) (CAPEX) and operational expenditures (OPEX) (resulting in an
characteristic. In the entry and exit phases, vehicles gain less expensive cellular service). Although Road-side WiFi network
throughput than when in production phase due to such as the can be a deployable solution, the problem of such a drive-thru
weak signal, connection establishment delay, and rate access is that in-vehicle users may have to tolerate intermittent
overestimation. To investigate the impact of backhaul connectivity so as to compromise the users’ satisfaction on QoS.
capability on drive-thru Internet, a measurement is conducted Therefore, there remains great uncertainty as to the cost effec-
in [94] on a traffic free road where the interference from other tiveness in applying the existing solutions to deliver Internet
vehicles does not exist. It is evidenced that the performance of the services to highly mobile vehicles. On the other hand, as the
drive-thru Internet suffers a lot from the limitation of the explosive growth of mobile data traffic have already congested
backhaul network. For example, with a 1 Mb/s bandwidth today’s cellular networks, off-the-shelf cellular infrastructures
backhaul capability, the data volume transferred within a might not be able to support a huge number of connected
drive-thru reduces from 92 to 25 MB. Moreover, a backhaul vehicles; in spite of free open WiFi network, the carrier WiFi,
connection with 100 ms one-way delay significantly degrades i.e., deployed by carrier operators, is also expected to provide
the performance of web services due to the time penalty of HTTP more robust and secure Internet access. As the deployment and
requests and responses. The problems that may cause the operation cost of the wireless network infrastructure is a domi-
performance degradation of the drive-thru Internet are nant factor in bringing Internet on wheels into reality, developing
296 IEEE INTERNET OF THINGS JOURNAL, VOL. 1, NO. 4, AUGUST 2014

TABLE IV
SUMMARY OF REAL-WORLD MEASUREMENT RESULTS

innovative cost-effective vehicular networking solution to en- scenarios. To combat outdoor optical noises, however, advanced
able high-quality vehicular Internet access is of great importance. receiver is required, such as high-speed camera [107], [108],
Research to this end is quite limited. A head-to-head compari- which may incur a high implementation cost. Compared to the
son of the performance characteristics of a 3G network and a mature IEEE 802.11-based technology, VLC is still in the
metro-scale commercial WiFi network is performed in [103], and introductory phase and substantial efforts are needed before it
a hybrid network design is suggested to enhance the network can be widely deployed for short-range ITSs applications.
performance. Fundamental relations between network through-
put and infrastructure cost (CAPEX and OPEX) are established VI. CONCLUSION
in [104] for cellular deployment and WiFi-based deployment,
respectively. The WiFi-based solution is suggested for providing In this paper, we have presented an overview of the state-of-
a cost-effective data pipe to vehicles, however, without consid- the-art wireless solutions to vehicle-to-sensor, vehicle-to-
ering service delay and opportunistic communications among vehicle, vehicle-to-Internet, and vehicle-to-road infrastructure
vehicles. In addition to the cost-performance issue of network connectivities. We have discussed the potential challenges and
infrastructure, WiFi offloading strategy is also a research focus, identified the space for future improvement. The biggest chal-
which is to offload cellular traffic through WiFi networks so that lenge for efficient and robust wireless connections is to combat
the cellular congestion can be alleviated and the usage cost of the harsh communication environment inside and/or outside the
access service can be reduced for mobile users. The performance vehicle. In addition, the significant research and development
of offloading cellular traffic via drive-thru Internet remains efforts are required to deal with the following issues.
unclear in the literature. Such an opportunistic WiFi offloading 1) To enable various wireless connectivity, multiple radio
has unique features. 1) A relatively small amount of data can be interfaces have to be implemented, such as DSRC/WAVE,
delivered to a vehicle in each drive-thru, due to the short WiFi, and 3G/4G-LTE interfaces, which may incur a high
connection time with WiFi APs and 2) the offloading perfor- cost and thereby impede the development of connected
mance can be significantly improved if the data service can be vehicles. A unified solution to provide V2X connectivity
deferred for a certain time, as vehicles with a high speed can have with low cost might be required.
multiple drive-thru opportunities in a short future. The chal- 2) In-vehicle systems have stringent requirement on latency
lenges and solutions of WiFi offloading in a vehicular environ- and reliability for control/monitoring purposes. The full
ment are discussed in [105]. adoption of V2S connectivity may not be feasible in the
near future unless V2S connectivity can provide the same
performance and reliability as the wired communication [43].
V. V2R CONNECTIVITY 3) Connected vehicle offers the driver a variety of informa-
V2R connectivity is critical to avoid or mitigate the effects of tion. However, research in [111] and [112] suggests an up
road accidents, and to enable the efficient management of ITSs. limit on information provided to the driver. Excessive
DSRC/WAVE is considered a key technology to enable con- information increases the driver’s workload and hence has
nections between vehicles and ITSs infrastructure, such as traffic a negative impact on safety. Therefore, the vehicle infor-
lights, street signs, and roadside sensors. Moreover, roadside mation system has to be appropriately designed for offer-
infrastructures can also be commercial content providers, such as ing information to drivers.
the roadside unit (RSU) broadcasting flyers of superstores [106].
RSU does not necessarily serve as Internet gateway as wireless REFERENCES
infrastructures in V2I communications. Visible light communi- [1] Transparency Market Research, “Connected car market—Global industry
cation (VLC), transmitting data by using light-emitting diodes analysis, size, share, growth, trends and forecast, 2013–2019,” Transpar-
(LEDs), has also been proposed for road-to-vehicle ITSs appli- ency Market Res., Albany, NY, USA, 2013.
[2] D. Schrank, B. Eisele, and T. Lomax, “TTI’s 2012 urban mobility report,”
cations, such as traffic light control at intersections [107]–[109]. Texas A&M Transport. Inst. and Texas A&M Univ. Syst., College Station,
As a key technology specified in the IEEE 802.15.7 standard, TX, USA, 2012.
VLC can support a data rate up to 96 Mb/s through fast [3] J. Anda, J. LeBrun, D. Ghosal, C.-N. Chuah, and M. Zhang, “VGrid:
Vehicular adhoc networking and computing grid for intelligent traffic
modulation of LED light sources [110]. VLC is becoming an control,” in Proc. IEEE 61st Veh. Technol. Conf. (VTC’05 Spring),
intriguing complement to DSRC for LOS communication Stockholm, Sweden, Jun. 2005, pp. 2905–2909.
LU et al.: CONNECTED VEHICLES: SOLUTIONS AND CHALLENGES 297

[4] C. Olaverri-Monreal, P. Gomes, R. Fernandes, F. Vieira, and M. Ferreira, [31] T. ElBatt, C. Saraydar, M. Ames, and T. Talty, “Potential for intra-vehicle
“The see-through system: A VANET-enabled assistant for overtaking wireless automotive sensor networks,” in Proc. IEEE Sarnoff Symp.,
maneuvers,” in Proc. IEEE Intell. Veh. Symp. (IV), San Diego, CA, USA, Princeton, NJ, USA, Mar. 2006, pp. 1–4.
Jun. 2010, pp. 123–128. [32] O. K. Tonguz, H.-M. Tsai, C. Saraydar, T. Talty, and A. Macdonald,
[5] A. Balasubramanian, R. Mahajan, A. Venkataramani, B. N. Levine, and “Intra-car wireless sensor networks using RFID: Opportunities and
J. Zahorjan, “Interactive WiFi connectivity for moving vehicles,” ACM challenges,” in Proc. IEEE Mobile Netw. Veh. Environ., Anchorage, AK,
SIGCOMM Comput. Commun. Rev., vol. 38, no. 4, pp. 427–438, 2008. USA, May 2007, pp. 43–48.
[6] J. Lin, S. Chen, Y. Shih, and S. Chen, “A study on remote on-line [33] K. Mandal et al., “Road traffic congestion monitoring and measurement
diagnostic system for vehicles by integrating the technology of OBD, using active RFID and GSM technology,” in Proc. 14th Int. IEEE
GPS, and 3G,” World Acad. Sci. Eng. Technol., vol. 56, p. 56, 2009. Conf. Intell. Transp. Syst. (ITSC), Washington, DC, USA, Oct. 2011,
[7] M. Ramadan, M. Al-Khedher, and S. Al-Kheder, “Intelligent anti-theft and pp. 1315–1379.
tracking system for automobiles,” Int. J. Mach. Learn. Comput., vol. 2, [34] H. Yue, C. Zhang, M. Pan, Y. Fang, and S. Chen, “A time-efficient
no. 1, pp. 88–92, Feb. 2012. information collection protocol for large-scale RFID systems,” in Proc.
[8] Telefnica, “Connected car industry report,” 2013. IEEE INFOCOM, Orlando, FL, USA, Mar. 2012, pp. 2158–2166.
[9] “eCall,” [Online]. Available: http://ec.europa.eu/digital-agenda/en/ecall- [35] I. Oppermann, M. Hämäläinen, and J. Iinatti, UWB: Theory and Applica-
time-saved-lives-saved tions. Hoboken, NJ, USA: Wiley, 2005.
[10] (2014) [Online]. Available: http://www.nhtsa.gov/About+NHTSA/Press+ [36] W. Niu, J. Li, S. Liu, and T. Talty, “Intra-vehicle ultra-wideband com-
Releases/2014/USDOT+to+Move+Forward+with+Vehicle-to-Vehicle+ munication testbed,” in Proc. IEEE Mil. Commun. Conf. (MILCOM’07),
Communication+Technology+for+Light+Vehicles Orlando, FL, USA, Oct. 2007, pp. 1–6.
[11] N. Liu, “Internet of Vehicles: Your next connection,” Huawei WinWin, [37] P. C. Richardson, W. Xiang, and W. Stark, “Modeling of ultra-wideband
vol. 11, pp. 23–28, 2011. channels within vehicles,” IEEE J. Sel. Areas Commun., vol. 24, no. 4,
[12] F. Qu, F.-Y. Wang, and L. Yang, “Intelligent transportation spaces: pp. 906–912, Apr. 2006.
Vehicles, traffic, communications, and beyond,” IEEE Commun. Mag., [38] J. Li and T. Talty, “Channel characterization for ultra-wideband
vol. 48, no. 11, pp. 136–142, Nov. 2010. intra-vehicle sensor networks,” in Proc. IEEE Mil. Commun. Conf.
[13] M. Faezipour, M. Nourani, A. Saeed, and S. Addepalli, “Progress and (MILCOM’06), Washington, DC, USA, Oct. 2006, pp. 1–5.
challenges in intelligent vehicle area networks,” Commun. ACM, vol. 55, [39] W. Xiang, “A vehicular ultra-wideband channel model for future wireless
no. 2, pp. 90–100, 2012. intra-vehicle communications (IVC) systems,” in Proc. IEEE 66th
[14] M. Pinelis, “Automotive sensors and electronics: Trends and developments Veh. Technol. Conf. (VTC’07 Fall), Baltimore, MD, USA, Oct. 2007,
in 2013,” in Automot. Sensors Electron. Expo, Detroit, MI, USA, Oct. 2013. pp. 2159–2163.
[15] M. Ahmed et al., “Intra-vehicular wireless networks,” in Proc. IEEE [40] W. Niu, J. Li, and T. Talty, “Intra-vehicle UWB channel measurements
Globecom Workshops, Washington, DC, USA, Nov. 2007, pp. 1–9. and statistical analysis,” in Proc. IEEE Global Telecommun. Conf.
[16] L. D’Orazio, F. Visintainer, and M. Darin, “Sensor networks on the car: (GLOBECOM’08), San Francisco, CA, USA, Dec. 2008, pp. 1–5.
State of the art and future challenges,” in IEEE Des. Autom. Test Eur. Conf. [41] M. Schack et al., “Measurements and analysis of an in-car UWB channel,”
Exhib. (DATE), 2011, pp. 1–6. in Proc. IEEE Veh. Technol. Conf. (VTC Spring’08), Singapore, May
[17] S. Tuohy, M. Glavin, E. Jones, M. Trivedi, and L. Kilmartin, “Next 2008, pp. 459–463.
generation wired intra-vehicle networks, a review,” in Proc. IEEE Intell. [42] W. Niu, J. Li, and T. Talty, “Intra-vehicle UWB channels in moving and
Veh. Symp. (IV), Gold Coast, Australia, Jun. 2013, pp. 777–782. staionary scenarios,” in Proc. IEEE Mil. Commun. Conf. (MILCOM’09),
[18] H.-M. Tsai et al., “Feasibility of in-car wireless sensor networks: A Boston, MA, USA, Oct. 2009, pp. 1–6.
statistical evaluation,” in Proc. IEEE SECON, San Diego, CA, USA, Jun. [43] C. U. Bas and S. C. Ergen, “Ultra-wideband channel model for intra-
2007, pp. 101–111. vehicular wireless sensor networks beneath the chassis: From statistical
[19] A. R. Moghimi, H.-M. Tsai, C. U. Saraydar, and O. K. Tonguz, “Character- model to simulations,” IEEE Trans. Veh. Technol., vol. 62, no. 1, pp. 14–
izing intra-car wireless channels,” IEEE Trans. Veh. Technol., vol. 58, 25, Jan. 2013.
no. 9, pp. 5299–5305, Nov. 2009. [44] Y. Sadi and S. Ergen, “Optimal power control, rate adaptation and
[20] T. Zhang, H. Antunes, and S. Aggarwal, “Defending connected vehicles scheduling for UWB-based intra-vehicular wireless sensor networks,”
against malware: Challenges and a solution framework,” IEEE Internet IEEE Trans. Veh. Technol., vol. 62, no. 1, pp. 219–234, Jan. 2013.
Things J., vol. 1, no. 1, pp. 10–21, Feb. 2014. [45] J. Qiao, L. X. Cai, X. S. Shen, and J. W. Mark, “Enabling multi-hop
[21] C. Bisdikian, “An overview of the Bluetooth wireless technology,” IEEE concurrent transmissions in 60 GHz wireless personal area networks,” IEEE
Commun. Mag., vol. 39, no. 12, pp. 86–94, Dec. 2001. Trans. Wireless Commun., vol. 10, no. 11, pp. 3824–3833, Nov. 2011.
[22] W. Ye, J. Heidemann, and D. Estrin, “Medium access control with [46] IEEE 802.15.3c, IEEE 802.15 WPAN Millimeter Wave Alternative PHY
coordinated adaptive sleeping for wireless sensor networks,” IEEE/ACM Task Group 3c (TG3c) Standard, 2009.
Trans. Netw., vol. 12, no. 3, pp. 493–506, Jun. 2004. [47] IEEE 802.11ad, IEEE P802.11 Task Group AD Standard, 2012.
[23] Bluetooth SIG. (2004). Specification of the Bluetooth system—2.0 + EDR [48] H. Sawada et al., “A sixty GHz intra-car multi-media communications
[Online]. Available: https://www.bluetooth.org/en-us/specification/ system,” in Proc. IEEE 69th Veh. Technol. Conf. (VTC Spring’09), Hilton
adopted-specifications Diagonal Mar, Spain, Apr. 2009, pp. 1–5.
[24] IEEE Standard for Information Technology—Telecommunications and [49] M. Peter, R. Felbecker, W. Keusgen, and J. Hillebrand, “Measurement-
Information Exchange Between Systems—Local and Metropolitan Area based investigation of 60 GHz broadband transmission for wireless in-car
Networks Specific Requirements Part 15.4: Wireless Medium Access communication,” in Proc. IEEE 70th Veh. Technol. Conf. Fall(VTC’09
Control (MAC) and Physical Layer (PHY) Specifications for Low-Rate Fall), Anchorage, AK, USA, Sep. 2009, pp. 1–5.
Wireless Personal Area Networks (LR-WPANs), IEEE Standard 802.15.4- [50] M. Schack, M. Jacob, and T. Kiirner, “Comparison of in-car UWB and
2006, 2006. 60 GHz channel measurements,” in Proc. Eur. Conf. Antennas Propag.
[25] O. K. Tonguz, H.-M. Tsai, T. Talty, A. Macdonald, and C. Saraydar, “RFID (EuCAP), Barcelona, Spain, Apr. 2010, pp. 1–5.
technology for intra-car communications: A new paradigm,” in Proc. [51] R. Nakamura and A. Kajiwara, “Empirical study on 60 GHz in-vehicle
IEEE Veh. Technol. Conf. (VTC’06 Fall), Montreal, QC, Canada, Sep. radio channel,” in Proc. IEEE Radio Wireless Symp. (RWS), Santa Clara,
2006, pp. 1–6. CA, USA, Jan. 2012, pp. 327–330.
[26] High Rate Ultra Wideband PHY and MAC Standard, ECMA International [52] F. Bai and B. Krishnamachari, “Exploiting the wisdom of the crowd:
Std. ECMA-368 (2008) [Online]. Available: http://www.ecma-international. Localized, distributed information-centric VANETs,” IEEE Commun.
org/publications/files/ECMA-ST/ECMA-368.pdf Mag., vol. 48, no. 5, pp. 138–146, May 2010.
[27] T. Baykas et al., “IEEE 802.15. 3c: The first IEEE wireless standard for [53] T. H. Luan, X. Shen, and F. Bai, “Integrity-oriented content transmission in
data rates over 1 Gb/s,” IEEE Commun. Mag., vol. 49, no. 7, pp. 114–121, highway vehicular ad hoc networks,” in Proc. IEEE INFOCOM, Turin,
Jul. 2011. Italy, Apr. 2013, pp. 2562–2570.
[28] “Zigbee specification,” ZigBee Alliance (2004 Dec.) [Online]. Available: [54] M. Boban, T. T. Vinhoza, M. Ferreira, J. Barros, and O. K. Tonguz, “Impact
http://www.zigbee.org of vehicles as obstacles in vehicular ad hoc networks,” IEEE J. Sel. Areas
[29] H.-M. Tsai et al., “Zigbee-based intra-car wireless sensor networks: A case Commun., vol. 29, no. 1, pp. 15–28, Jan. 2011.
study,” IEEE Wireless Commun., vol. 14, no. 6, pp. 67–77, Dec. 2007. [55] L. Cheng, B. E. Henty, D. D. Stancil, F. Bai, and P. Mudalige, “Mobile
[30] R. de Francisco, L. Huang, G. Dolmans, and H. de Groot, “Coexistence of vehicle-to-vehicle narrow-band channel measurement and characterization of
ZigBee wireless sensor networks and bluetooth inside a vehicle,” in Proc. the 5.9 GHz dedicated short range communication (DSRC) frequency band,”
IEEE PIMRC, Tokyo, Japan, Sep. 2009, pp. 2700–2704. IEEE J. Sel. Areas Commun., vol. 25, no. 8, pp. 1501–1516, Oct. 2007.
298 IEEE INTERNET OF THINGS JOURNAL, VOL. 1, NO. 4, AUGUST 2014

[56] C. F. Mecklenbrauker et al., “Vehicular channel characterization and its [80] S. Pagadarai, A. Wyglinski, and R. Vuyyuru, “Characterization of vacant
implications for wireless system design and performance,” Proc. IEEE, UHF TV channels for vehicular dynamic spectrum access,” in Proc. IEEE
vol. 99, no. 7, pp. 1189–1212, Jul. 2011. Veh. Netw. Conf. (VNC’09), Tokyo, Japan, Oct. 2009, pp. 1–8.
[57] G. Karagiannis et al., “Vehicular networking: A survey and tutorial [81] V. Bychkovsky, B. Hull, A. Miu, H. Balakrishnan, and S. Madden,
on requirements, architectures, challenges, standards and solutions,” “A measurement study of vehicular internet access using in situ wi-fi
IEEE Commun. Surv. Tuts., vol. 13, no. 4, pp. 584–616, Fourth Quart. networks,” in Proc. ACM MobiCom, Los Angeles, CA, USA, Sep. 2006,
2011. pp. 50–61.
[58] J. B. Kenney, “Dedicated short-range communications (DSRC) standards [82] A. Jameel, A. Fuchs, and M. Stuempfle, “Internet multimedia on wheels:
in the United States,” Proc. IEEE, vol. 99, no. 7, pp. 1162–1182, Jul. 2011. Connecting cars to cyberspace,” in Proc. IEEE Conf. Intell. Transp. Syst.
[59] H. Hartenstein and K. Laberteaux, VANET: Vehicular Applications and (ITSC’97), Boston, MA, USA, Nov. 1997, pp. 637–642.
Inter-Networking Technologies. Hoboken, NJ, USA: Wiley, 2010. [83] A. Jameel, M. Stuempfle, D. Jiang, and A. Fuchs, “Web on wheels: Toward
[60] H. Hartenstein and K. P. Laberteaux, “A tutorial survey on vehicular internet-enabled cars,” IEEE Comput., vol. 31, no. 1, pp. 69–76, Jan. 1998.
ad hoc networks,” IEEE Commun. Mag., vol. 46, no. 6, pp. 164–171, [84] R. Lind et al., “The network vehicle-a glimpse into the future of mobile
Jun. 2008. multi-media,” IEEE Aerosp. Electron. Syst. Mag., vol. 14, no. 9, pp. 27–
[61] J. Yin et al., “Performance evaluation of safety applications over DSRC 32, Sep. 1999.
vehicular ad hoc networks,” in Proc. 1st ACM Int. Workshop Veh. ad hoc [85] Car Connectivity Consortium, MirrorLink [Online]. Available: http://
Netw. (ACM VANET), Philadelphia, PA, USA, Oct. 2004, pp. 1–9. www.mirrorlink.com/
[62] V. Taliwal, D. Jiang, H. Mangold, C. Chen, and R. Sengupta, “Empirical [86] Toyota Touch 2 [Online]. Available: http://www.toyota-europe.com/
determination of channel characteristics for DSRC vehicle-to-vehicle about/news_and_events/editorial/toyota-touch-2.tmex
communication,” in Proc. 1st ACM Int. Workshop Veh. ad hoc Netw. [87] Apple CarPlay [Online]. Available: http://www.apple.com/ca/ios/carplay/
(ACM VANET), Philadelphia, PA, USA, Oct. 2004, p. 88. ?cid=wwa-ca-kwg-features-com
[63] F. Bai, D. D. Stancil, and H. Krishnan, “Toward understanding character- [88] OnStar [Online]. Available: https://www.onstar.com
istics of dedicated short range communications (DSRC) from a perspective [89] BMW ConnectedDrive [Online]. Available: http://www.bmw.com/com/
of vehicular network engineers,” in Proc. ACM MobiCom, Chicago, IL, en/insights/technology/connecteddrive/2013/index.html
USA, Sep. 2010, pp. 329–340. [90] Audi connect [Online]. Available: http://www.audiusa.com/help/audi-
[64] X. Wu et al., “Vehicular communications using DSRC: Challenges, connect
enhancements, and evolution,” IEEE J. Sel. Areas Commun., vol. 31, [91] LTE connected car–The ng Connect Program [Online]. Available: http://
no. 9, pp. 399–408, Sep. 2013. www.ngconnect.org/program/connected-car.htm
[65] T. Jiang, H.-H. Chen, H.-C. Wu, and Y. Yi, “Channel modeling and [92] J. Ott and D. Kutscher, “Drive-thru Internet: IEEE 802.11 b for automo-
inter-carrier interference analysis for V2V communication systems in bile,” in Proc. IEEE INFOCOM, Hong Kong, China, Mar. 2004.
frequency-dispersive channels,” Mobile Netw. Appl., vol. 15, no. 1, [93] J. Ott and D. Kutscher, “The drive-thru architecture: WLAN-based Intemet
pp. 4–12, 2010. access on the road,” in Proc. IEEE 59th Veh. Technol. Conf. (VTC’04
[66] F. Yu and S. Biswas, “Self-configuring TDMA protocols for enhancing Spring), May 2004, pp. 2615–2622.
vehicle safety with DSRC based vehicle-to-vehicle communications,” [94] R. Gass, J. Scott, and C. Diot, “Measurements of in-motion 802.11
IEEE J. Sel. Areas Commun., vol. 25, no. 8, pp. 1526–1537, Oct. 2007. networking,” in Proc. 7th IEEE Workshop Mobile Comput. Syst. Appl.,
[67] N. Lu, Y. Ji, F. Liu, and X. Wang, “A dedicated multi-channel MAC Semiahmoo Resort, WA, USA, Aug. 2005, pp. 69–74.
protocol design for VANET with adaptive broadcasting,” in Proc. IEEE [95] D. Hadaller, S. Keshav, T. Brecht, and S. Agarwal, “Vehicular opportu-
Wireless Commun. Netw. Conf. (WCNC’10), Sydney, N.S.W., Australia, nistic communication under the microscope,” in Proc. ACM MobiSys, San
Apr. 2010, pp. 1–6. Juan, PR, USA, Jun. 2007, pp. 206–219.
[68] H. Omar, W. Zhuang, and L. Li, “VeMAC: A TDMA-based MAC [96] J. Eriksson, H. Balakrishnan, and S. Madden, “Cabernet: Vehicular
protocol for reliable broadcast in VANETs,” IEEE Trans. Mobile Comput., content delivery using WiFi,” in Proc. ACM MobiCom, San Francisco,
vol. 12, no. 9, pp. 1724–1736, Sep. 2012. CA, USA, Sep. 2008, pp. 199–210.
[69] S. Bharati and W. Zhuang, “CAH-MAC: Cooperative ADHOC MAC for [97] Wi-Fi Certified Passpoint [Online]. Available: http://www.wi-fi.org/
vehicular networks,” IEEE J. Sel. Areas Commun., vol. 31, no. 9, pp. 470– discover-wi-fi/wi-fi-certified-passpoint
479, Sep. 2013. [98] T. H. Luan, X. Ling, and X. Shen, “Mac in motion: Impact of mobility on
[70] Y. Ihara et al., “Distributed autonomous multi-hop vehicle-to-vehicle the mac of drive-thru internet,” IEEE Trans. Mobile Comput., vol. 11,
communications over TV white space,” in Proc. IEEE Consum. Commun. no. 2, pp. 305–319, Feb. 2012.
Netw. Conf. (CCNC’13), Las Vegas, NV, USA, Jan. 2013, pp. 336–344. [99] J. Zhao, T. Arnold, Y. Zhang, and G. Cao, “Extending drive-thru data
[71] M. Di Felice, R. Doost-Mohammady, K. R. Chowdhury, and L. Bononi, access by vehicle-to-vehicle relay,” in Proc. ACM VANET, San Francisco,
“Smart radios for smart vehicles: Cognitive vehicular networks,” IEEE CA, USA, Sep. 2008, pp. 66–75.
Veh. Technol. Mag., vol. 7, no. 2, pp. 26–33, Jun. 2012. [100] O. Trullols-Cruces, M. Fiore, and J. Barcelo-Ordinas, “Cooperative down-
[72] N. Cheng et al., “Opportunistic spectrum access for CR-VANETs: A game load in vehicular environments,” IEEE Trans. Mobile Comput., vol. 11,
theoretic approach,” IEEE Trans. Veh. Technol., vol. 63, no. 1, pp. 237– no. 4, pp. 663–678, Apr. 2012.
251, Jan. 2014. [101] F. Malandrino, C. Casetti, C. Chiasserini, and M. Fiore, “Optimal content
[73] M. I. Hassan, H. L. Vu, and T. Sakurai, “Performance analysis of the IEEE downloading in vehicular networks,” IEEE Trans. Mobile Comput.,
802.11 MAC protocol for DSRC safety applications,” IEEE Trans. Veh. vol. 12, no. 7, pp. 1377–1391, Jul. 2013.
Technol., vol. 60, no. 8, pp. 3882–3896, Oct. 2011. [102] Z. Zheng, P. Sinha, and S. Kumar, “Sparse WiFi deployment for vehicular
[74] T. Wang, L. Song, and Z. Han, “Coalitional graph games for popular internet access with bounded interconnection gap,” IEEE/ACM Trans.
content distribution in cognitive radio VANETs,” IEEE Trans. Veh. Netw., vol. 20, no. 3, pp. 956–969, Jun. 2012.
Technol., vol. 62, no. 8, pp. 4010–4019, Oct. 2013. [103] P. Deshpande, X. Hou, and S. R. Das, “Performance comparison of 3G and
[75] O. Altintas et al., “Demonstration of vehicle to vehicle communications metro-scale WiFi for vehicular network access,” in Proc. ACM SIGCOMM
over TV white space,” in Proc. IEEE Veh. Technol. Conf. (VTC Fall), San Conf. Internet Meas., Melbourne, Vic., Australia, Nov. 2010, pp. 301–307.
Francisco, CA, USA, Sep. 2011, pp. 1–3. [104] N. Lu et al., “Vehicles meet infrastructure: Towards capacity-cost trade-
[76] O. Altintas et al., “Field tests and indoor emulation of distributed autono- offs for vehicular access networks,” IEEE Trans. Intell. Transp. Syst.,
mous multi-hop vehicle-to-vehicle communications over TV white vol. 14, no. 3, pp. 1266–1277, Sep. 2013.
space,” ACM SIGMOBILE Mobile Comput. Commun. Rev., vol. 16, no. 4, [105] N. Cheng, N. Lu, N. Zhang, X. Shen, and J. W. Mark, “Vehicular WiFi
pp. 54–57, 2013. offloading: Challenges and solutions,” Veh. Commun., vol. 1, no. 1,
[77] S. J. Shellhammer, A. K. Sadek, and W. Zhang, “Technical challenges for pp. 13–21, 2014.
cognitive radio in the TV white space spectrum,” in Proc. Inf. Theory Appl. [106] T. H. Luan, L. X. Cai, J. Chen, X. Shen, and F. Bai, “Vtube: Towards the
Workshop, San Diego, CA, USA, Feb. 2009, pp. 323–333. media rich city life with autonomous vehicular content distribution,” in
[78] A. Flores, R. Guerra, E. Knightly, P. Ecclesine, and S. Pandey, “IEEE Proc. IEEE 8th Annu. IEEE Commun. Society Conf. Sensor Mesh Ad Hoc
802.11 af: A standard for TV white space spectrum sharing,” IEEE Commun. Networks (SECON), Salt Lake City, UT, USA, Jun. 2011,
Commun. Mag., vol. 51, no. 10, pp. 92–100, Oct. 2013. pp. 359–367.
[79] C. R. Stevenson et al., “IEEE 802.22: The first cognitive radio wireless [107] M. Wada, T. Yendo, T. Fujii, and M. Tanimoto, “Road-to-vehicle
regional area network standard,” IEEE Commun. Mag., vol. 47, no. 1, communication using led traffic light,” in Proc. IEEE Intell. Veh. Symp.,
pp. 130–138, Jan. 2009. Jun. 2005, pp. 601–606.
LU et al.: CONNECTED VEHICLES: SOLUTIONS AND CHALLENGES 299

[108] S. Iwasaki et al., “Visible light road-to-vehicle communication using Dr. Shen served as the Technical Program Committee Chair/Co-Chair for
high-speed camera,” in Proc. IEEE Intell. Veh. Symp., Eindhoven, The IEEE Infocom’14, IEEE VTC’10 Fall, the Symposia Chair for IEEE ICC’10, the
Netherlands, Jun. 2008, pp. 13–18. Tutorial Chair for IEEE VTC’11 Spring and IEEE ICC’08, the Technical Program
[109] C. B. Liu, B. Sadeghi, and E. W. Knightly, “Enabling vehicular visible Committee Chair for IEEE Globecom’07, the General Co-Chair for Chinacom’07
light communication (V2LC) networks,” in Proc. 8th ACM Int. Workshop and QShine’06, the Chair for IEEE Communications Society Technical Com-
Veh. Inter-Netw. (VANET), Las Vegas, NV, USA, Sep. 2011, pp. 41–50. mittee on Wireless Communications, and P2P Communications and Networking.
[110] S. Rajagopal, R. D. Roberts, and S.-K. Lim, “IEEE 802.15. 7 visible light He also serves/served as the Editor-in-Chief for IEEE Network, Peer-to-Peer
communication: Modulation schemes and dimming support,” IEEE Networking and Application, and IET Communications; a Founding Area Editor
Commun. Mag., vol. 50, no. 3, pp. 72–82, Mar. 2012. for the IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS; an Associate Editor for
[111] J.-M. Girard, N. Tricot, K. Younsi, and J.-C. Popieul, “When does the the IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, Computer Networks, and
driver workload reaches its limits?” in Proc. IEEE Intell. Transp. Syst. ACM/Wireless Networks, etc., and the Guest Editor for the IEEE JOURNAL ON
Conf. (ITSC’06), Toronto, ON, Canada, Sep. 2006, pp. 578–583. SELECTED AREAS IN COMMUNICATIONS, IEEE Wireless Communications, IEEE
[112] C. Wu and Y. Liu, “Queuing network modeling of driver workload and Communications Magazine, and ACM Mobile Networks and Applications, etc.
performance,” IEEE Trans. Intell. Transp. Syst., vol. 8, no. 3, pp. 528–537, He is a Registered Professional Engineer of Ontario, Canada, an Engineering
Sep. 2007. Institute of Canada Fellow, a Canadian Academy of Engineering Fellow, and a
Distinguished Lecturer of the IEEE Vehicular Technology Society and Com-
munications Society. He was the recipient of the Excellent Graduate Supervision
Ning Lu (S’12) received the B.Sc. and M.Sc. degrees Award in 2006, and the Outstanding Performance Award in 2004, 2007, and 2010
in electrical engineering from Tongji University, from the University of Waterloo, the Premier’s Research Excellence Award
Shanghai, China, in 2007 and 2010, respectively, and (PREA) in 2003 from the Province of Ontario, Canada, and the Distinguished
is currently working toward the Ph.D. degree in Performance Award in 2002 and 2007 from the Faculty of Engineering, Univer-
electrical and computer engineering at the University sity of Waterloo.
of Waterloo, Waterloo, ON, Canada.
His research interests include capacity and delay
analysis, real-time scheduling, and cross-layer design Jon W. Mark (M’62–SM’80–F’88–LF’03) received
for vehicular networks. He also did some collaborative the Ph.D. degree in electrical engineering from
research on cooperative cognitive radio networks. McMaster University, Hamilton, ON, Canada, in
Mr. Lu served as a Technical Program Committee 1970.
member for IEEE PIMRC’12, WCSP’13, WCSP’14, and ICNC’15. In September 1970, he joined the Department of
Electrical and Computer Engineering, University of
Waterloo, Waterloo, ON, Canada, where he is cur-
Nan Cheng (S’13) received the B.S. and M.S. degrees rently a Distinguished Professor Emeritus. He served
in electrical engineering from Tongji University, as the Department Chairman from July 1984 to June
Shanghai, China, in 2009 and 2012, respectively, and 1990. In 1996, he established the Center for Wireless
is currently working toward the Ph.D. degree in Communications (CWC), University of Waterloo,
electrical and computer engineering at the University and is currently serving as its Founding Director. He had been on sabbatical
of Waterloo, Waterloo, ON, Canada. leave at the following places: IBM Thomas J. Watson Research Center, Yorktown
Since 2012, he has been a Research Assistant Heights, NY, USA, as a Visiting Research Scientist (1976–1977); AT&T Bell
with the Broadband Communication Research Group, Laboratories, Murray Hill, NJ, USA, as a Resident Consultant (1982–1983);
ECE Department, University of Waterloo. His research Laboratoire MASI, UniversitPierre et Marie Curie, Paris France, as an Invited
interests include vehicular communication networks, Professor (1990–1991); and the Department of Electrical Engineering, National
cognitive radio networks, and cellular traffic offloading. University of Singapore, as a Visiting Professor (1994–1995). He worked in the
areas of adaptive equalization, image and video coding, spread spectrum com-
munications, computer communication networks, ATM switch design and traffic
Ning Zhang (S’12) received the B.Sc. degree from management. His research interests include broadband wireless communications,
Beijing Jiaotong University, Beijing, China, in 2007, resource and mobility management, and cross domain interworking.
the M.Sc. degree from Beijing University of Posts and Dr. Mark is a Fellow of the Canadian Academy of Engineering. He was an
Telecommunications, Beijing, China, in 2010, and is editor of the IEEE TRANSACTIONS ON COMMUNICATIONS (1983–1990), a member of
currently working toward the Ph.D. degree in electri- the Inter-Society Steering Committee of the IEEE/ACM TRANSACTIONS ON
cal and computer engineering at the University of NETWORKING (1992–2003), a member of the IEEE Communications Society
Waterloo, Waterloo, ON, Canada. Awards Committee (1995–1998), an editor of Wireless Networks (1993–
His research interests include cooperative net- 2004), and an Associate Editor of Telecommunication Systems (1994–2004).
working, cognitive radio networks, physical layer He was the recipient of the 2000 Canadian Award for Telecommunications
security, and vehicular networks. Research and the 2000 Award of Merit of the Education Foundation of the
Federation of Chinese Canadian Professionals.
Xuemin Shen (M’97–SM’02–F’09) received the
B.Sc. degree from Dalian Maritime University,
Dalian, China, in 1982, and the M.Sc. and Ph.D.
degrees from Rutgers University, New Brunswick,
NJ, USA, in 1987 and 1990, respectively, all in
electrical engineering.
He is a Professor and University Research Chair
with the Department of Electrical and Computer
Engineering, University of Waterloo, Waterloo, ON,
Canada. He was the Associate Chair for Graduate
Studies from 2004 to 2008. He is a coauthor/editor of
six books and has authored or coauthored more than 600 papers and book chapters
in wireless communications and networks and control and filtering. His research
interests include resource management in interconnected wireless/wired net-
works, wireless network security, social networks, smart grid, and vehicular ad
hoc and sensor networks.