Evaluating the Dedicated Short-range Communication for Connected
Vehicles against Network Security Attacks
Tu Le, Ingy Elsayed-Aly, Weizhao Jin, Seunghan Ryu, Guy Verrier, Tamjid Al Rahat, B. Brian Park
and Yuan Tian
School of Engineering and Applied Science, University of Virginia, Charlottesville, Virginia, U.S.A.
Keywords:
Cybersecurity, Network, DSRC, Connected Vehicles, VANET.
Abstract:
According to the National Highway Traffic Safety Administration, there are more than 5 million road crashes
every year in the U.S. More than 90 people die in car crashes every day. Even though the number of people sur-
viving crashes has increased significantly thanks to safety features, such as airbags and anti-lock brakes, many
people experience permanent injuries. The U.S. Department of Transportation introduced connected vehicle
technologies, which enables vehicles to “talk” to each other and exchange important data on the roads, with
the goal of preventing crashes from happening in the first place. With the rapid development of autonomous
driving technology, vehicles in the near future will be able to operate completely without human drivers,
increasing the need of reliable connected vehicle technologies. Due to the safety-critical characteristics of
autonomous vehicles, it is important to evaluate the technologies extensively prior to deployment to ensure the
safety of drivers, passengers, and pedestrians. In this paper, we evaluate the safety of Dedicated Short-Range
Communication (DSRC), which is a popular low-latency wireless communication technology specifically de-
signed for connected vehicles. We present three real-world network security attacks and conduct experiments
on real DSRC-supported modules. Our results show that DSRC is vulnerable to these dangerous attacks and
such attacks can be easily implemented by adversaries without significant resources. Based on our evaluation,
we also discuss potential countermeasures to better improve the security and safety of DSRC and connected
vehicles.
1 INTRODUCTION
With utilization of cutting-edge technologies in trans-
portation, communication, and control, roadways are
becoming connected. The connections among vehi-
cles and infrastructure enable the urban transportation
system to play a vital role in addressing mobility and
sustainability concerns. Those connections have been
introduced as vehicle-to-vehicle (V2V) and vehicle-
to-infrastructure (V2I) communication. In addition,
smart vehicle’s features such as lane departure warn-
ing, obstacle avoidance, and autonomous driving sup-
port the role of the urban transportation system to im-
prove its performance.
Autonomous driving contains ve levels based
on their degree of autonomy. Higher level of
autonomous driving requires less human involve-
ment (BMW, 2019; NHTSA, 2020). For example,
level 5 represents fully autonomous without human
controls (NHTSA, 2020). The automated driving sys-
tem overcomes distractions of human drivers; how-
ever, communication security becomes more impor-
tant. As drivers’ control will decrease along with the
level of autonomy, the automated driving system will
heavily rely on communication system.
Dedicated Short-Range Communication (DSRC)
has been adopted in smart vehicles as the most popu-
lar communication protocol. It is a low-latency wire-
less communication architecture for node-to-node
communication among hardware-enabled vehicles
and roadside equipment. DSRC includes Road Side
Units (RSUs; roadside infrastructure) and On-Board
Units (OBUs; travelling vehicles) with transceivers
and transponders. DSRC over different radio spec-
trum bands is already being used in North America,
Europe, and Japan for several applications such as
electronic toll collection (Abboud et al., 2016).
It is expected that DSRC will be more widely used
in the smart vehicles due to its growing popularity.
However, it is important to carefully investigate the
DSRC because one of the greatest security threats to
future automotive systems is DSRC itself. Therefore,
this paper studies the security of DSRC communica-
tion and seeks to determine how secure the protocol
Le, T., Elsayed-Aly, I., Jin, W., Ryu, S., Verrier, G., Al Rahat, T., Park, B. and Tian, Y.
Evaluating the Dedicated Short-range Communication for Connected Vehicles against Network Security Attacks.
DOI: 10.5220/0009355500370044
In Proceedings of the 6th International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS 2020), pages 37-44
ISBN: 978-989-758-419-0
Copyright
c
2020 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
37
is against network security attacks in practice.
In this paper, our main contributions are summa-
rized as follows:
We present three potential security attacks target-
ing the V2V and V2I communications: commu-
nication jamming, man-in-the-middle forwarding,
and false alert.
We evaluate the robustness of the DSRC protocol
for connected vehicles against these attacks.
We introduce some countermeasures to better pro-
tect DSRC against these network security attacks.
The remainder of this paper is organized as fol-
lows. Section 2 describes our literature review of re-
lated research. Section 3 presents the threat model
and the network security attacks we implement to
evaluate the DSRC protocol. Section 4 reports
our evaluation of the proposed attacks on DSRC-
supported devices. Section 5 evaluates our approach,
suggests possible countermeasures, and outlines some
future research directions. Finally, Section 6 con-
cludes this paper.
2 LITERATURE REVIEW
Throughout the transportation community, cybersecu-
rity vulnerabilities are a key concern for connected
vehicle (CV) applications. Bertini et al. (Bertini et al.,
2016) presented a survey result regarding the percep-
tion of connected and automated vehicle systems by
the Oregon Department of Transportation. Over a
hundred respondents were asked for potentials and
concerns about CV deployment in the survey. As a
result, 79% of responses were concerned about cy-
bersecurity risks where 50% of them inferred the sys-
tem is too premature to be implemented yet. The pa-
per concluded that CV applications may be beneficial
given that safe communication is guaranteed. Kaur
et al. (Kaur et al., 2016) described potential security
attacks on CV technology and classified the attacks
based on the impacts of their outcomes. They also
provided a comprehensive view of security practices
and theoretical risk assessment for the attacks.
With increasing interests in cybersecurity, re-
searchers attempted to investigate an impact of cyber-
attacks and communication errors through simulation
platform. Bhavsar et al. (Bhavsar et al., 2017) in-
vestigated the risk of communication threat to the
mixed traffic stream. The communication threat in
this paper were identified through fault tree model for
both vehicular component and infrastructure compo-
nent. Risk analysis was conducted through simula-
tion; however, the paper defined the threat as commu-
nication failure probability from reviewing the litera-
ture. Cui et al. (Cui et al., 2018) proposed a simula-
tion platform for Cooperative Adaptive Cruise Con-
trol (CACC) vehicles. The platform improved the
quality of simulation considering vehicle dynamics,
communication uncertainty, quantifying crash sever-
ity and CACC stability, and so on. Cyber-attack in
this paper was defined as randomly varying radar and
GPS sensor errors. Through sensitivity analysis, the
authors resulted GPS and radar sensor errors may cre-
ate a collision when the errors happened to be the
same direction (e.g., sensor error of positive or neg-
ative value) and GPS jamming is the most dangerous
cyber-attack. Heijden et al. (van der Heijden et al.,
2017) analyzed security attacks on CACC and their
impacts using simulation, suggesting the necessity
of mis-behavior detection along with resilient con-
trollers with graceful degradation.
When it comes to urban traffic, cybersecurity
plays an even more critical role in road safety since
street intersections are where most vehicle collisions
happen. Ernst and Michaels (Ernst and Michaels,
2017) studied the severity of cyber vulnerability on
signalized intersections. They investigated possible
threats to the traffic controller (e.g., changes on signal
phase and timing). Traffic simulation was conducted
to measure the increase in travel time along with the
threat level. Feng et al. (Feng et al., 2018) inves-
tigated vulnerabilities of traffic control system under
cyber-attacks based on falsified data. They assumed
the attacker’s goal is to maximize the system delay on
actuated and adaptive signal intersections. One of the
achievements was identifying the critical intersection
with the highest potential for heavy congestions.
In order to improve the security but not to
compromise the low-latency feature of DSRC, the
US Department of Transportation (USDOT) intro-
duced their Security Credential Management System
(SCMS). The Security Credential Management Sys-
tem (SCMS) is a proof-of-concept (PoC) message
security solution for vehicle-to-vehicle (V2V) and
vehicle-to-infrastructure (V2I) communication (US-
DOT, 2020b). This PoC uses a Public Key Infras-
tructure (PKI) approach that issues digital certificates
to authorized system participants to facilitate trusted
communication. This ensures the messages originate
from a legitimate source and guarantees the integrity
of the messages communicated. Amoozadeh et al.
(Amoozadeh et al., 2015) analyzed security vulner-
abilities of connected vehicles in the context of multi-
ple attacks whose impact was observed through sim-
ulation using VENTOS. They presented various types
of attacks including message falsification in which the
adversary listens to the wireless communication be-
VEHITS 2020 - 6th International Conference on Vehicle Technology and Intelligent Transport Systems
38
tween vehicles and manipulates the contents of the
messages. However, due to the fact that the mes-
sages are modified by the adversary, the attack is
subjected to integrity checks from defense mecha-
nisms such as SCMS. Laurendeau et al. (Lauren-
deau and Barbeau, 2006) performed an analysis on
DSRC/WAVE architecture and implied that DSRC
is potentially vulnerable towards cyber-attacks such
as replay attacks where the adversary intercepts and
retransmits malicious messages. Recent work (Le
et al., 2019) presented some attacks that adversaries
can cost-effectively launch on DSRC. However, they
did not conduct experiments to evaluate the attacks
on the DSRC protocol. Motro et al. (Motro et al.,
2016) evaluated the effectiveness of the DSRC-based
V2V communication system in VANET simulation
environment. Various DSRC characteristics including
power settings, communication range, packet errors,
sensor errors, and estimation inaccuracy were tested
to detect collision on highway. Throughout the sim-
ulation, the authors concluded that the primary factor
to be considered is communication range. The ma-
jority of communication failures were caused by the
vehicle being out of the range for the power settings.
The research provided a promising result that packet
errors at a rate of up to 50% did not have a significant
impact on collision detection. However, this research
is based on a strong assumption that their network is
secured.
State-of-the-art research studied the impacts of
cyber-attacks on transportation performance; how-
ever, their arguments are based on measurements ob-
tained through simulation platforms or probabilities
from literatures. In this paper, all experiments are
conducted with real On-Board Units (OBU) and a
Road Side Unit (RSU). Three possible cyber-attack
scenarios are presented and evaluated.
3 METHODOLOGY
DSRC is a wireless communication technology
specifically designed for high data transmission
among vehicles and infrastructures. DSRC defines
several sets of messages and fields for each message
which can be customized for V2X applications (US-
DOT, 2009). DSRC bandwidth vary by region. In
the U.S., the Federal Communications Commission
(FCC) allocated 75 MHz of spectrum in the 5.9 GHz
band for DSRC (FCC, 2019). Japan reserved 760
MHz of spectrum in the 5.8 GHz band for intelli-
gent transportation systems, in particular for DSRC
(Tachikawa, 2003). The European Union assigned 30
MHz of spectrum at 5.875 5.905 GHz for safety-
related applications and 20 MHz at 5.855 5.875
GHz for other applications (ACEA, 2018). DSRC
aims at providing a low latency protocol for Vehic-
ular Ad-Hoc Networks (VANETs) and V2X commu-
nication (Torabi and Ghahfarokhi, 2014). The main
goals for the protocol are to be reliable, fast, and safe
for both passengers and pedestrians. However, in this
work, we show that the DSRC protocol design has not
met the reliability and safety goals to be deployed in
real-world use. This section describes our approach
to evaluate the robustness of the DSRC protocol.
Threat Model. In our attack model, the adversary
has access to a vehicle with an On-Board Unit (OBU)
or to a Road Side Unit (RSU). It is reasonable to as-
sume that the adversary can have his/her vehicles or
devices with DSRC capabilities like a legitimate user
and exploit the vulnerabilities to attack others. Basi-
cally, any user can act maliciously. The goal of the
adversary is to jam the communication channels or
target other OBUs/RSUs to spread false information
among the vehicles, which will then lead to physical
damage and/or prevent other vehicles from commu-
nicating as a form of Denial-of-Service (DOS) attack.
The adversary may also position himself/herself to in-
tercept communication and relaying it giving the im-
pression that two units are in range. In this paper, we
inspect three different types of attacks targeting com-
munication layer as follows.
Communication Jamming. We explore how a ma-
licious RSU can jam the communication between two
OBUs (see Figure 1). In this scenario, the adversary
aims to disrupt communications between the victims
by decreasing the signal-to-noise ratio. This could
have very serious implications in the case where one
OBU is trying to transmit a safety critical message
and the other OBU cannot receive it properly.
Man-In-The-Middle (MITM) Forwarding. We
look into the case that a malicious RSU can trick
two OBUs into thinking they are within range of each
other by forwarding the messages (see Figure 2). This
attack would be dangerous to transportation opera-
tions such as platooning, which uses distance as a
safety measure.
False Alert. A malicious OBU that decides to
spread false information is just as dangerous as a ma-
licious RSU. In this case, the malicious OBU could
cause traffic to reroute by sending messages announc-
ing a collision alert (see Figure 3). The ability to con-
trol how traffic is rerouted is dangerous because the
Evaluating the Dedicated Short-range Communication for Connected Vehicles against Network Security Attacks
39
Figure 1: Communication Jamming Attack Scenario.
Figure 2: Man in the Middle Forwarding Attack Scenario.
victims could be rerouted through areas that are un-
safe. It is important to note that this attack considers
a malicious user sending messages instead of an at-
tacker spoofing other users’ messages. The original
messages are not modified between source and des-
tination, thus bypassing integrity checks of defense
mechanisms such as the Security Credential Manage-
ment System (SCMS) (USDOT, 2020b).
4 EVALUATION
In this section, we describe our evaluation of the pro-
posed attacks on DSRC. We used LocoMate products
made by Arada Systems, which was a leading auto-
motive technology recently acquired by Lear Corpo-
ration (Lear, 2015). In order to experiment the pro-
posed attacks, we use two LocoMate Mini 2 devices
as the OBUs and a Locomate Classic device as an
RSU. We control the RSU via Ethernet connection
and control the OBUs via Bluetooth connection. Fig-
ure 4 shows an overview of our setup.
Figure 3: False Alert Attack Scenario.
Figure 4: Experiment Setup to Evaluate Proposed Attacks
on DSRC Protocol.
4.1 Communication Jamming
In this attack, the goal is to prevent messages sent
by one OBU from reaching the other OBU (see Fig-
ure 1). The idea is that strategically placed malicious
RSUs could create issues by flooding the communi-
cation channels (in our case the control channel).
We used the RSU as a jammer to try to block the
messages going from the OBU A (Unit A) to the OBU
B (Unit B). The two OBUs were placed within 2 me-
ters from each other. At first, we attempted to manip-
ulate the power of the transmissions in order to make
the signal transmissions of the RSU much stronger
than the others but that had no effect on the drop rate
of the messages from Unit A to Unit B. Thus, we tried
increasing significantly the number of messages being
sent by the RSU in an amount of time. The default
delay between each message was 100ms. We modi-
fied the RSU’s transmission interval to 1ms, allowing
it to transmit 100 times more messages in the same
amount of time.
In our experiment, we also configured so that Unit
A and Unit C are both sending Basic Safety Message
VEHITS 2020 - 6th International Conference on Vehicle Technology and Intelligent Transport Systems
40
(BSM) but with different vehicle length information
in order to differentiate them. More specifically, the
messages being sent from Unit A to Unit B will in-
clude an additional byte C8 as shown in Figure 5.
Figure 5: Basic Safety Message Sent by Unit a with the
Additional Byte C8 in the End.
Before deploying the jamming attack (i.e., mak-
ing the modification to the transmission interval), we
calculated the message loss percentage (i.e., number
of messages that are not received out of the total of
sent messages) from Unit A to Unit B and found it to
be 19%. We first found that the change in the trans-
mission interval of the malicious RSU dramatically
increased the message loss between OBU A and B to
more than 50% when launching the attack (as can be
seen in Table 1) indicating that the RSU (i.e., the jam-
mer) can prevent a large proportion of the messages
from reaching Unit B. We further observed the mes-
sage reception rate of Unit B. The attack reduced the
reception rate (from Unit A to B) by about a factor
of 10 (10.4 msg/sec and 1.1 msg/sec respectively). In
Table 2, we show the average number of messages
per second that Unit B received from both message
sources (i.e., from Unit A and the jammer) before and
after launching the jamming attack.
4.2 Man In The Middle (MITM)
Forwarding
In this attack, the goal is to make two OBUs (Unit
A and Unit B) believe that they are within range
of each other and communicate through a malicious
RSU (Unit C) which can eavesdrop on the messages,
vehicle information and/or cause misinformation (see
Figure 2).
We first measured the maximum range in which
our two OBUs could communicate with each other.
This range is estimated to be 20 feet without line of
sight. For the experiment, we placed the two OBUs
out of their range. We then used the RSU as the “mid-
dle man” to forward the messages from one OBU to
the other. We found that the RSU was able to relay
the messages and tamper with them as they were not
encrypted or signed with security credentials. This
attack shows that it is possible to interfere with vehic-
ular applications such as platoon. In particular, this
attack can lead to unexpected merging or changes in
speed of the vehicles on the road, thus causing traffic
chaos.
4.3 False Alert
In this attack, the attacker’s goal is to exploit an OBU
(unit A) to spread false information to the other OBU
(Unit B) and the RSU (Unit C) (see Figure 3). The
success of this attack means that a malicious OBU
could reroute traffic or cause another vehicle brake
unexpectedly.
We found that a malicious actor could send mes-
sages containing false information to other units and
change the contents of the alerts and fields to arbi-
trary values. In Figure 6, we show an Intersection
Collision Alert (ICA) message being received by the
RSU (identical message received by the other OBU).
Similar to the BSM shown in Figure 5, this message
can be modified. There are also other types of mes-
sages used for critical alerts that can be sent. Those
available message types that can be exploited for a
false alert/information attack are: Intersection Colli-
sion Alert (ICA), Probe Vehicle Date (PVD), Basic
Safety Message (BSM), and Road Side Alert (RSA).
Figure 6: Intersection Collision Alert Message.
Depending on the way the connected vehicles han-
dle the reception of these alerts, the attack can be very
effective or less effective. For example, it is reason-
able to assume that when an ICA message is received,
the vehicle would certainly brake to prevent the col-
lision. Obviously, this critical message must be sent
and received with the smallest latency possible (thus
the possibility of sending it without registration to an
application). Therefore, being able to manually send
this type of message or modify it arbitrarily is dan-
gerous. For example, someone who would like their
route to be less congested could issue alerts to reroute
the other vehicles to alternative routes resulting in un-
balanced traffic and congestion. Although sensors at-
tached to the vehicle (e.g., radar) may help with iden-
Evaluating the Dedicated Short-range Communication for Connected Vehicles against Network Security Attacks
41
Table 1: Message Loss in Normal Operation and in Presence of Jamming Attack.
Messages Sent by Unit A Message Received by Unit B Message Loss
Normal Operation 393 318 19.0%
Jamming Attack 193 106 54.4%
Table 2: Message Reception Rate of Unit B before and after Jamming.
Message Source
Unit B’s Reception Rate (messages/second)
Before Jamming After Jamming
From Unit A 10.40 1.10
From The Jammer 0 12.81
tifying false alerts, researchers have shown that sen-
sors are vulnerable to spoofing attacks in which the
adversary can easily manipulate outputs of the sen-
sors (Fu and Xu, 2018).
5 DISCUSSION
We have shown that DSRC is vulnerable to the three
network security attacks. In this section, we discuss
some limitations of our work. We also describe po-
tential defenses that can be adopted to enhance the
security of DSRC. Finally, we outline some future re-
search directions to extend our work.
5.1 Limitation
5.1.1 Scalability
The scalability of our attacks needs to be further re-
searched. Our experiments were conducted using Lo-
coMate devices (as described in Section 4). There can
be other different manufacturers producing DSRC in-
frastructures and devices, which may introduce dif-
ferent variants of DSRC-supported hardware. Al-
though these variants may vary in capabilities, we fo-
cus on evaluating DSRC rather than other capabilities.
Besides, in our attack experiments, we use station-
ary hardware setup due to testing environment con-
straints. Hence, some attacks may produce different
outcomes when it comes to real vehicles in motion.
However, we believe that this does not undermine our
overall findings and insights in this paper.
5.1.2 Security Credential Management System
Our proposed attacks reasonably assumed that SCMS
was not in use because of the following reasons. First,
SCMS is still a new technology under research and
development. We tried to send an enrollment request
to the USDOT. However, the access to enroll in the
SCMS was limited to research deployment sites that
receive funding from the USDOT (USDOT, 2020b).
In addition, it is important to note that although se-
curity credentials issued by defense mechanisms such
as SCMS may protect the authenticity and integrity
of the messages, it can be difficult to defend against
malicious users who possess several legitimate cre-
dentials and exploit their own credentials to launch
attacks. In fact, assuming that units A, B, and C
have valid credentials, all the aforementioned attacks
are still possible. SCMS does not prevent users from
changing message transfer rate, relaying messages, or
sending messages that are not consistent with reality.
Although, in the case of the MITM forwarding attack,
SCMS might be able to prevent the malicious unit C
from modifying the messages in between the source
and the destination.
5.1.3 Testing on Real Vehicles
There are few vehicles actually equipped with DSRC.
Many have hardware support for it but it is not nec-
essarily supported by the car’s manufacturer software
yet. However, the USDOT has already equipped some
trucks and infrastructure elements in order to test
DSRC via pilot programs in Wyoming, Florida, and
New York (USDOT, 2020a). Also, testing security
attacks on a real vehicle in motion could be very dan-
gerous if the vehicle has some of the basic responses
to DSRC messages and alerts implemented (e.g., hard
stop while in high speed motion due to a collision
alert). For this reason, we decided to conduct our ex-
periments with stationary hardware setup instead.
5.2 Potential Countermeasures
There are two solutions we suggest to prevent net-
work congestion-based types of attack. The first
would be to set a limit for the message transmission
interval so that legitimate RSU and OBU devices can-
not be exploited for spamming messages. The sec-
ond solution is to monitor the network traffic to de-
VEHITS 2020 - 6th International Conference on Vehicle Technology and Intelligent Transport Systems
42
tect anomaly and prevent attackers from flooding the
message queue. The limitation of this defense is that
it may increase latency and it is not effective against
distributed attacks which use a large number of ex-
ploited devices. In practice, both defense mechanisms
should be combined to increase the coverage against
different types of attack.
Integrity check in defense mechanisms such as the
SCMS will be useful in providing a way of ensuring
that the messages are not modified on the way; how-
ever, it will neither prevent the attacker from relaying
the messages to an unexpected destination nor prevent
the attacker from sending originally malicious mes-
sages. One solution to prevent the MITM forward-
ing attack is to measure the time it takes at maximum
range to transmit a message. If the difference between
the sent time and the received time is too great then
that message should be marked as malicious item (i.e.,
distance bounding).
In order to prevent the false alert attack, anomaly
detection is again a great method. Once malicious be-
haviors are detected, the adversary’s credential must
be suspended. As part of the system, all vehicles need
to maintain a list of revoked credentials to avoid ad-
versaries. However, there might be a tradeoff between
security and user experience since the accuracy of the
anomaly detection algorithm matters. It is important
to have an accurate detection system to avoid losing
too much user experience.
5.3 Future Work
5.3.1 DSRC Integrated with Visual
Classification
Autonomous vehicles now are implemented with vi-
sual classification technology in the system to recog-
nize road signs and traffic lights. Methods like deep
neural networks (DNNs) are capable of doing that
with the help of sensors such as camera and lidar. Vi-
sual classification based on DNNs has been proven to
be vulnerable to adversarial machine learning exam-
ples, which would lead to dangerous attacks such as
tricking a vehicle to mistake a stop sign as a speed
limit sign. However, integrating data and the deci-
sions based on DSRC and visual classification could
help detect the adversaries and mitigate the threats. It
would be much harder for the attacker to successfully
spoof on both channels. Designing this type of inte-
grated mechanisms and testing against attacks will be
meaningful.
5.3.2 DSRC with Security Certificates Deployed
With security certificates implemented along with
DSRC, our intuition is that the jamming and the false
alert attacks can still be possible. However, after the
incident happens, it might be easier to identify the
culprit thanks to the certificates attached to the ma-
licious messages. In the MITM forwarding attack the
malicious actor will be able to forward the message,
but not modify its content. In this case, the certifi-
cates will not prevent the incident but will be useful
to trace the adversary after the attack happens. In the
false alert attack, the attacker will still be able to cre-
ate custom messages but will need to sign the message
with a valid certificate.
5.3.3 Defense Evaluation
Our proposed defenses need to be further evaluated
for each attack to validate their performance against
the attack scenarios discussed in this paper. It is also
important to evaluate the feasibility of our defenses,
considering possible interference with other critical
aspects of the protocol implementation such as la-
tency.
5.3.4 Attack Implementation in Simulation
Simulation allows us to evaluate the attacks in a bet-
ter controlled environment. In addition, we can com-
pare the results between real hardware and simulation
testing. A simulated environment also enables attack
implementations without worrying about some phys-
ical limitations, for example, we had the constraints
on power supply for the RSU when setting up the ex-
periments. Moreover, a simulated environment may
give us an opportunity to study the performance of the
attacks without having to worry about hardware com-
patibility issues. Furthermore, a comprehensive sim-
ulator that enables modifications to the DSRC imple-
mentation may help us better evaluate our proposed
countermeasures.
6 CONCLUSION
With the active growth of autonomous vehicles and
the goal of improving automotive safety, a robust
wireless communication technology used for con-
nected vehicles is very important. Due to its safety-
critical use case, this technology needs to be carefully
evaluated and tested prior to deployment for real-
world use. In this paper, we have shown that DSRC,
which is a popular wireless communication technol-
ogy designed for connected vehicles, is vulnerable to
Evaluating the Dedicated Short-range Communication for Connected Vehicles against Network Security Attacks
43
three network security attacks: jamming, MITM for-
warding, and false alert. In the near future, connected
and fully autonomous vehicles will soon replace tra-
ditional vehicles. As such, more attack surfaces will
lead to more safety issues. Therefore, it is important
to ensure that we have effective defense mechanisms
in the first place.
REFERENCES
Abboud, K., Omar, H. A., and Zhuang, W. (2016). Inter-
working of dsrc and cellular network technologies for
v2x communications: A survey. IEEE transactions on
vehicular technology, 65(12):9457–9470.
ACEA (2018). Frequency bands for v2x. https://www.
acea.be/uploads/publications/ACEA position paper-
Frequency bands for V2X.pdf.
Amoozadeh, M., Raghuramu, A., Chuah, C.-N., Ghosal, D.,
Zhang, H. M., Rowe, J., and Levitt, K. (2015). Secu-
rity vulnerabilities of connected vehicle streams and
their impact on cooperative driving. IEEE Communi-
cations Magazine, 53(6):126–132.
Bertini, R. L., Wang, H., Knudson, T., Carstens, K., and
Rios, E. (2016). Assessing state department of trans-
portation readiness for connected vehicle–cooperative
systems deployment: Oregon case study. Transporta-
tion Research Record, 2559(1):24–34.
Bhavsar, P., Das, P., Paugh, M., Dey, K., and Chowd-
hury, M. (2017). Risk analysis of autonomous vehi-
cles in mixed traffic streams. Transportation Research
Record, 2625(1):51–61.
BMW (2019). The path to autonomous driv-
ing. https://www.bmw.com/en/automotive-
life/autonomous-driving.html.
Cui, L., Hu, J., Park, B. B., and Bujanovic, P. (2018).
Development of a simulation platform for safety im-
pact analysis considering vehicle dynamics, sensor
errors, and communication latencies: assessing co-
operative adaptive cruise control under cyber attack.
Transportation research part C: emerging technolo-
gies, 97:1–22.
Ernst, J. M. and Michaels, A. J. (2017). Framework
for evaluating the severity of cybervulnerability of
a traffic cabinet. Transportation Research Record,
2619(1):55–63.
FCC (2019). Dedicated short range communications
(dsrc) service. https://www.fcc.gov/wireless/bureau-
divisions/mobility-division/dedicated-short-range-
communications-dsrc-service.
Feng, Y., Huang, S., Chen, Q. A., Liu, H. X., and Mao,
Z. M. (2018). Vulnerability of traffic control system
under cyber-attacks using falsified data. In 97th An-
nual Meeting of the Transportation Research Board.
Fu, K. and Xu, W. (2018). Risks of trusting the physics of
sensors. Communications of the ACM, 61(2):20–23.
Kaur, M., Martin, J., and Hu, H. (2016). Comprehensive
view of security practices in vehicular networks. In
2016 International Conference on Connected Vehicles
and Expo (ICCVE), pages 19–26.
Laurendeau, C. and Barbeau, M. (2006). Threats to security
in dsrc/wave. In International Conference on Ad-Hoc
Networks and Wireless, pages 266–279. Springer.
Le, T., ElSayed-Aly, I., Jin, W., Ryu, S., Verrier, G., Al Ra-
hat, T., Park, B. B., and Tian, Y. (2019). Poster:
Attack the dedicated short-range communication for
connected vehicles. Poster presented at the 40th IEEE
Symposium on Security and Privacy.
Lear (2015). Lear acquires arada systems, a leading
automotive technology company that specializes
in vehicle-to-vehicle and vehicle-to-infrastructure
communications. https://www.lear.com/Press-
Room/4279/lear-acquires-arada-systems-a-leading-
automotive-technology-company-that-special.aspx.
Motro, M., Chu, A., Choi, J., Lavieri, P. S., Pinjari, A. R.,
Bhat, C. R., Ghosh, J., and Heath, R. W. (2016). Ve-
hicular ad-hoc network simulations of overtaking ma-
neuvers on two-lane rural highways. Transportation
Research Part C: Emerging Technologies, 72:60–76.
NHTSA (2020). Automated vehicles for safety. https:
//www.nhtsa.gov/technology-innovation/automated-
vehicles-safety.
Tachikawa, K. (2003). Global standards on its ra-
dio communications. https://itsforum.gr.jp/Public/
E4Meetings/P01/tachikawa5
5 3.pdf.
Torabi, N. and Ghahfarokhi, B. S. (2014). Implementa-
tion of the ieee 802.11 p/1609.4 dsrc/wave in ns-2.
In 2014 4th International Conference on Computer
and Knowledge Engineering (ICCKE), pages 519–
524. IEEE.
USDOT (2009). Standard specification for dedicated short
range communication (dsrc) physical layer using mi-
crowave in the 902-928 mhz band. https://www.
standards.its.dot.gov/Factsheets/Factsheet/5.
USDOT (2020a). Connected vehicle pilot deployment pro-
gram. https://www.its.dot.gov/pilots.
USDOT (2020b). Security credential management system.
https://www.its.dot.gov/resources/scms.htm.
van der Heijden, R., Lukaseder, T., and Kargl, F. (2017).
Analyzing attacks on cooperative adaptive cruise con-
trol (cacc). In 2017 IEEE Vehicular Networking Con-
ference (VNC), pages 45–52. IEEE.
VEHITS 2020 - 6th International Conference on Vehicle Technology and Intelligent Transport Systems
44