sensors

Article
Smart Road Traffic Accidents Reduction Strategy
Based on Intelligent Transportation Systems (TARS)
Abdulaziz Aldegheishem 1, *, Humera Yasmeen 2 , Hafsa Maryam 2 , Munam Ali Shah 2 ID
,
Amjad Mehmood 3 ID , Nabil Alrajeh 4 and Houbing Song 5 ID
1 Traffic Safety Technologies Chair, Urban Planning Department, College of Architecture and Planning,
King Saud University, Riyadh 11574, Saudi Arabia
2 Department of Computer Science, COMSATS University Islamabad, Islamabad 45550, Pakistan;
humera_9@ymail.com (H.Y.); hafsa.maryam09@gmail.com (H.M.); mshah@comsats.edu.pk (M.A.S.)
3 Institute of IT, Kohat University of Science & Technology, Kohat KP 26000, Pakistan;
dramjad.mehmood@ieee.org
4 Biomedical Technology Department, College of Applied Medical Sciences, King Saud University,
Riyadh 11633, Saudi Arabia; nabil@ksu.edu.sa
5 Department of Electrical, Computer, Software, and Systems Engineering, Embry-Riddle Aeronautical
University, Daytona Beach, FL 32114, USA; h.song@ieee.org
* Correspondence: aldeghei@ksu.edu.sa




Received: 4 May 2018; Accepted: 12 June 2018; Published: 21 June 2018


Abstract: Traffic accidents have become an important problem for governments, researchers and
vehicle manufacturers over the last few decades. However, accidents are unfortunate and frequently
occur on the road and cause death, damage to infrastructure, and health injuries. Therefore, there
is a need to develop a protocol to avoid or prevent traffic accidents at the extreme level in order
to reduce human loss. The aim of this research is to develop a new protocol, named as the Traffic
Accidents Reduction Strategy (TARS), for Vehicular Ad-hoc NETworks (VANETs) to minimize the
number of road accidents, decrease the death rate caused by road accidents, and for the successful
deployment of the Intelligent Transportation System (ITS). We have run multiple simulations and the
results showed that our proposed scheme has outperformed DBSR and POVRP routing protocols in
terms of the Message Delivery Ratio (MDR), Message Loss Ratio (MLR), Average Delay, and Basic
Safety Message.

Keywords: Intelligent Transportation System; accident; reduction; VANETs

1. Introduction
Over the past few decades, the number of vehicles has significantly increased which has resulted in
populated roads. This massive traffic load has uplifted the road accident graph, which has consequently
raised injuries and death rates worldwide. In 2010, the states presented in the World Health Report
(WHR) documented road traffic injuries as the ninth most common reason for the disability of people.
According to the report by the World Health Organization (WHO) [1], about 1.24 million casualties
and 50 million injuries were reported globally. Automobile accidents are rated as the eighth cause of
injuries and death toll worldwide. In the KSA, compared to other developed countries like the UK
and USA, the death toll caused by road accidents has increased from seventeen percent to twenty-four
percent per 10 million people over a decade [2] and is the primary reason for a young person’s death [3].
Friends of the Red Crescent Committee reported 526,000 accidents yearly in KSA with approximately
seventeen casualties per day. According to the report published in the Aleqtisadiah newspaper, there
were 2.8% more car accidents in 2016 than those reported in 2015. On the basis of these statistics, it is

Sensors 2018, 18, 1983; doi:10.3390/s18071983 www.mdpi.com/journal/sensors

Sensors 2018, 18, 1983 2 of 23

estimated that the death toll will further grow by 2020 due to road accidents [4,5]. Attention must be
given towards this severe road traffic condition in the KSA to control this, otherwise it will probably
increase to four million crashes per year by 2030 according to the prediction made in [6].
These road traffic accidents not only cause an increased death toll or injury, but also result in
communal and financial damage to the state [7]. The deaths or disabilities of youngsters caused by
accidents have a solemn effect on families and more widely on society [8]. The loss or disability of an
earner not only causes financial suffering for family members, but also changes the domestic dynamics.
On the other hand, the state expends SR 13 billion approximately per year to matters associated with
road accidents [9]. Therefore, it is essential to find a way to avoid such severe conditions so that the
state can continue to prosper in terms of productivity, as well as socially and economically.
Numerous efforts have been made to avoid such drastic situations through the implications of
speed limits, imposition of traffic rules, deployment of seat belts and air bags, boosting stiffness into the
physical structure of a vehicle, and so on [10]. Furthermore, researchers have directed their attention
to avoiding accidents and have introduced a number of active electronic and computer-controlled
mechanisms such as braking systems, intelligent speed adaptation, collision prevention, self-governing
cruise control systems, electronic stability control systems, and so on [11]. In recent times, Intelligent
Transportation Systems (ITS) have gained much attention from the research community [12]. ITS
introduced smart vehicles that are enabled to wirelessly communicate with one another through a
communication device called an On-Board Unit (OBU). A network of inter-connected vehicles equipped
with Electronic Control Modules (ECMs) are called Vehicular Ad-hoc NETworks (VANETs) [13].
Vehicular Networks are the most important and emerging technology in the field of Intelligent
Transportation Systems (ITS). A large number of new routing schemes and architectures have been
suggested in recent years in order to successfully deploy ITS. The widespread deployment of ITS
is a complex and challenging task. Numerous projects have been developed and their performance
evaluated either through simulation or deployment in a real environment. Various projects developed
by different countries include the project related with Automobile Safety Communication in the United
State of America (USA) and Car-to-Car communication in the European Union. The European Union,
USA, and Japan have already deployed a smart vehicular environment in both urban and highway
areas. Many car manufacturing companies like Ford, Daimler, BMW, General Motors, and Audi have
been inspired by the ITS and are motivated to manufacture smart vehicles for passenger safety.
In VANETs, the topology of the network keeps changing due to the high mobility of vehicles,
so the network is considered as an ad-hoc network. It enables vehicles to sense their environment
and exchange their sensed data with surrounding vehicles. An infrastructure named Road Side
Units (RSU) is installed along the roads to assist the vehicles moving in its vicinity. Standard IEEE
802.11p [14] has been introduced, particularly for vehicular communication, which allows ad-hoc
communication (the p denotes the specific version for communication between vehicles). There are
three modes of communication in VANETs: Vehicle to Infrastructure (V2I), Vehicle to Vehicle (V2V),
and Vehicle 2 Hybrid (V2X) communication. Vehicles communicate with each other in order to get a
better understanding of the surrounding environment to prevent any hazardous situations. Drivers
must be given timely warnings about any expected hazardous situations in order to avoid accidents.
Another concern is that drivers become doubtful as to which route to follow next even if warned
about emergency situations, which can cause inefficient traffic flow and congestion on alternate
routes [15]. Traffic safety and smooth flow can be achieved by assisting the driver with appropriate
suggestions. The aim of this project was to design and develop a possible accident detection strategy
that provides timely warnings to the driver about possible accident situations and gives appropriate
suggestions accordingly.
The objectives of this research were focused on the following aspects:

(1) The design of a new protocol for accident prevention and, hence, reduction.
(2) To forecast/guess the probability of the occurrence of an accident in advance before it occurs.
(3) To re-route vehicle traffic to prevent traffic jams on the road that may cause accidents.
Sensors 2018, 18, 1983 3 of 23

(4) To maintain traffic flow efficiently.

(5) To assist drivers to reach a destination on time.
(6) To reduce safety messages broadcasting in order to avoid a broadcasting storm and
network congestion.
(7) To minimize the delay in re-routing of traffic to other available paths.

The roadmap of this research is arranged as follows. An overview of the existing protocols
proposed to avoid traffic accidents is presented in Section 2. In Section 3, a smart road traffic accident
reduction strategy is presented in detail. The simulation parameters of the proposed strategy are
presented in Section 4. The simulation results of the proposed scheme are discussed and evaluated in
Section 5. Finally, the research work conclusions are presented in Section 6.

2. Literature Review
This section describes a list of the extensive research conducted to avoid road accidents at the
maximum level with a specific focus on the vehicular environment in order to reduce human loss.
The Literature Review section is further divided into two sections for each category of traffic accident
avoidance and prevention schemes, based on routing schemes and warning messages as shown in
Figure 1.

2.1. Traffic Accident Avoidance and Prevention Based on Warning Messages

Lozano et al. [16] proposed the warning message scheme to avoid traffic accidents between
vehicles by sending a warning message to alert drivers about the condition of the current accident.
The proposed scheme used the distance-based flooding scheme. The author calculated the time of
the vehicle’s reaction after the occurrence of the accident in order to prevent further traffic accidents.
Furthermore, a warning message dissemination scheme has been introduced for low priority messages
in order to achieve efficient bandwidth utilization. The proposed routing scheme achieved the best high
and low priority message dissemination in drastic weather conditions, i.e., rain, sun, etc. Furthermore,
the routing scheme achieved less delay and an efficient utilization of the bandwidth under different
traffic conditions. However, the limitation of this routing scheme is that it does not work in a highly
dense vehicular environment. Another limitation of this scheme is the highest priority message that
the author used in the simple flooding scheme. So, the time utilized in taking the forwarding decision
was less. Therefore, the proposed scheme faces scalability issues related to the dissemination of high
priority messages.
Gokulakrishnan et al. [17] proposed an accident avoidance routing scheme named Road Accident
Prevention (RAP). This scheme introduced the Early Warning (EW) message in order to make essential
decisions—selecting alternate routes, slowing down the vehicle, and changing lanes. Furthermore, the
Road Side Unit (RSU) detects any unusual activity in the highway scenario and broadcasts an Early
Warning (EW) message to all the vehicles in the range of the RSU. In this way, the EW message must
be sent at the exact time to prevent the road accident. The authors introduced different types of risk
zone, namely, high, average, and low risk zones. The high-risk zone comprises those vehicles that are
nearest to vehicles that are involved in road accidents. The routing scheme successfully achieved less
delay. However, the limitation of the proposed scheme was the high network processing overhead due
to the extra warning message transmitted, the EW message and identification of the risk zone.
Dawood et al. [18] proposed an accident avoidance scheme named the Efficient Emergency
Message Broadcasting (EEMB) routing scheme. The objective of the proposed scheme was to reduce
traffic bottlenecks and prevent multiple road accidents by broadcasting emergency messages with low
overheads at high velocity. After an accident has occurred between two vehicles, the affected vehicle
will select the best forwarder and broadcast the emergency message until the emergency message
covers the whole risk zone. The forwarding scheme achieved the minimal overhead caused by the
beacon message by using the method of the early prediction system. The beauty of this forwarding
Sensors 2018, 18, 1983 4 of 23

scheme is that it overcomes the problem of network fragmentation by using the mechanisms of
store-carry and forward.
Benslimance et al. [19] considered that the consumer with the emergency message must
disseminate the message intermittently until it selects another forwarder vehicle. However, this
dissemination causes a duplication of the broadcasting message due to multiple broadcasting attempts
that do not help the vehicle to obtain another vehicle. Therefore, the network of the traffic will be
affected from the processing overhead. In order to minimize the processing overhead, the author
proposed a scheme named Optimized Dissemination of Alarm Message (ODAM) while restricting the
regions and forwarder vehicles. The ODAM routing scheme efficiently achieves less overhead due
to the periodical broadcasting message. However, the ODAM scheme experiences a high delay by
sending the emergency messages to the forwarder vehicle.
Roy et al. [20] proposed accident avoidance and congestion prevention in a vehicular environment.
In this scheme, after the accident occurs, the affected vehicle generates the emergency message. The
RSU receives the emergency message and then retransmits a further emergency message to the other
RSUs which lie in its range. The beauty of this routing scheme is that it prevents the duplication of an
emergency message. Before receiving the emergency message, the RSU will check whether the same
emergency message has been received or not. In the case of not being received, the RSU will accept the
emergency message, otherwise not. This routing scheme has successfully achieved less delay, a high
delivery ratio, and maximum throughput.

2.2. Traffic Accident Avoidance and Prevention Based on Routing Scheme

Kumar et al. [21] proposed a Distance-Based routing scheme. The basic idea behind this routing
scheme is to avoid the accident at the intersection as, most of the time, anonymous vehicles appear
from the other side of the road and cause accidents at the intersection. This scheme for accident
prevention starts by obtaining the vehicle location and calculating the distance of the vehicles near
to the intersection. After calculating the location of each vehicle, all vehicles will obtain information
on the location of other vehicles coming towards the intersection. As a result, the vehicle with the
shortest distance will send an alert to other vehicles about its distance and location. The proposed
scheme successfully achieved less traffic congestion, which helps to prevent or reduce traffic accidents.
However, the scope of this proposed routing scheme was limited and faces scalability issues. Moreover,
it only works for an intersection, where three different roads meet/join at a single location, and a
ring road.
Kshirsagar et al. [22,23] proposed an intelligent vehicular management scheme to prevent traffic
accidents. The proposed scheme used basic warning safety messages to prevent traffic congestion. The
traffic signal and management scheme successfully achieved high throughput, a high delivery ratio,
and less delay.
Nzouonta et al. [24] proposed a scheme named Spatio-Temporal Emergency Information
Dissemination (STEID). The goal of this routing scheme is to suggest a hybrid network architecture
by using cellular links and proxy servers. The STEID routing protocol achieved a maximum ratio
of successfully delivered alert emergency messages, less traffic load, and fewer delays related to the
message reception rate. However, the STEID routing scheme is costly as a large-scale infrastructure is
needed in order to cover a long path. Furthermore, the hybrid network architecture does not match
the requirements related to traffic safety applications. This generally requires V2V communication to
be efficient for delivering the data at the lowest cost [25,26].
Devdhara et al. [27] considered a traffic avoidance routing protocol named Inter-Vehicle Collision
(IVC). All vehicles in the cluster broadcast a secure warning message to give further information to the
other vehicles, for example, a traffic jam [28]. This scheme only detects collisions when the vehicle
changes its path. In order to ensure IVC performance, the author used methods such as K-Mean
Clustering and Agglomerative Hierarchical Clustering (AHC) [28]. The AHC algorithms had an
Sensors 2018, 18, 1983 5 of 23

outstanding performance when compared to K-Mean Clustering in order to detect accidents that lie in
the identical cluster as well as to send data about the accident to those vehicles which lie in the cluster.
Bhumkar et al. [29] proposed a driver fatigue detection scheme by using real time sensors. In
the proposed scheme, the driver’s fatigue will be detected immediately; if the driver is found to have
drunk alcohol, they are warned through a notification message and the ignition is turned off. In this
way, the possibility of traffic accidents is avoided. The fatigue detection scheme has an outstanding
performance in human behavior detection in vehicles.
Nzouonta et al. [30] considered a routing scheme named Road Based Vehicular Traffic (RBVT).
The RBVT scheme uses real time information based on the vehicular environment to create a
road-based intersection with network connectivity and high probability among vehicles. The proposed
routing scheme uses the geographical forwarding scheme to transmit interest packets between road
intersections on the route. The RBVT achieved the average packet delivery ratio and average delay
due to its traffic overhead.
Manoj et al. [31] proposed a congestion detection algorithm to avoid traffic accidents caused
by traffic congestion. After detecting traffic congestion, the drivers of the vehicles provide multiple
options about the magnitude and location to avoid getting stuck in the traffic congestion. In Vehicle
2 Infrastructure (V2I) communication, after detecting traffic congestion in the lane due to a traffic
accident, the affected vehicle, which is involved in the traffic accident, broadcasts the warning message
to the other forwarder vehicle and RSU to inform them of the present condition of the lane. In this
way, the next upcoming vehicle changes their decision to avoid traffic congestion and road accidents.
This scheme achieved efficient bandwidth utilization and minimum message overhead.
Khatri et al. [32] proposed a traffic congestion detection routing scheme to avoid traffic accidents.
The beauty of this proposed scheme is that it depends on the data collection and central network
infrastructure. In data collision, data are gathered from the real-time environment. Vehicles are
equipped with GPS to communicate with other vehicles. Collected data only become beneficial when it
is shared with other vehicles, including velocity and present location. When any vehicle broadcasts the
message to the path, the forwarder vehicle collects the congestion message and then checks whether
the area is congested or not. If not, then the vehicle stores this message in its own memory. Otherwise,
it simply shares the congestion location with the other vehicles. The proposed scheme achieved less
transmission overhead and efficient utilization of the bandwidth. However, this scheme only worked
in a homogenous vehicular environment, but was not tested in a heterogeneous vehicular environment.

2.3. Discussion
In this section, we discussed the traffic accident prevention and avoidance scheme with a specific
focus on VANETs. We evaluated research papers from 2011–2017. In Tables 1 and 2, we provide a
concise and detailed review of different road accident and prevention schemes based on warning
messages and routing schemes. It can be observed that road congestion and traffic accidents cause
big trouble after an accident happens. In the accident area, vehicles get congested due to a lack of
awareness about the road accident. Furthermore, vehicles coming from the road intersections cause
traffic jamming. The scenario affects the routine of human life. To address the above problems there is
a need to develop a routing protocol in order to avoid the road accidents as well as traffic congestion in
VANETs’ environment. Furthermore, the protocol must warn the vehicle drivers promptly about any
expected hazardous situations in order to avoid accidents. The protocol also gives an alternate route
to drivers in order to follow the next shortest path, even if warned about the emergency situation, to
overcome the traffic congestion situation, otherwise inefficient traffic flow and congestion at different
routes will result [15].
Sensors 2018, 18, 1983 6 of 23

Table 1. Literature review of traffic accident prevention and avoidance schemes based on Safety/Warning Messages.

Proposed Scheme Scalability Simulation

Authors Proposed Scheme Methodology Issues Identified Benefits
Name Issue Tool
The warning message scheme has been
introduced for low priority messages in order
Drivers’ reaction time when
Warning Message to achieve efficient bandwidth utilization. Less delay efficient
Garcia-Lozano et al. [16] some situations occur. Yes NS-2.34
routing scheme The scheme achieves the best high and low utilization of the bandwidth
Delay in message broadcasting
priority message dissemination in drastic
weather conditions
When RSU detects any unusual activity,
broadcast EM message to all those vehicles
Road Accident Timeliness of alert messages
Gokulakrishnan et al. [17] which are lying in the range of the RSU. Less delay No NS-2.0
Prevention (RAP). Delay in message broadcasting
EW message must be sent at the exact period
to prevent road accidents.
The affected vehicle selects the best forwarder
and broadcasts the emergency message.
The forwarder vehicle rebroadcasts the
The Collision probability to
Efficient Emergency emergency message to the RSU. MATLAB,
evaluate the communication
Salman Dawood et al. [18] Message Broadcasting This procedure continues until the emergency Less message overhead No R2011b
in VANETs.
(EEMB) message covers the whole risk zone. Version
Timeliness of alert messages
The proposed scheme overcome the problem
of network fragmentation by using store-carry
and forward.
Optimized
Vehicles which have the alarm message must Detection of the Traffic
Dissemination of Less overhead
Benslimane et al. [19] broadcast the message until it selects another Congestion No MATLAB
Alarm Message Less end-to-end delay
forwarder vehicle. Flooding
(ODAM)
An RSU broadcasts the emergency message to
the other RSUs which lies in its range. Distribution of the capacity of
Traffic congestion High throughput
This prevents the duplication of an emergency the bandwidth.
Roy et al. [20] detection and Less end-to-end delay No NS-2.0
message. Broadcasting storm problem
avoidance scheme High packet delivery ratio
These messages help the driver to select Detection of Traffic congestion.
another route to avoid accidents.
Sensors 2018, 18, 1983 7 of 23

Table 2. Literature review of traffic accident prevention and avoidance schemes based on Routing Schemes.

Proposed Scheme Scalability

Authors Proposed Scheme Methodology Issues Identified Benefits Simulation Tool
Name Issue
The scheme obtains the vehicle location and
calculates the distance of the vehicle that is near to
the intersection.
Possibilities presented at the Less traffic congestion
Distance Based After calculating the location, the vehicle with the
Rajesh-Kumar et al. [21] intersection. Maximum Packet Yes NS-2.34
routing scheme shortest distance to the intersection will send an alert
Timeliness of alert messages Delivery Ratio.
to the other vehicles about its distance and location.
All other vehicles will be alerted about the upcoming
vehicle and are protected from the traffic accident
Spatio-Temporal The routing scheme satisfies both temporal and Maximum delivery
Emergency spatial reliability by guaranteeing the delivery of an ratio of alert messages
Nzouonta et al. [24] Information alert message in a short interval Delay in message broadcasting Less traffic load No NS-2.29
Dissemination The message passes to all the vehicles that are passing Less delay w.r.t
(STEID) through the zone during the lifetime of an emergency. message reception rate
All vehicles in the cluster broadcast secure the SUMO
message to give further information to the other OMNET++
Inter-Vehicle Cluster size
vehicles. VEINS
Devdhara et al. [27,28] Collision (IVC) Simple Flooding Driver reaction time No
A traffic accident which lies in the same cluster is Clustering
scheme Transmission range
detected and information about the accident is sent to Algorithm
the other vehicles in the cluster. Techniques
The proposed scheme used real time sensors to detect ARM7
the driver’s fatigue immediately. MQ-3 gas sensor
Driver fatigue Upon detection, it warns through a notification Human behavior causes traffic Driver reaction time GPS Smart
Bhumkar et al. [29] No
detection scheme message and then turns the ignition off. accidents Transmission range Receiver
The scheme performs very well on human behavior Google Map API
like drinking alcohol. Visual Basic (VB)
The proposed routing scheme used the geographical
Possibilities use the presented Average packet
Road Based forwarding scheme in order to transmit interest NS-2.30
intersection. delivery ratio
Vehicular Traffic packets between road intersections on the route. SUMO
Nzouonta et al. [30] The collision probability is used High Average delay No
(RBVT) routing The proposed scheme performs very well on the DCF standard
to evaluate the communication due to its traffic
scheme intersections. with IEEE 802.11p
in VANETs. overhead.
AODV protocol is used to discover the route.
After the detection of traffic congestion, the drivers of
the vehicles provide magnitude and location of the
Distribution of the capacity of Efficient bandwidth
Traffic congestion vehicle.
the bandwidth. utilization Net Beans IDE 7.0
Manoj et al. [31] detection and The affected vehicle, which is involved in the traffic No
Detection of the Traffic Minimum message Java
avoidance scheme accident, broadcasts the warning message to the RSU.
Congestion. overhead
The RSU further rebroadcasts the warning message to
make a decision on an alternate route.
The best forwarder vehicle collects the congestion
message. Distribution of the capacity of Less transmission
Traffic congestion
It checks if the area is congested or not. If not, then the bandwidth. overhead NS-2.0
Khatri et al. [32] detection and No
the vehicle stores this message in its own memory. Detection of the Traffic Efficient utilization of SUMO
avoidance scheme
Otherwise, it simply shares the congestion location Congestion. the bandwidth
with the other vehicles [33].
Sensors 18, 1983
2018,2018,
Sensors 18, x FOR PEER REVIEW 8 of 23
9 of 23

Figure 1. Taxonomy of Literature Review.

Figure 1. Taxonomy of Literature Review.
3. Smart Road Traffic Accidents Reduction Strategy
3. Smart Road Traffic Accidents Reduction Strategy
In this section, we propose a strategy named as the Traffic Accidents Reduction Strategy (TARS)
to
Inavoid accidents
this section, weand to prevent
propose congestion
a strategy named inasVANETs. In the
the Traffic proposed
Accidents strategy, Strategy
Reduction the warning
(TARS)
message is transmitted in time; therefore, potential vehicle collisions are avoided.
to avoid accidents and to prevent congestion in VANETs. In the proposed strategy, the warning Moreover, vehicles
are assisted
message to another
is transmitted inroute.
time; In this section,
therefore, first, anvehicle
potential introduction to some
collisions are of the participating
avoided. Moreover, entities
vehicles
is presented. Second, a few assumptions were made for the smooth execution of our
are assisted to another route. In this section, first, an introduction to some of the participating entities proposed
strategy. Then, phases of Traffic Accidents Reduction Strategy (TARS) are described in detail.
is presented. Second, a few assumptions were made for the smooth execution of our proposed strategy.
Following are the entities involved in the execution of the TARS.
Then, phases of Traffic Accidents Reduction Strategy (TARS) are described in detail. Following are the
entities
3.1. involved
Entities in the execution of the TARS.
The entities involved in the traffic accident reduction strategy are as follows:
3.1. Entities
(1)
The Smart Vehicles
entities involved in the traffic accident reduction strategy are as follows:
(2) Road Side Units (RSUs)
(1) (3)
Smart Vehicles
Trusted Authority (TA)
(2) (4)
RoadGovernmental Authority (GA).
Side Units (RSUs)
(3) (1)
Trusted
SmartAuthority
Vehicles (TA)
(4) Governmental Authority (GA).
In VANETs, a communicating device called the On-Board-Unit (OBU) is installed in each vehicle
(1) Smart allows
which Vehiclesit to communicate with other entities in the system. Vehicles communicate with other
vehicles and the RSU through the OBU to obtain a better awareness of their surroundings in order to
In VANETs,
react a communicating
in a timely device called
manner to any unusual thefor
situation, On-Board-Unit (OBU)
example, accident, is installed
traffic in blockage,
jam, road each vehicle
which
andallows
so on. it to communicate with other entities in the system. Vehicles communicate with other
vehicles and the RSU through the OBU to obtain a better awareness of their surroundings in order to
(2) Road Side Units (RSUs)
react in a timely manner to any unusual situation, for example, accident, traffic jam, road blockage,
and so on.The RSU keeps the information of the automobiles (vehicles) moving in its transmission range
so that it has a global view of the traffic. This feature enables the RSU to maintain the efficient traffic
Roadand,
(2) flow Sidehence,
Units accidents
(RSUs) are also avoided.
The RSU keeps the information of the automobiles (vehicles) moving in its transmission range so
that it has a global view of the traffic. This feature enables the RSU to maintain the efficient traffic flow
and, hence, accidents are also avoided.
Sensors 2018, 18, 1983 9 of 23

(3) Trusted Authority (TA)

The Trusted Authority (TA) is a trusted administrative authority. Both vehicles and RSUs are
registered with the TA. The TA assigns unique identities to both the RSUs and vehicles. The TA has a
record of all the registered entities so that they remain liable to it.
(4) Governmental Authority (GA)
The Governmental Authority (GA) plays the role of a passive entity. The passive entity is not
directly involved in the execution of TARS; however, it plays a vital role if a non-registered vehicle is
encountered so that it should not affect the security of the VANET system at all.

3.2. Assumptions
For the smooth and efficient flow of our proposed strategy, we made the following assumptions:

1. All vehicles must be equipped or armed with a Global Positioning System (GPS) in order to
obtain the exact location so that it can be used in the Basic Safety Message (BSM).
2. The RSU is an uncompromised entity. An uncompromised entity means that it can neither be
effected by any attack nor can its record can be modified.
3. All vehicles and RSUs are registered with the TA.
4. The Vehicle to Infrastructure (V2I) communication mode is considered.

3.3. Phases of Traffic Accidents Reduction Strategy (TARS)

The Traffic Accidents Reduction Strategy (TARS) is divided into three phases. The first is the setup
phase, which describes the process of establishing the VANET according to our proposed strategy; the
authorization phase, the second phase of TARS, demonstrates the registration process of the entities.
Finally, the execution of the proposed strategy is presented in the execution phase. We discuss these
three phases in detail.
(1) Setup Phase
In this phase, the entities of VANETs are set up for the TARS scheme. A RSU periodically
broadcasts its identity and location in its vicinity so that vehicles moving in its range can communicate
with it. A vehicle broadcasts BSM containing VID, location, speed and direction to the RSU when
it enters into the range of RSU and listens to RSU’s beacons. The RSU keeps the information of the
automobiles (vehicles) travelling in its vicinity through the information received in the BSMs. The RSU
data include the vehicle-ID, their velocity (V), location, and Tour-Span (TS). Tour-Span is the minimum
time a vehicle spends in the range of a RSU. On reception of the BSM, the RSU extracts the vehicle-ID
and verifies it by looking it up in its database. After verifying the vehicle-ID, the RSU calculates the
tour-span ‘TS’ of the vehicle in its region as shown in Equation (1).

Flocation RSU − Positioni

TSi = (1)
Velocityi

where FlocationRSU is the end point of the transmission range of the RSU; Positioni is the current location
of a vehicle received in the BSM; and Velocityi is the velocity of vehicle i.
(2) Authorization Phase
This phase describes the procedure of registering the RSUs and vehicles with a Trusted Authority
(TA). The TA is a trusted administrative authority that is responsible for the accountability of entities
of the VANET system. Once the entities, RSUs, and vehicles become registered with the TA, then they
remain liable to it. During the registration process, the TA provides a key and a vehicle ID to a vehicle
as shown in Figure 2. A vehicle owner must keep this key in a safe place, for example, at home or at the
bank, so that the owner can claim ownership of a vehicle if a vehicle becomes physically compromised.
Sensors 2018, 18, 1983 10 of 23

Then, the vehicle credentials can be revoked and any possible attacks, for example, misleading attack,
can be avoided. The TA hands out the identities of registered vehicles to the RSUs installed at that
Sensors 2018, 18, x FOR PEER REVIEW 11 of 23
region as people often travel in the region they live in. A question may arise as to what will happen if
aasvehicle
to what registered in another
will happen region
if a vehicle arrives/enters
registered into region
in another the transmission range
arrives/enters intoofthe
a RSU. For this
transmission
purpose, we present different scenarios in the next subsection.
range of a RSU. For this purpose, we present different scenarios in the next subsection.

Figure 2. Authorization Phase.

Figure 2. Authorization Phase.
(3) Execution Phase
(3) Execution Phase
The TARS focuses on the use of the distance measurement approach for warning message
The TARS
generation focusesaccidents
to prevent on the use of the
along withdistance measurement
the efficient reroutingapproach
of traffic. forAn warning
illustration message
of the
generation
proposed technique is shown in Figure 3. The RSU broadcasts the beacon message containingthe
to prevent accidents along with the efficient rerouting of traffic. An illustration of its
proposed
identity and technique
locationisinshownits range in soFigure 3. The RSU
that vehicles moving broadcasts the beacon
in its vicinity message containing
can communicate its
with it. The
identity and location
vehicle broadcasts theinBasic
its range
Safety so that vehicles
Message (BSM)moving
to theinRSU its vicinity
when itcan communicate
listens to the beacon withmessage
it. The
vehicle broadcasts the Basic Safety Message (BSM) to the RSU when
of a RSU. The BSM consists of the vehicle id (Vehicle-ID), location, speed, and direction. The RSU it listens to the beacon message
of a RSU. The
maintains a map BSM consiststhe
between ofvehicle
the vehicle id (Vehicle-ID),
identity and locationlocation, speed, moving
of each vehicle and direction. The RSU
in its range. The
maintains
RSU addsathe mapinformation
between theabout vehicle identityaccording
a vehicle and location of each
to the vehicle
latest BSM moving
receivedinfrom its range. The
a vehicle.
RSU addsthen
Vehicles the information
broadcast the about
BSM a vehicle
every taccording
milliseconds. to the latest
The RSU BSM received
updates thefrom
record a vehicle. Vehicles
at the reception
then broadcast
of each BSM. Ifthe BSM every
a distance t milliseconds.
between The RSU moving
the two vehicles, updates on thesame
recordroadat the reception
lane, of each
is less than theBSM.
safe
Ifdistance
a distance(Dv)between
then thethe two vehicles,
leading moving
vehicle sends on same
warning road lane,
message to theis following
less than the safe distance
vehicle. (Dv )
On the other
then
hand, thethe
leading
RSU also vehicle sends warning
monitors message
the distance (D) toandthevelocity
following vehicle.i) On
(Velocity the other
of each hand,It the
vehicle. RSUa
sends
also monitors
warning messagethe distance
to warn(D) thatand velocity
vehicle. Dv (Velocity
is calculatedi ) of each
using vehicle.
the It
method sends a warning
presented in message
Section to
3.4.
warn that vehicle.
Whenever the distance D v is calculated using the method presented in Section
between the two vehicles is less than Dv, the RSU sends a warning message 3.4. Whenever the distance
between the twovehicle
to the following vehiclestoiskeep less the
than Dvdistance.
safe , the RSUIfsends a warning
in a certain timemessage
period (Tto p),the
thefollowing
RSU has sent vehicle
Talert
to keep thethen
messages, safeitdistance.
means there If inisaacertain time load
high traffic period (Tp ),area.
in that the RSU has sent
The values Talert messages,
considered then
for Tp and it
Talert
means
were 10 there
s and is a4,high traffic load
respectively. Thesein that area.were
values The taken
valuesfrom considered for Tp and the
[34]. Afterwards, TalertRSUwere 10 s aand
sends high4,
respectively.
traffic congestion Thesemessage
values were to the taken from [34].
previous Afterwards,
neighboring RSUthe RSU sends
to reroute the atraffic.
high traffic congestion
The neighboring
message
RSU willto the previous
reroute the upcoming neighboring
vehicles RSU to reroute
as shown the traffic.1. The
in Algorithm neighboring
In case vehicles are RSU will reroute
moving outside
the
the upcoming
range of the vehicles
RSU, and as shown in Algorithm
the distance between 1. the
In case
twovehicles
vehiclesare moving
becomes outside
less than D the
v atrange
someoftime,
the
RSU, and thevehicle
the leading distance between
sends the two
a warning vehiclestobecomes
message the followingless than Dv atto
vehicle some
keeptime,
a safethedistance.
leading vehicle
sends a warning message to the following vehicle to keep a safe distance.
Meanwhile, it also reports it to the previously connected RSU through neighboring vehicles.
Neighboring vehicles only act as relay nodes and they pass on the message to the trailing vehicles until
it reaches that particular RSU.

Figure 3. Illustration of Traffic Accidents Reduction Strategy (TARS).

 to the following vehicle to keep the safe distance. If in a certain time period (Tp), the RSU has sent Talert
messages, then it means there is a high traffic load in that area. The values considered for Tp and Talert
were 10 s and 4, respectively. These values were taken from [34]. Afterwards, the RSU sends a high
traffic congestion message to the previous neighboring RSU to reroute the traffic. The neighboring
RSU will reroute the upcoming vehicles as shown in Algorithm 1. In case vehicles are moving outside
Sensors
the2018,
range18, of
1983
the RSU, and the distance between the two vehicles becomes less than Dv at some time, 11 of 23
the leading vehicle sends a warning message to the following vehicle to keep a safe distance.

Sensors 2018, 18, x FOR PEER REVIEW 12 of 24

Meanwhile, it also reportsFigure 3.it Illustration

Figure 3.toIllustration
the previously
of Trafficconnected
of Traffic Accidents RSU through
Reduction
Accidents Reduction neighboring
Strategy
Strategy
(TARS). vehicles.
(TARS).
Neighboring vehicles only act as relay nodes and they pass on the message to the trailing vehicles
until itAlgorithm
reaches1:that particular RSU.
Traffic Congestion Avoidance Algorithm

Algorithm 1: Traffic Congestion

A flowchart Avoidanceof
of the methodology Algorithm
the proposed strategy is shown in Figure 4.
1. Input: BSM
2. RSU extracts VID;
4. if VID is registered then
5. Calculate TSi;
6. Calculate Dv;
7. if TSi expired then
8. Wait for 10 secs;
9. if BSM received then
10. Go to the line Calculate TSi;
Sensors 2018, 18, 1983 12 of 23
Sensors 2018, 18, x FOR PEER REVIEW 13 of 23

Figure 4. Workflow of TARS.

Figure 4. Workflow of TARS.
 Sensors 2018, 18, 1983 13 of 23

Scenario 1:
If a vehicle travels outside of the region in which it is registered, then the RSU of the newly
entered region contacts the TA for the verification of a vehicle. The TA checks whether the vehicle is
a registered with it or not. The TA then sends a verification message to the RSU if it is a registered
vehicle
Sensors 2018,as
18,illustrated
x FOR PEERin REVIEW
Figure 5. 14 of 23

Sensors 2018, 18, x FOR PEER REVIEW 14 of 23

Figure5.5.Scenario
Figure Scenario 1 illustration.
1 illustration.

Scenario 2:
If a non-registered vehicle enters into the range of a RSU, it then contacts the TA for vehicle
verification. The TA reports it to the Governmental Authorities (GA) for further action if that vehicle is
not registered at the TA as shown in Figure
Figure 5. 6.
Scenario 1 illustration.

Figure 6. Scenario 2 illustration.

3.4. Safety Distance Calculation Method

Let a and b represents the following and leading vehicle. The change in position of a leading
vehicle w.r.t. the following vehicle can be calculated as shown in Equation (2):
Figure
Figure6.
6. Scenario
Scenario 22 illustration.
illustration.
(𝑣 − 𝑣 ( , ))
∆( , )= (2)
3.4. Safety Distance Calculation Method 2 𝑑𝑒𝑐𝑙𝑟

Let a𝑑𝑒𝑐𝑙𝑟
where and b is the maximum
represents declaration
the following allowed
and leadinginvehicle.
the system
Theand 𝑣 ( in
change is the relative
, )position velocity
of a leading
between
vehicle vehicle
w.r.t. a and b which
the following can
vehicle bebe
can calculated
calculatedasas
shown
shownininEquation
Equation(3):
(2):
𝑣 (𝑣( , )−= 𝑣𝑣 (−, 𝑣) ) (3)
∆( , ) = (2)
The change in position of a following vehicle 2 𝑑𝑒𝑐𝑙𝑟
can be calculated as shown in Equation (4):
where 𝑑𝑒𝑐𝑙𝑟 is the maximum declaration allowed in𝑣the system and 𝑣 ( , ) is the relative velocity
∆( ) =
as2shown (4)
between vehicle a and b which can be calculated 𝑑𝑒𝑐𝑙𝑟 in Equation (3):
Sensors 2018, 18, 1983 14 of 23

3.4. Safety Distance Calculation Method

Let a and b represents the following and leading vehicle. The change in position of a leading
vehicle w.r.t. the following vehicle can be calculated as shown in Equation (2):
 2
v a − vrel (a,b)
∆(a,b) = (2)
2 declrm

where declrm is the maximum declaration allowed in the system and vrel (a,b) is the relative velocity
between vehicle a and b which can be calculated as shown in Equation (3):

vrel (a,b) = vb − v a (3)

The change in position of a following vehicle can be calculated as shown in Equation (4):

va 2
∆( a) = (4)
2 declrm

Safety distance is denoted by S.D(a,b) and it is difference of a change in position of following and
leading vehicles as illustrated in Equation (5).

S.D(a,b) = ∆(a) − ∆(a,b) (5)

By putting all the values ∆(a) − ∆(a,b) in Equation (5), it can be simplified as shown in Equations (6)–(11):
 2 
va 2
 v a − vrel (a,b)
S.D(a,b) = − (6)

2 declrm 2 declrm


" #
va 2 v a 2 + vrel (a,b) 2 − 2 v a vrel (a,b)
S.D(a,b) = − (7)
2 declrm 2 declrm

va 2 v a 2 − vrel (a,b) 2 + 2 v a vrel (a,b)

S.D(a,b) = − (8)
2 declrm 2 declrm
v a 2 − v a 2 − vrel (a,b) 2 + 2 v a vrel (a,b)
S.D(a,b) = (9)
2 declrm
−vrel (a,b) 2 + 2 v a vrel (a,b)
S.D(a,b) = (10)
2 declrm
h i
vrel (a,b) 2 v a − vrel (a,b)
S.D(a,b) = (11)
2 declrm

4. Performance Evaluation
The performance of the proposed TARS routing protocol is analyzed and evaluated by using a
real traffic simulation tool called the Simulation of Urban Mobility (SUMO) [35] integrated with the
NS-2.35 [36]. The TARS protocol simulation is implemented in a 1000 × 1000 m2 grid map with a
different number of vehicular nodes. All vehicular nodes were deployed randomly with a randomly
generated velocity range from 60 km/h to 120 km/h. Vehicles are uniformly distributed over four
lanes of a road as shown in Figures 7 and 8.
4. Performance Evaluation
The performance of the proposed TARS routing protocol is analyzed and evaluated by using a
The performance
real traffic of the
simulation tool proposed
called TARS routing
the Simulation protocol
of Urban is analyzed
Mobility (SUMO) and
[35]evaluated bywith
integrated usingthea
real traffic
NS-2.35 simulation
[36]. The TARS toolprotocol
called the Simulation
simulation of Urban Mobility
is implemented (SUMO)
in a 1000 [35]mintegrated
× 1000 2 grid map with
withthea
NS-2.35 [36].
different number The of
TARS protocol
vehicular simulation
nodes. is implemented
All vehicular nodes werein a 1000 ×randomly
deployed 1000 m2 grid
with map with a
a randomly
different
2018,number
generated
Sensors 1983 ofrange
velocity
18, vehicular
fromnodes.
60 km/hAll to
vehicular nodes
120 km/h. were deployed
Vehicles randomly
are uniformly with a randomly
distributed over15 four
of 23
generated
lanes velocity
of a road rangeinfrom
as shown 60 km/h
Figures 7 andto8. 120 km/h. Vehicles are uniformly distributed over four
lanes of a road as shown in Figures 7 and 8.

Figure 7. Highway Scenario using SUMO.

Figure 7. Highway Scenario using SUMO.
Figure 7. Highway Scenario using SUMO.

Figure 8. Real time traffic on highway using SUMO.

Figure 8. Real time traffic on highway using SUMO.
Figure 8. Real time traffic on highway using SUMO.

The TARS protocol simulation parameters are shown in Table 3.

Table 3. Properties of the TARS simulation parameters.

Network Simulator NS-2.35

Transmission Range RSU 400 m
Waiting Time 10 s
Total No. of Vehicles 40, 50, 60, 70, 80, 90, 100
Packet Size 1024 bytes
MAC Layer IEEE 802.11p
Total number of Lanes on highway 4
Velocity Threshold 40 m/h

4.1. Simulation Tool of the TARS Routing Protocol

4.1.1. Simulation of Urban Mobility (SUMO)

The SUMO tool is used to simulate the real vehicular traffic. In order to implement a realistic
highway scenario, we obtained the mobility file of the real time traffic using SUMO. A screenshot of a
real highway environment using SUMO is shown in Figure 8.
4.1. Simulation Tool of the TARS Routing Protocol
4.1. Simulation Tool of the TARS Routing Protocol
4.1.1. Simulation of Urban Mobility (SUMO)
4.1.1. Simulation of Urban Mobility (SUMO)
The SUMO tool is used to simulate the real vehicular traffic. In order to implement a realistic
highway
The
Sensors scenario,
2018,SUMO weisobtained
18, 1983tool used to the mobility
simulate thefile
realofvehicular
the real time traffic
traffic. usingto
In order SUMO. A screenshot
implement 16 of of
a realistic
23
a real highway environment using SUMO is shown in Figure 8.
highway scenario, we obtained the mobility file of the real time traffic using SUMO. A screenshot of
a real highway environment using SUMO is shown in Figure 8.
4.1.2. Network
4.1.2. Network Simulator
Simulator (NS-2.35)
(NS-2.35)
4.1.2.NS-2.35
NetworkisSimulator
another(NS-2.35)
tool used
used in
in the
the TARS
NS-2.35 is another tool TARS routing
routing scheme
scheme to to implement
implement the
the main
main
purpose/procedure
purpose/procedure
NS-2.35 is another of the TARS
of the TARS routing
tool routing protocol
used inprotocol
the TARS as shown
as shown in Figure
in Figure
routing 9.
9. to implement the main
scheme
purpose/procedure of the TARS routing protocol as shown in Figure 9.

Figure 9. Simulation of the TARS protocol.

Figure 9. Simulation of the TARS protocol.
Figure 9. Simulation of the TARS protocol.
The mobility trace file is imported in NS-2.35. The TARS protocol simulation is presented in NS-
The
2.35 as
The mobility
shown tracefile
in Figure
mobility trace fileisisimported
10. importedinin NS-2.35.
NS-2.35. TheThe TARS
TARS protocol
protocol simulation
simulation is presented
is presented in
in NS-
NS-2.35
2.35 as shown
as shown in Figure
in Figure 10. 10.

Figure 10. TARS simulation by using NS-2.35.

Figure 10. TARS simulation by using NS-2.35.
Figure 10. TARS simulation by using NS-2.35.

We focused on the highway scenario for the TARS simulation. The TARS scenario was
implemented in a VANETS environment by using NS-2 and SUMO. We ran the simulation twenty-five
times for each vehicle density parameter, and average results are shown in graphs.

4.2. Performance Evaluation Metrics

The TARS, DBSR [37], and POVRP [38] routing schemes were evaluated using a highway scenario
in VANETs’ environment on the basis of the following performance parameters: Message Delivery
Ratio (MDR), Message Loss Ratio (MLR), Average Delay, and Basic Safety Message (BSM). All of the
above parameters are discussed in detail as follows.
Sensors 2018, 18, 1983 17 of 23

4.2.1. Message Delivery Ratio (MDR)

The MDR shows the percentage of messages that are successfully delivered to the vehicles. The
basic objective of each routing protocol is to gain a high delivery ratio. The following mathematical
formula is used to calculate the MDR.

MDR = (Total No. of Received Messages/Total No. of Sent Messages) × 100

4.2.2. Message Loss Ratio (MLR)

The MLR shows the percentage of lost packets that are dropped during the transmission. The
following formula is used to calculate the MLR.

MLR = 100 − (Total No. of Received Messages/Total No. of Sent Messages) × 100

4.2.3. Average Delay

The Average Delay is the time taken by the RSU to send the safety warning message to the vehicle
when two vehicles cannot maintain the predefined threshold distance between each other.

4.2.4. Basic Safety Message (BSM)

The Basic Safety Message (BSM) shows the number of messages transmits from RSU to the vehicle
or vehicle to vehicle.

5. Results and Discussion

Now, we discuss the experimental results of our proposed protocol, TARS, and compare it with
the results obtained for DBSR and POVRP protocols using the highway scenario on the basis of the
performance parameters which are discussed in the previous section.

5.1. Basic Safety Message (BSM)

The BSM plays a very important role in VANETs’ environment to transfer safety messages. Safety
messages need to be transmitted all the time to all neighbor vehicles. In order to provide reliability in
the traffic network, safety messages must reach their destinations without causing channel congestion.
Furthermore, sending safety messages without using a congestion control or avoidance mechanism
causes the broadcasting storm problem. In TARS protocol, when the number of vehicles increases,
then the number of basic safety messages also increases. Moreover, RSU makes intelligent decisions,
sending safety messages to the other vehicles while considering the dynamic nature of topology and
the broadcasting storm problem in order to adjust, and considers the current road traffic information.
The RSU monitors the distance between two vehicles, if the distance is less than the defined threshold,
then the RSU sends warning messages in order to make the vehicle aware of the status of the traffic
network (whether it is congested or not) and distance. In order to provide the stability of the channel
network, the RSU and vehicle only send warning and safety messages when it is really needed. In
TARS, the number of basic safety messages are fewer compared to DBSR and POVRP, meaning TARS
decreases the chances of the occurrence of channel congestion in order to provide the stability of the
network protocol as well as improving the network performance depending on the capacity of the
channel. The RSU sends warning messages after verifying the possibility of congestion occurrence in
terms of high traffic load in the network, and informs the upcoming vehicle to take alternate routes to
avoid congestion as shown in Figure 11.
 and POVRP,
and POVRP, meaning
meaning TARSTARS decreases
decreases the
the chances
chances of
of the
the occurrence
occurrence of of channel
channel congestion
congestion inin order
order
to provide
to provide the the stability
stability of
of the
the network
network protocol
protocol as
as well
well asas improving
improving thethe network
network performance
performance
depending on the capacity of the channel. The RSU sends warning messages
depending on the capacity of the channel. The RSU sends warning messages after verifying the after verifying the
possibility of
possibility of congestion
congestion occurrence
occurrence inin terms
terms of
of high
high traffic
traffic load
load in
in the
the network,
network, and
and informs
informs the
the
Sensors 2018, 18, 1983 18 of 23
upcoming vehicle to take alternate routes to avoid congestion as shown
upcoming vehicle to take alternate routes to avoid congestion as shown in Figure 11. in Figure 11.

Figure 11.
Figure 11. Basic
Basic Safety
Safety Message
Message in
in the
the highway
highway scenario.
scenario.
Figure 11. Basic Safety Message in the highway scenario.
5.2. Average
5.2. Average Delay
Delay
5.2. Average Delay
The TARS
The TARS protocol
protocol shows
shows lessless average
average delaydelay asas compared
compared to to the
the DBSR
DBSR and and POVRP.
POVRP. TARS
TARS usesuses
a The TARS
traffic protocol
congestion shows less
avoidance average delay
algorithm which as reduces
comparedthe to the DBSR andmessages
unnecessary POVRP. TARS usesby
caused a
a traffic congestion avoidance algorithm which reduces the unnecessary messages caused by
traffic congestion
vehicular congested avoidance
congested traffic. algorithm
traffic. The
The RSURSU only which reduces
only transmits
transmits warningthe unnecessary
warning messages
messages andmessages
and vehiclescaused by
vehicles exchange vehicular
exchange safety
safety
vehicular
congested
messages traffic. ThetoRSU
according the only
presenttransmits
traffic warning
condition messages
(i.e., and vehicles
vehicle velocity exchange
and the safety messages
distance between
messages according to the present traffic condition (i.e., vehicle velocity and the distance between
according
each vehicle to the
vehicle present
to the
the other
other traffic condition
vehicle). TARS (i.e., vehicle velocity
overcomes and the in
the condition
condition distance
whichbetween
vehicleseachmove vehicle to
at high
high
each to vehicle). TARS overcomes the in which vehicles move at
the other
speed and vehicle).
and broadcast TARS
broadcast lots overcomes
lots of
of messages, the
messages, where condition
where there in which
there are vehicles
are aa large
large number move
number of at high
of vehicles speed
vehicles causing and broadcast
causing congestion
congestion
speed
lots
due of
to messages,
message where there are
dissemination. a large number
Moreover, after of vehicles
detecting causing the
congestion, congestion
RSU due to message
transmits warning
due to message dissemination. Moreover, after detecting congestion, the RSU transmits warning
dissemination.
messages at at the Moreover,
the exact
exact time after
time for detecting
for the
the upcoming congestion,
upcoming vehicle the
vehicle to RSU
to selecttransmits
select an warning
an alternate
alternate route messages
route in in order at
order to the exact
to prevent
prevent
messages
time for the
congestion and upcoming
and avoid vehicle
avoid accidents, to
accidents, the select an
the upcoming alternate
upcoming vehicle route
vehicle makes in order
makes more to prevent
more intelligent congestion
intelligent decisions
decisions and and avoid
and selects
selects aa
congestion
accidents,
less congestedthe upcoming
route while vehicle
applying makesa more
basic intelligent
safety message decisions
strategy and selectsshare
(vehicles a lesstheir
congested
velocity route
and
less congested route while applying a basic safety message strategy (vehicles share their velocity and
while applying
vehicle-ID with a basic
with other safety
other vehicles) message
vehicles) to to follow strategy
follow the (vehicles
the specified share
specified path.
path. In their
In this velocity
this way,
way, the and
the TARS vehicle-ID
TARS protocol with other
protocol provides
provides
vehicle-ID
vehicles)
less delay toinfollow
order the
to specifiedwarning
transmit path. Inmessages,
this way, the as a TARS
result,protocol
driver provides
safety is less delayasinshown
increased, order toin
less delay in order to transmit warning messages, as a result, driver safety is increased, as shown in
transmit
Figure warning messages, as a result, driver safety is increased, as shown in Figure 12.
12.
Figure 12.

Figure 12. Average Delay in the highway scenario.

Sensors 2018, 18, 1983 19 of 23
Sensors 2018, 18, x FOR PEER REVIEW 19 of 23

Figure 12. Average Delay in the highway scenario.

5.3. Message Delivery Ratio (MDR)
5.3.
The Message
TARSDelivery
protocol Ratio
shows(MDR) a high MDR, as compared to the DBSR and POVRP with different
numbersThe of vehicles.
TARS protocol shows a high of
When the number vehicles
MDR, increases,tothen
as compared the rate
the DBSR andofPOVRP
messages also
with increases.
different
If the RSU and
numbers vehiclesWhen
of vehicles. both transmit
the number theofmessages, then heavy
vehicles increases, thencongestion in the channel
the rate of messages will result,
also increases.
the If
frequency of transmitting messages gets reduced, and the messages cannot
the RSU and vehicles both transmit the messages, then heavy congestion in the channel will result, be delivered to the
neighboring vehicles. Moreover, due to the unsuccessful reception of messages,
the frequency of transmitting messages gets reduced, and the messages cannot be delivered to the there are a lot of
chances of the data
neighboring packet
vehicles. loss and due
Moreover, due toto the
thisunsuccessful
packet loss some safety
reception of messages
messages, also thereexperience
are a lot ofloss
andchances
cannot of the data
reach packet lossdestination.
the specified and due to this In packet
order toloss some safety
overcome thismessages
challenge, alsothe
experience loss
TARS protocol
playsand cannot
a very reach the role
important specified destination.
and shows In order
high PDR, to overcome
meaning this challenge,
the protocol builds the TARS protocol
a reliable connection
plays athe
between very
twoimportant
vehiclesrole and shows
as well as thehigh PDR, meaning
connection between the the
protocol
RSU builds a reliable
and vehicle, connection
with minimum
between the two vehicles as well as the connection between the
packet and message loss [39–43]. The RSU transmits the warning message up to one hop RSU and vehicle, with minimum
neighbor
packet and message loss [39–43]. The RSU transmits the warning message
(which lies in its transmission range) and the direction based forwarding mechanism minimizes the up to one hop neighbor
(which lies in its transmission range) and the direction based forwarding mechanism minimizes the
packet loss chances, which also helps to improve the packet delivery ratio [44]. The protocol shows
packet loss chances, which also helps to improve the packet delivery ratio [44]. The protocol shows
the successful packet transmission of warnings as well as safety messages with maximum reach, and
the successful packet transmission of warnings as well as safety messages with maximum reach, and
the ability to improve driving safety through more reliable delivery of BSMs in VANETs, as shown in
the ability to improve driving safety through more reliable delivery of BSMs in VANETs, as shown
Figure 13. Furthermore,
in Figure the MDR
13. Furthermore, achieves
the MDR achievessuccessful transmission
successful transmission to to
the
thespecified
specifieddestination;
destination;forfor the
MDR required, 7, 8, and 9 out of 10 messages per second must be calculated
the MDR required, 7, 8, and 9 out of 10 messages per second must be calculated and successfully and successfully received
from the other
received fromvehicles or vehicles
the other RSU. or RSU.

Figure 13. Message Delivery Ratio w.r.t. the Varied Vehicle Density in the highway scenario.
Figure 13. Message Delivery Ratio w.r.t. the Varied Vehicle Density in the highway scenario.

5.4. Confidence Interval in Terms of Message Delivery Ratio

5.4. Confidence Interval in Terms of Message Delivery Ratio
In order to calculate the confidence interval, the highway scenario is evaluated and analyzed.
In order
The to calculate
geographical the confidence
(highway) interval,
trace file was the highway
retrieved from SUMO scenario is evaluated
in order to achieveandtheanalyzed.
real worldThe
geographical (highway) was
scenario. Simulation traceconducted
file was retrieved
in NS-2, from SUMO in
neighboring order to
vehicles achieve
with 40–200the real world
vehicles scenario.
travelling
Simulation was conducted
random routes with the help in NS-2, neighboring
of a random vehicles
mobility model.with 40–200
For each vehicles travelling
configuration, 50 trials random
were
executed
routes for help
with the 500 s of
and the simulation
a random mobilityhasmodel.
been conducted
For each with different random
configuration, seeds;
50 trials wereasexecuted
a result, for
500 MDR
s andresults were achieved
the simulation in the conducted
has been transmissionwith
range of 30 m–2100
different randomm that are assumed
seeds; to beMDR
as a result, normally
results
distributed.
were achieved in the transmission range of 30 m–2100 m that are assumed to be normally distributed.
TheThe
MDR MDR showed
showed thetheaverage
averageprobability
probability of
of receiving
receiving aasingle
singlemessage
message with
withthethe
confidence
confidence
interval. Figure 14 shows the 95% confidence interval in the highway scenario.
interval. Figure 14 shows the 95% confidence interval in the highway scenario. The MDR shows The MDR shows the the
coverage of message dissemination, when the source vehicle transmits the message and the
coverage of message dissemination, when the source vehicle transmits the message and the simulation
simulation computes the increase of message coverage with time. If the message coverage greatly
computes the increase of message coverage with time. If the message coverage greatly increases in a
increases in a short interval, then it means the proposed protocol achieves high message
short interval, then it means the proposed protocol achieves high message dissemination performance.
TARS shows a 95% confidence interval as shown in Figure 14. In the highway scenario, the distance
between vehicles is larger as compared to the dense areas. Therefore, by using a predefined threshold
 Sensors 2018, 18, x FOR PEER REVIEW 20 of 23
Sensors 2018, 18, 1983 20 of 23

dissemination performance. TARS shows a 95% confidence interval as shown in Figure 14. In the
highway
value, the MDR
Sensors scenario,
2018, x FOR the
18,requires distance
PEERthat between
6, 8,
REVIEW and 9 outvehicles is larger as
10 messages compared
per second are to the dense areas.
received Therefore,
and are calculated
20 of 23
to have an approximately 75%, 88%, and 95% confidence interval, respectively, in a case second
by using a predefined threshold value, the MDR requires that 6, 8, and 9 out 10 messages per when the
aredissemination
received andperformance.
are calculated to have
TARS showsan aapproximately
95% confidence75%, 88%,asand
interval 95% in
shown confidence
Figure 14.interval,
In the
distances between the vehicles are smaller [45]. However, when the number of vehicles is increased
respectively, in a case
highway scenario, thewhen the between
distance distancesvehicles
between is the vehicles
larger are smaller
as compared to the[45]. However,
dense when the
areas. Therefore,
then the number of messages also increases as well as the communication becoming difficult in cases
number
by using ofa predefined
vehicles is threshold
increasedvalue,
thenthe theMDR
number
requiresof that
messages
6, 8, andalso increases
9 out as well
10 messages as the
per second
whenarethereceived
distanceand between the vehicles
are calculated to in
is greater.
have an when
So, there are
approximately
a lot
75%,between
of chances
88%, andthe 95%
that the single
confidence
message
interval,
communication becoming difficult cases the distance vehicles is greater. So,
doesthere
not reach
respectively,
are theof
a lot inspecified
a case when
chances destination
that the single
the and
distances the rate
between
message doesof message
thenot reach loss
vehicles becomes
are specified
the smaller greater
[45]. when
However,
destination the
thenumber
when
and the
rate
of vehicles
ofnumber
messageis increasing
of
lossvehicles
becomes [45].
is greater
increased
when then
thethe number
number of messages
of vehicles also increases
is increasing [45]. as well as the
communication becoming difficult in cases when the distance between the vehicles is greater. So,
there are a lot of chances that the single message does not reach the specified destination and the rate
of message loss becomes greater when the number of vehicles is increasing [45].

Figure 14. Confidence Interval w.r.t Message Delivery Ratio in the highway scenario.
Figure 14. Confidence Interval w.r.t Message Delivery Ratio in the highway scenario.
5.5. Message Loss Ratio (MLR)
5.5. Message LossFigure
Ratio14. Confidence Interval w.r.t Message Delivery Ratio in the highway scenario.
(MLR)
We evaluated and analyzed the loss warning messages or safety messages due to an unstable
5.5.connection.
We
link Message Loss
evaluated Ratio
and
The (MLR)
analyzed
TARS the loss
protocol warning
transmits messages
messages or safety messages
successfully due todestination.
at the specified an unstable Inlink
connection.
this way The
WeTARS TARS
shows
evaluated protocol
andless transmits
loss ratio
analyzed theasloss messages
compared successfully
warningtomessages
other similar at the
studies.
or safety specified destination.
The warning
messages due to an and beaconthis
unstableIn
waysafety
TARS message
shows less
link connection. must
The be
loss delivered
ratioprotocol
TARS to each
as compared neighboring
to other
transmits messages vehicle
similar without
studies.
successfully delay.
atThe A single
the warning
specified anddelay or loss
beacon
destination. safety
In
ofthis
message anymust
packet
way becould
TARS result
delivered
shows in
lesstoloss
eachof life.asHowever,
neighboring
ratio compared iftotheother
vehicle number
without of vehicles
similar studies.Aincreases
delay. single
The thenand
delay
warning the chances
or beacon
loss of any
ofsafety
packet lost packets
couldmessagealso
result inincrease,
must be of
loss as result
delivered
life. tocausing
However, ifinstability
each neighboring
the number of theofnetwork.
vehicle without Therefore,
vehicles delay. the
A single
increases thenTARS
theprotocol
delay or loss of
chances
shows
of anyless loss
packet packets
could as
result compared
in loss of to
life. DBSR
However, and POVRP.
if the It
number means
of the
vehicles
lost packets also increase, as result causing instability of the network. Therefore, the TARS protocol proposed
increases protocol
then the provides
chances
theofless
shows stability of also
lost packets the increase,
loss packets network
as comparedasand
resultmore packets
tocausing
DBSR or messages
instability
and POVRP. of the are delivered
Itnetwork.
means Therefore,
the up
proposedthetoprotocol
the protocol
TARS specified
provides
destination.
shows lessSo,lossTARS provides
packets a better
as compared toand
DBSRreliable routingItscheme
and POVRP. means astheper shown protocol
proposed in Figureprovides
15.
the stability of the network and more packets or messages are delivered up to the specified destination.
the stability of the network and more packets or messages are delivered up to the specified
So, TARS provides a better and reliable routing scheme as per shown in Figure 15.
destination. So, TARS provides a better and reliable routing scheme as per shown in Figure 15.

Figure 15. Message Loss Ratio w.r.t. the Varied Vehicle Density in the highway scenario.
Sensors 2018, 18, 1983 21 of 23

5.6. Discussion
We have highlighted the issue of road safety in Section 1. In Section 2, we summarized and
discussed the literature review. In Section 3, we proposed a strategy named as the Traffic Accidents
Reduction Strategy (TARS) that aimed to avoid vehicle collision and re-route road traffic to another
route so that the traffic keeps flowing. In Section 4, we analyzed and evaluated the performance
of our proposed protocol scheme. TARS showed a better performance in Message Delivery Ratio
(MDR), Message Loss Ratio (MLR), Average Delay, and Basic Safety Message Rate. In the end, we
concluded that all of the step-by-step work in scientific steps, theories, mathematical calculation and
performance results evaluation and discussion were verified by supporting explanations. In Section 5,
we discussed the experimental results and compared the performance of TARS protocol with the other
similar studies of routing protocols like DBSR and POVRP [37,38].

6. Conclusions
Our research paper has highlighted the issue of road safety. Road hazards have drastically
increased with the increase in automobiles over the last ten years. Many efforts have been made to
avoid such severe conditions. However, these efforts are restricted in providing a holistic solution. The
proposed protocol, TARS, forecasts the probability of the occurrence of an accident in advance before
it occurs. It also re-routes vehicle traffic to prevent traffic jams on the road that may cause accidents.
TARS is aimed at maintaining the traffic flow efficiently. It also assists drivers in reaching a destination
on time. Moreover, TARS reduces the safety messages broadcasting in order to avoid a broadcasting
storm and network congestion and minimizes the delay in re-routing of traffic to other available paths.
Our results demonstrated that TARS had a good performance in Message Delivery Ratio, Message
Loss Ratio, Average Delay, and the Basic Safety Message Rate. In future, our goal is to propose a
mathematical model and perform further simulations on performance metrics that are associated with
TARS in an urban scenario. Our results and discussion showed a successful demonstration of the
TARS proposed scheme as compared to other similar studies.

Author Contributions: A.A. and N.A. worked on and finalized the Literature Review section and finalized two
protocols which are compared with our proposed protocol known as TARS from similar studies. They also built
tables related to the literature review. H.Y. and H.M. have finalized the initial proposed scheme and worked on
the proposed methodology section. Both also suggested further improvements in the Results and Discussion
section. They also worked on all phases including the proposed methodology section, drew and finalized the
proposed architecture diagram of the TARS protocol, and developed all equations included in the threshold for
safe distance calculation method and setup phase section. M.S. and A.M. worked on and finalized the Results
and Discussion section. Both also suggested further improvements for the proposed methodology section. They
also built tables and made all diagrams included in the results and discussion section. H.S. has worked on and
finalized both the Introduction and Conclusion section. He also proof-read the whole paper, finalized the quality
of all diagrams, and made further suggestions to the whole team regarding improvement of the journal paper.
Funding: This research was funded by Deanship of Scientific Research, King Saud University through the Vice
Deanship of Scientific Research.
Acknowledgments: The authors are grateful to the Deanship of Scientific Research, King Saud University, KSA
for funding through the Vice Deanship of Scientific Research Chairs.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. World Health Organization. World Health Statistics 2010 Indicator Compendium Interim Version; World Health
Organization: Geneva, Switzerland, 2010.
2. Ganeshkumar, P.; Gokulakrishnan, P. Emergency Situation Prediction Mechanism: A Novel Approach for
Intelligent Transportation System Using Vehicular Ad Hoc Networks. Sci. World J. 2015, 2015, 218379.
[CrossRef] [PubMed]
3. World Health Organization. Global Status Report on Road Safety 2013; World Health Organization: Geneva,
Switzerland, 2013.
Sensors 2018, 18, 1983 22 of 23

4. Peden, M.; McGee, K.; Sharma, G. The Injury Chart Book: A Graphical Overview of the Global Burden of Injuries;
World Health Organization: Geneva, Switzerland, 2002.
5. Peden, M.; Scurfield, R.; Sleet, D.; Mohan, D.; Hyder, A.A.; Jarawan, E.; Mathers, C. (Eds.) World Report on
Road Traffic Injury Prevention; World Health Organization: Geneva, Switzerland, 2004.
6. Mansuri, F.A.; Al-Zalabani, A.H.; Zalat, M.M.; Qabshawi, R.I. Road safety and road traffic accidents in Saudi
Arabia: A systematic review of existing evidence. Saudi Med. J. 2015, 36, 418–424. [CrossRef] [PubMed]
7. Mohamed, H.A. Estimation of Socio-Economic Cost of Road Accidents in Saudi Arabia: Willingness-To-Pay
Approach (WTP). Adv. Manag. Appl. Econ. 2015, 5, 43.
8. Al Turki, Y.A. How can Saudi Arabia use the Decade of Action for Road Safety to catalyse road traffic injury
prevention policy and interventions? Int. J. Inj. Control Saf. Promot. 2014, 21, 397–402. [CrossRef] [PubMed]
9. DeNicola, E.; Aburizaize, O.S.; Siddique, A.; Khwaja, H.; Carpenter, D.O. Road Traffic Injury as a Major
Public Health Issue in the Kingdom of Saudi Arabia: A Review. Front. Public Health 2016, 4, 215. [CrossRef]
[PubMed]
10. Wenger, J. Automotive Radar—Status and Perspectives. In Proceedings of the Compound Semiconductor
Integrated Circuit Symposium, Palm Springs, CA, USA, 30 October–2 November 2005.
11. Spacey, J. Active Safety vs. Passive Safety. Simplicable, 14 August 2016.
12. Abbasi, H.I.; Voicu, R.C.; Copeland, J.A.; Chang, Y. Cooperative BSM architecture to improve transportation
safety in VANETs. In Proceedings of the 13th International Wireless Communications and Mobile Computing
Conference, Valencia, Spain, 26–30 June 2017; pp. 1016–1022.
13. Sommer, C.; Dressler, F. Vehicular Networking; Cambridge University Press: Cambridge, UK, 2015.
14. Jiang, D.; Delgrossi, L. IEEE 802.11p: Towards an international standard for wireless access in vehicular
environments. In Proceedings of the 2008 Vehicular Technology Conference, Singapore, 11–14 May 2008;
pp. 2036–2040.
15. De Souza, A.M.; Brennand, C.A.; Yokoyama, R.S.; Donato, E.A.; Madeira, E.R.; Villas, L.A. Traffic
management systems: A classification, review, challenges, and future perspectives. Int. J. Distrib. Sens. Netw.
2017, 13. [CrossRef]
16. Garcia-Lozano, E.; Barba, C.T.; Igartua, M.A.; Campo, C. A distributed, bandwidth-efficient accident
prevention system for interurban VANETs. In Proceedings of the 2013 International Conference on Smart
Communications in Network Technologies (SaCoNeT), Paris, France, 17–19 June 2013.
17. Gokulakrishnan, P.; Ganeshkumar, P. Road Accident Prevention with Instant Emergency Warning Message
Dissemination in Vehicular Ad-Hoc Network. PLoS ONE 2015, 10, e0143383.
18. Dawood, H.S.; Wang, Y. An Efficient Emergency Message Broadcasting Scheme in Vehicular Ad Hoc
Networks. Int. J. Distrib. Sens. Netw. 2013, 11, 232916. [CrossRef]
19. Benslimane, A. Optimized Dissemination of Alarm Messages in Vehicular Ad-Hoc Networks (VANET).
In Proceedings of the 7th IEEE International Conference on High Speed Networks and Multimedia
Communications, Toulouse, France, 30 June–2 July 2004; pp. 655–666.
20. Roy, A.; Chakraborty, J. Communication based accident avoidance and congestion control mechanism in
VANETs. In Proceedings of the 2015 International Symposium on Advanced Computing and Communication
(ISACC), Silchar, India, 14–15 September 2015; pp. 320–327.
21. Kumar, R.R.; Begum, S.W.; Manikandan, M.; Student, P.G. Distance Based Accident Prevention in Intersection
Using Vanet. Int. J. Innov. Res. Comput. Commun. Eng. 2007, 2, 3624–3629.
22. Kshirsagar, N.S.; Sutar, U.S.; Student, P.G. Review on Intelligent Traffic Management System Based on
VANET. Int. J. Innov. Res. Comput. Commun. Eng. 2015, 3, 2001–2004.
23. Kshirsagar, N.; Sutar, U.S. An Intelligent Traffic Management and Accident Prevention System based on
VANET. Int. J. Sci. Res. 2013, 14, 2319–7064.
24. Nzouonta, J.; Borcea, C. STEID: A Protocol for Emergency Information Dissemination in Vehicular Networks *;
Report; Department of Computer Science, New Jersey Institute of Technology: Newark, NJ, USA, 2006.
25. Ma, X.; Chen, X. Delay and Broadcast Reception Rates of Highway Safety Applications in Vehicular Ad Hoc
Networks. In Proceedings of the 2007 Mobile Networking for Vehicular Environments, Anchorage, AK,
USA, 11 May 2007.
Sensors 2018, 18, 1983 23 of 23

26. Kuribayashi, S.; Sakumoto, Y.; Ohsaki, H.; Hasegawa, S.; Imase, M. Performance Evaluation of Epidemic
Broadcast with Directional Antennas in Vehicular Ad-Hoc Networks. In Proceedings of the 2011 IEEE/IPSJ
11th International Symposium on Applications and the Internet (SAINT), Munich, Germany, 18–21 July 2011;
pp. 260–265.
27. Devdhara, G.; Gohil, D.; Akhade, P.; Vala, M. Inter-Vehicular Collision Detection and Avoidance using
Ad-hoc Network. Int. J. Res. Emerg. Sci. Technol. 2015, 2, 1–7.
28. Raut, S.B.; Bajaj, P.R.; Malik, L.G. Prediction of Vehicle Collision Probablity at Intersection using V2V
Communication. Int. J. Sci. Eng. Res. 2015, 6, 295–300.
29. Bhumkar, S.P.; Deotare, V.V.; Babar, R.V. Accident Avoidance and Detection on Highways. Int. J. Eng. Trends
Technol. 2012, 3, 247–252.
30. Nzouonta, J.; Rajgure, N.; Wang, G.; Borcea, C. VANET Routing on City Roads Using Real-Time Vehicular
Traffic Information. IEEE Trans. Veh. Technol. 2009, 58, 3609–3626. [CrossRef]
31. Manoj, D.; Narendra, B. Traffic Congestion Detection by Using VANET to Improve Intelligent Transportation
System (ITS). Int. J. Netw. Commun. 2015, 5, 74–82.
32. Khatri, D.; Rathod, P.D.; Gajjar, P.A.; Khanna, D.; Professor, A. Traffic Congestion Detection and Avoidance
in VANET. Int. J. Sci. Res. Dev. 2014, 2, 2321–2613.
33. Wei, Y.C.Y. SafeAnon: A safe location privacy scheme for vehicular networks. Telecommun. Syst. 2012, 50,
339–354.
34. Boualouache, A.; Moussaoui, S. TAPCS: Traffic-aware pseudonym changing strategy for VANETs. Peer-to-Peer
Netw. Appl. 2017, 10, 1008–1020. [CrossRef]
35. Krajzewicz, D.; Hertkorn, G.; Rössel, C.; Wagner, P. Sumo (simulation of urban mobility)-an open-source
traffic simulation. In Proceedings of the 4th Middle East Symposium on Simulation and Modelling
(MESM20002), Sharjah, UAE, 2002; pp. 183–187.
36. Issariyakul, T.; Hossain, E. Introduction to Network Simulator NS2; Springer Science & Business Media: Berlin,
Germany, 2011.
37. Chang, S.W.; Lee, S.S. Distance-Based Stable Routing Decision Scheme in Urban Vehicular Ad Hoc Networks.
Int. J. Distrib. Sens. Netw. 2015, 11, 245439. [CrossRef]
38. Jung, S.-D. A Position-Based on-Demand Routing Protocol in VANET. Ph.D. Thesis, Hanyang University,
Seoul, Korea, 2011.
39. Mughal, B.M.; Wagan, A.A.; Hasbullah, H. Impact of Safety Beacons on the Performance of Vehicular Ad Hoc
Networks; Springer: Berlin/Heidelberg, Germany, 2011; pp. 368–383.
40. Anyameluhor, N.; Peytchev, E.; Akhlaghinia, J.; Lane, C.; Kingdom, U. Improving Message Delivery In
Vehicular Ad-Hoc Networks. In Proceedings of the 31st European Conference on Modelling and Simulation
ECMS 2017, Budapest, Hungary, 23–26 May 2017; Volume 6.
41. Kalpande, P.B. Reliable broadcast of safety messages in VANET. Int. J. Res. Advent Technol. 2015, 3, 73–78.
42. Garg, D.; Singh, R.; Kaur, A. Performance Evaluation of Data Delivery Mechanism for CognitiveRadio
Vehicular and Vehicular Ad-Hoc Networks. Procedia Comput. Sci. 2015, 57, 596–605. [CrossRef]
43. Wahid, A.; Shah, M.A.; Qureshi, F.F.; Maryam, H.; Iqbal, R.; Chang, V. Big data analytics for mitigating
broadcast storm in Vehicular Content Centric networks. Future Gener. Comput. Syst. 2017. [CrossRef]
44. Abbasi, I.A. Dynamic Multiple Junction Selection Based Routing Protocol for VANETs in City Environment.
Appl. Sci. 2018, 8, 687. [CrossRef]
45. Paul, M.; Sanyal, G. Traffic Analysis of Vehicular Ad-Hoc Networks of V2I Communication. Procedia Comput.
Sci. 2015, 54, 215–223. [CrossRef]

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).