Content uploaded by Kashif Naseer Qureshi
Author content
All content in this area was uploaded by Kashif Naseer Qureshi on Jul 21, 2015
Content may be subject to copyright.
SSNM-Smart Sensor Network Model for Vehicular
Ad hoc Networks
Kashif Naseer Qureshi, Abdul Hanan Abdullah, Majid Bukhari and Raja Waseem Anwar
Department of Communication, Faculty of Computing,
Universiti Teknologi Malaysia, Skudai, Malaysia
kashifnq@gmail.com
Abstract— Vehicular ad hoc network is an emerging
tremendous technology that facilitates ubiquitous wireless
communication between vehicles and possibly with road side
infrastructure to providing safety, driver assistance and comfort.
Real-time vehicular information gathering is a significant and
crucial part in vehicular communication about vehicle position
and velocity, etc. For broadcasting and gathering real time
vehicles information, usually base stations are used for
communication with vehicles and backbone infrastructure. This
vehicular- to-infrastructure architecture installed with different
types of power cables and other heterogeneous devices, this type
of solution elevate the installation and maintenance cost. Wireless
sensor network is considering low cost technology to bridge the
gap between physical and digital world. These wireless sensors
are deployed without cables restriction in vehicular network to
overcome the traditional system issues. In this paper, we propose
a sensor based model for reliable and efficient communication in
vehicular ad hoc networks. The complete process of data
transmission and aggregation are designed and prototyped.
Keywords— Transmission, Aggregation, Wireless Sensor,
Vehicular network
I. INTRODUCTION
Vehicular ad hoc networks (VANETs) are self-organizing
distributed communication networks for providing road safety
and comfort by frequent exchange the data between vehicles
and infrastructure. At the present time increasing number of
fatalities due to different road dangers and accidents have been
recognized as a serious issue. According to WHO (World
Health Organization) the total number of road accident deaths
are 1.24 million per year [1]. Recently, VANETs have emerged
to turn the attention of researchers in the field of mobile and
wireless communication to deal with road dangers. The
accident detection and safety messages are most significant
services to enhance driving safety and comfort.
Communication between vehicles or with infrastructure is
possible through short and long range communication
standards such as wireless access for vehicular environment
(WAVE) and IEEE 802.11p, etc. [2, 3]. Moving cars are
communicating approximately 100 to 300 m with each other
and with infrastructure and create a wide range network.
VANET is a sub class of traditional mobile ad hoc network
(MANET) but with some unique features differentiate it with
MANET such as high mobility, frequently changing network
topology, etc. The communication performance is degrade and
suffered from dynamic change topology, interference because
of obstacles, lack of infrastructure, channel fading, etc. [4, 5].
The convergence of wireless sensor network with vehicular
communication facilitates highways and roads with an efficient
communication platform. The recent advances in embedded
technologies the wireless sensors networks have been used to
perform different monitoring tasks. Wireless sensor networks
breaks traditional end-to-end communications and introduced a
new type of distributed information exchange. The tiny sensor
nodes are deployed on roads with sensing capabilities for
information exchange.
In this paper a SSNM-Smart Sensor Network Model
propose for better and reliable communication in VANETs.
Through this model reduces the network delays, load and
removes redundant messages.
The paper is organized as follows: in section two, we
discuss related work and existing network issues in VANETs.
In section three, we discuss the propose model in detail. The
last section presents the simulation results and compared
propose model with traditional model in terms of packet loss
and end-to-end delay.
II. RELATED WORK
Various different types of solutions have been proposed to
improve vehicular network applications performance. The
VANETs applications performed different functions such as
provide traffic safety and monitoring, control, traffic law
enforcement and infotainment, etc. The role of wireless sensors
in these applications are limited such as for information
retrieval, etc. The sensors devices work proactively to warn
drivers about prior dangerous situation such as sense road
conditions, presence of obstacles, animal detection, overtaking
assistance, etc. [6, 7].
Fanyu et al. [8] proposed a system framework based on
static road side sensors for information retrieval. Mobile sensor
nodes in vehicles monitor and collected road condition and
disseminate to road side sensors. This hybrid method addressed
the traffic congestion control issue by collecting data through
vehicles and road side sensors. This approach is limited for
2015 International Conference on Smart Sensors and Application (ICSSA)
978-1-4799-7364-4/15/$31.00 ©2015 IEEE 82
events by emplacing static sensor nodes without base station.
Hua et al. [7] proposed a sensor based scheme to address the
inherent limitations of traditional VANET based systems with
two communication interfaces: WiFi (IEEE 802.11) and
ZigBee (IEEE 802.15.4). The sensor node based on ZigBee
standard for communication with other sensors. The prototype
of the system tested in field and results showed that through
suitable parameters the energy and safety achieved
simultaneously. This system also limited with sensor
performance and its energy consumption issues. Wolfgang et
al. [6] proposed cooperative road infrastructure system with
overtaking assistance and based on wireless sensors and
actuators for measuring and estimating road properties.
Carolina et al. [9] proposed HSVN (Hybrid sensor
vehicular network) platform for disseminating road information
between vehicles. Author tested different routing protocols to
evaluate network performance in the presence of wireless
sensor nodes. According to this model the road divided into
segments, but the VANETs topology is dynamically change so
the proposed approach is suitable for pre plan highways and
not effect for urban and un-planned rural areas.
For our propose model we use two standard routing
protocols DSR (Dynamic Source Routing) [10] and AODV
(Ad Hoc on Demand Distance Routing) [11] protocols, which
are effective in the terms of throughput, end-to-end delay.
AODV is multihop topology based routing protocol, it's pure
on demand; need base protocol and maintained all the listed
routes in the table. In AODV, the source node initiates by using
Hello beacons for discover and detect its neighbor, then the
source transmit a route request packet (RREQ), and then its
neighbor repeat same procedure with their neighbors. The
RREQ packets do not know about route to the destination
target before sending the packet to their neighbors. The RREQ
message has IP address of source and destination nodes and
contains current and last known sequence numbers. When
RREQ packet are reach to destination node, it add the entry of
address and send this information to their routing table, it is
called backward learning. Finally, a reply packet (RREP)
transmitted to the source node.
The DSR protocol was proposed in 1996, and it is
straightforward and competent routing protocol similar with
AODV, It forms the route on demand and depend on source
routing as a substitute of table. The meaning of source routing
is the header of every packet bear sequence number list of
nodes for transmission to the destination. The approach of DSR
is depend on two processes: route discovery and maintenance;
in discovery process source node first check the node cache for
existence of a route. If the route entry does not exist then it
starts the discovery phase. During the packet transfer between
source and destination, in maintenance process if source node
noticed any broken route then use any other alternative route to
the destination.
III. SYSTEM OVERVIEW
The proposed SSNM (Smart sensor network model) based
on four subsystems: (1) Sensing subsystem (acquisition of
information) (2) Distribution subsystem (information
distribution) (3) Decision subsystem (information processing)
(4) Execution subsystem (execution of information). These
subsystems are multi directional subsystems to interact with
each other and equipped with different heterogonous devices.
The Fig 1, shows the complete smart sensor network model
with four sub system.
The sensing subsystem is based on prevailing wireless
sensor network technology and deployment of tiny size sensors
throughout the observation area. These sensors detects vehicles
and communicate wirelessly with each other. If the area of
sensing is larger to communicate with other sensors in network,
then various different types of routing protocols are available to
create self-forming and self-healing capabilities. On the other
hand vehicles nodes are also used for large areas to relay the
information with the help of on board sensors installed in
vehicles. The distribution subsystem is responsible to exchange
the information between sensing and decision subsystems.
Because it is placed in center and serving as transmission in-
charge of sensed data. Another responsibility of this subsystem
is forwarding every event received from sensing subsystem and
need significant amount of power. To cover up this place
Fig 1: Smart sensor network model
2015 International Conference on Smart Sensors and Application (ICSSA)
978-1-4799-7364-4/15/$31.00 ©2015 IEEE 83
researchers proposed different solutions such as on board
devices installed in vehicles, base stations, smart phones and
cellular networks [7, 12].
Devices in vehicles with onboard facilities are used to
broadcast the information without any energy constraints, but
the problem in this approach is one hop communication, which
enables data sharing from source to every vehicle. The second
multi-hop method is base station (BS) faster and scalable
because it is equipped with various technologies, but it requires
more system resources. Another issue is negative impact of low
technology penetration rates on system performance. Smart
phones also used for distribution subsystems but these devices
need user interaction and used 3G (third generation) cellular
networks. The cellular system is an attractive choice for
distributed subsystem, some gateway require to be placed close
to cellular base station and some sensors nodes connect with it
with different radio technologies such as 3G (Third
generation), WLAN (Wide Local Area Network), etc. The third
subsystem is decision making use for planning and important
actions in order to complete the objective of system. Basically
this subsystem responsibilities divided into three different
categories: data storage, processing and managing the network.
The fourth and last subsystem is execution for necessary
actions which foster changes in traffic flow. It is composed
with visual and acoustic drivers and stimuli.
3.1 Main Features of SSNM
The one of the main key feature of SSNM is its ability to
obtain information from road environment. Sensor nodes
acquire raw data and process them for occurrence of some
events such as vehicle velocity and length, etc. Another feature
is vehicle identification and classification and making statistics
from road by different vehicles. The process of re-identification
is enabling the vehicle tracking, travel time, origin demands
and for environment monitoring [13]. These sensor based
systems are helpful for detect the presence of different
obstacles and animals on the roads.
The vehicle detection is significant for traffic safety and for
control applications, with sensor nodes, it will be easily detect
moving vehicle and at least one measurement recorded.
Previous traditional vehicle detection systems used video
cameras instead of sensors. The collaborative sensors
deployment are useful for detect the speed of vehicle because
the single sensor node require additional length [14]. The
above Fig 1, showed the one cellular tower covering a large
road junction, the system is cost effective in terms of many
base station installation.
3.2 SSNM Hardware
For sensing subsystem we used different types of sensors
with different functions. The AMR (Anisotropic Magneto-
Resistance Sensors) are used for detecting vehicles because of
their tiny size, low power features and provide accurate
detection functionality [15, 16]. These sensors measure the
variation produced by earth magnetic field. These are placed in
encapsulation plastic box on the roadside. Acoustic sensors are
used to detect vehicles engine noise through microphone. It is
used for long range and used in multiple lanes for reliable
detection [17]. Passive Infrared sensor used for vehicle
velocities and used in multi-lanes detection. For road
monitoring we used accelerometers to detect humidity, snow,
ice, etc. The Fig 2, shows the internal and external view of
sensor network.
3.3 Data Transferring and Aggregation
The sensor subsystem nodes detect vehicle distance and
direction and send to base station because base station need
these information all the time. The base station installed at
intersection and always need the information such as how
many vehicles reach in every lane before signal ends. Base
station is responsible for collecting all the messages and send
to decision subsystem. Denote the information that vehicle
Fig 2: Internal and external structure view of sensor node
2015 International Conference on Smart Sensors and Application (ICSSA)
978-1-4799-7364-4/15/$31.00 ©2015 IEEE 84
node provided is Γ (v),
Denoted the information a base station unit b provided is β (b),
(2)
Then,
(3)
In the above formula, f
1( ) present aggregation function such
as average, count, max, min, etc. and Φ (b) shows the vehicle
set monitored through base station B and VJ shows the vehicle
unit. The brief function is depend on VANET application.
The distributed subsystem is not only gathering vehicle
information it also connect with upper level with decision
subsystem. In last the data will send to execution subsystem
and this information set denote with Si and βT (Si), then
The f2( ) is another aggregation function and depends on the
application.
A vehicle unit Vj sends vehicle information denote with г(Vj)
to the sensor nodes Si and then sensors collects the
information after several data aggregations and send to base
station, the complete procedure shows in Fig 3.
As mentioned above, the data aggregates divided into three
main steps: vehicles to sensors (V→S), relay by sensor unit
(Si→Si-1), and sensor unit to base station unit (Si→B).
• V→S: S sends broadcast message
(vehicle_info_Request), with parameter of SID. V
sends Vehicle_Info back to S with parameter г(V).
• Si → Si-1: Road information with parameter βT (Vi).
• V1→ B: the message is the same as above, with
different receiver.
The vehicles communication, sensors and base station
systems are different with each other but the basic architecture
is same with different requirements such as hardware,
operating system and protocols for communication. In system
hardware the on board units are installed in cars, and different
types of sensors deployed on cellular base station. Operating
system in sensors is WSN:FDCX08-W and also its depend on
applications. The protocol standard IEEE 802.15.4 is used to
meet the system requirement in sensor side and for vehicles
and cellular system the dedicated short range communication
(DSRC) and WAVE IEEE 1609 are implemented.
(a) Data transmission and aggregation
(b) Message procedure
Fig 3 (a and b): Data transmission and aggregation
IV. SIMULATION RESULTS
In this section, the simulation results obtained for proposed
model have been presented. To analyze the performance of
proposed SSNM model we used NS2 simulator with SUMO
mobility generator [18]. We have two scenarios to test our
proposed sensor based scheme, the first is traditional VANETs
environment and second one is with sensors subsystem. We
used two well knows routing protocols for routing called ad
hoc networks AODV (Ad hoc on demand) and DSR (Dynamic
source routing) [19]. In first scenario vehicles node exchange
the information with road side unit and RSU send and gather
data with traffic management system for further process. This
communication is multi-hop and depends on applications. In
second scenario the data will pass through sense subsystem
where various different types of nodes deployed on the road
(1)
(4)
2015 International Conference on Smart Sensors and Application (ICSSA)
978-1-4799-7364-4/15/$31.00 ©2015 IEEE 85
side and send the gather data to cellular tower for further
process. The below Table 1, shows the simulation parameters
to test these two scenarios.
TABLE I. SIMULATION PARAMETERS
S/No Description Parameters
1 Vehicle Velocity µ=30 to 70 km/h
2 Road lanes 2
3 Road length 2 KM
4 Number of vehicle 10 to 20
5 Transmission Range of
WSN and VANET nodes
150 m
6 Routing Protocols AODV, DSR
7 Packet Size 500,1000,1500 bytes
8 MAC IEEE 802.11b
We have analyzed the performance losses and compared
the performance of AODV to DSR in terms of end-to-end
delay and packet losses. The below Fig 3 (a) shows the
evolution of packet loses in traditional VANET scenario with
different packet sizes and show the 70% confidence interval for
these values, where six simulations per point have been carried
out with different vehicle speeds.
(a) Traditional VANET scenario
(b) WSN based VANET scenario
Fig 4 (a and b): Packet loses evaluation in traditional and WSN
based VANETs scenarios with AODV and DSR routing
protocols
The result depicted in Fig 4. show the comparison of
traditional and WSN based VANET with different vehicle
speeds. It can be clearly observed that packet loses of proposed
model is below, whereas with two routing protocols the packet
loss of traditional model is higher. The difference between
traditional and WSN based model can be attributed to the fact
that proposed model is based on wireless sensors.
An end-to-end delay has been given in Fig 5. The proposed
system shows the traditional system end-to-end time is higher
compared with WSN based system results. The reason behind
the better results of end-to-end delay is its sensor based
strategy of data collection and aggregations. The stable nature
of proposed system will decrease the network load and increase
throughput in vehicular ad hoc network.
(a) End-to-end delay in traditional scenario
(b) End-to-end delay in WSN based System
Fig 5 (a and b): End-to-end delay of traditional and WSN based
VANET scenarios with DSR and AODV protocols
The results clearly shows that SSNM (Smart sensor
network model) outperforms in terms of packet loss and end-
2015 International Conference on Smart Sensors and Application (ICSSA)
978-1-4799-7364-4/15/$31.00 ©2015 IEEE 86
to-end delay. The low cost and effective wireless sensor nodes
deployment with already exist cellular base station can help to
forward the data effectively in the presence of obstacles in city
environment.
V. CONCLUSION
To address the issues of information collection and
transferring for further process in vehicular ad hoc networks,
this paper propose a wireless sensor based model with its low
cost and power capabilities. In the presence of obstacles the
traditional base station deployment is not suitable especially in
city environment. These base stations are high in cost with fix
radio range. Wireless sensors deployment on road side is an
effective solution for data gathering and aggregation between
different subsystems. The VANET environment is portioned
frequently due to the absence of vehicles, these road side
sensors can help to forward the packets to cellular base station,
which is already exist in network. The results showed the better
performance of proposed system in terms of packet loss and
end-to-end delay in network. The proposed sensor based
system also saves space, cost effective, fast transmission and
able to reduce communication overhead and bandwidth.
ACKNOWLEDGMENT
This research is supported by the Ministry ofEducation
Malaysia (MOE) and in collaboration with Research
Management Centre (RMC) Universiti Teknologi Malaysia
(UTM). This paper is funded by the GUP Grant (vote
Q.J130000.2528.06H00).
REFERENCES
[1] World Health Organization. WHO global status report on road safety
2013: supporting a decade of action. World Health Organization, 2013.
[2] Standards Committee. "Wireless LAN medium access control (MAC)
and physical layer (PHY) specifications: Amendment 8: Medium Access
Control (MAC) Quality of Service enhancements." IEEE Computer
Society (2005).
[3] K. N. Qureshi and A. H. Abdullah, "A survey on intelligent
transportation systems," Middle-East Journal of Scientific Research, vol.
15, pp. 629-642, 2013.
[4] S. Mao, S. Lin, S. S. Panwar, Y. Wang, and E. Celebi, "Video transport
over ad hoc networks: multistream coding with multipath transport,"
Selected Areas in Communications, IEEE Journal on, vol. 21, pp. 1721-
1737, 2003.
[5] K. N. Qureshi and A. H. Abdullah, "Localization-Based System
Challenges in Vehicular Ad Hoc Networks: Survey," Smart Computing
Review, vol. 4, pp. 515-528, 2014.
[6] W. Birk, E. Osipov, and J. Eliasson, "iRoad—Cooperative Road
Infrastructure Systems for Driver Support," in Proceedings of the 16th
ITS World Congress, 2009.
[7] H. Qin, Z. Li, Y. Wang, X. Lu, W. Zhang, and G. Wang, "An integrated
network of roadside sensors and vehicles for driving safety: Concept,
design and experiments," in Pervasive Computing and Communications
(PerCom), 2010 IEEE International Conference on, 2010, pp. 79-87.
[8] F. Kong and J. Tan, "A collaboration-based hybrid vehicular sensor
network architecture," in Information and Automation, 2008. ICIA 2008.
International Conference on, 2008, pp. 584-589.
[9] Tripp Barba, Carolina, Karen Ornelas Aguirre, and Mónica Aguilar
Igartua. "Performance evaluation of a hybrid sensor and vehicular
network to improve road safety." Proceedings of the 7th ACM workshop
on Performance evaluation of wireless ad hoc, sensor, and ubiquitous
networks. ACM, 2010.
[10] Ad hoc On-Demand Distance Vector (AODV) Routing.
http://www.ietf.org/rfc/rfc3561.txt
[11] The Dynamic Source Routing Protocol (DSR. http://www.rfc-
editor.org/rfc/rfc4728.txt
[12] L. Li, Y.-A. Liu, and B.-H. Tang, "SNMS: an intelligent transportation
system network architecture based on WSN and P2P network," The
journal of China universities of posts and telecommunications, vol. 14,
pp. 65-70, 2007.
[13] C. Sun, S. G. Ritchie, K. Tsai, and R. Jayakrishnan, "Use of vehicle
signature analysis and lexicographic optimization for vehicle
reidentification on freeways," Transportation Research Part C: Emerging
Technologies, vol. 7, pp. 167-185, 1999.
[14] S. Y. Cheung, S. Coleri, B. Dundar, S. Ganesh, C.-W. Tan, and P.
Varaiya, "Traffic measurement and vehicle classification with single
magnetic sensor," Transportation research record: journal of the
transportation research board, vol. 1917, pp. 173-181, 2005.
[15] A. Haoui, R. Kavaler, and P. Varaiya, "Wireless magnetic sensors for
traffic surveillance," Transportation Research Part C: Emerging
Technologies, vol. 16, pp. 294-306, 2008.
[16] K. N. Qureshi and A. H. Abdullah, "Adaptation of Wireless Sensor
Network in Industries and Their Architecture, Standards and
Applications," World Applied Sciences Journal, vol. 30, pp. 1218-1223,
2014.
[17] D. H. Goldberg, A. G. Andreou, P. Julian, P. O. Pouliquen, L. Riddle,
and R. Rosasco, "A wake-up detector for an acoustic surveillance sensor
network: Algorithm and vlsi implementation," in Proceedings of the 3rd
international symposium on Information processing in sensor networks,
2004, pp. 134-141.
[18] M. Behrisch, L. Bieker, J. Erdmann, and D. Krajzewicz, "Sumo-
simulation of urban mobility-an overview," in SIMUL 2011, The Third
International Conference on Advances in System Simulation, 2011, pp.
55-60.
[19] K. N. Qureshi And Abdul. H. Abdullah, "Topology Based Routing
Protocols for Vanet and Their Comparison with Manet," Journal of
Theoretical and Applied Information Technology, vol. 58, 2013.
2015 International Conference on Smart Sensors and Application (ICSSA)
978-1-4799-7364-4/15/$31.00 ©2015 IEEE 87
