SDN based Network Traffic Routing in Vehicular Networks: A Scheme
and Simulation Analysis
Jitendra Bhatia
1
, Mohammad S. Obaidat
2,3,4,∗
, Tirath Savasaiya
1
, Hardik Trivedi
5
, Sudeep Tanwar
5,†
and Kuei-Fang Hsiao
6,7
1
Vishwakarma Government Engineering College, Gujarat Technological University, Ahmedabad, Gujarat, India
2
College of Computing and Informatics, University of Sharjah, U.A.E.
3
KAIST, University of Jordan, Jordan
4
University of Science and Technology Beijing, China
5
Department of CSE, Institute of Technology, Nirma University, Ahmedabad, Gujarat, India
6
Department of Information Systems, University of Sharjah, U.A.E.
7
Department of Information Management, Ming Chuan University, Taiwan
Keywords:
VANET, Software Defined Networks, Routing, Traffic Optimization.
Abstract:
This paper focuses on exploiting the service architecture that exercises the Software-Defined Network (SDN)
concept in heterogeneous vehicular networks for handling data traffic using optimal path selection and routing
in the situation like congestion in Vehicle Ad Hoc Networks (VANETs). In particular, we consider the scenario
where a set of information is requested by a vehicular node from other vehicular nodes. Software-Defined
Networking (SDN) have caught much attention in vehicular networks where decoupling of data and control
plane enables the centralized control of the network; providing flexibility to a great extent. In this paper,
we design an efficient network traffic routing algorithm assisted by SDN. Finally, we built a realistic traffic
based simulation model. Simulation results show that the proposed protocol leveraged by the SDN framework
outperforms the conventional network for data traffic management in terms of Round Trip Time (RTT) and
Packet Delivery Ratio (PDR).
1 INTRODUCTION
The design and development of the routing algorithms
that enhance the efficiency of the vehicular networks
have been motivated by recent advancements in ve-
hicular communications. VANET consists of vehi-
cles where they behave as mobile nodes and make
requests for some data as per the application require-
ment. An example of traditional VANET is shown in
Fig.1where all the vehicles use GPS and OBU (On-
Board Unit) equipped with sensors and communica-
tion hardware. Traditional IP networks are not suf-
ficient to meet all the needs of the VANET require-
ments. A traditional network is unable to provide
flexibility, scalability and is not capable of manag-
ing huge number of mobile nodes in VANET(Chahal
et al., 2017). As traditional network has high latency;
∗
Fellow of IEEE and Fellow of SCS
†
Member, IEEE
some of the VANET applications can not work prop-
erly. All VANET applications have a different kind
of requirements like high packet delivery ratio, man-
Figure 1: Traditional VANET(Ku et al., 2014a).
Bhatia, J., Obaidat, M., Savasaiya, T., Trivedi, H., Tanwar, S. and Hsiao, K.
SDN based Network Traffic Routing in Vehicular Networks: A Scheme and Simulation Analysis.
DOI: 10.5220/0009911400130020
In Proceedings of the 10th International Conference on Simulation and Modeling Methodologies, Technologies and Applications (SIMULTECH 2020), pages 13-20
ISBN: 978-989-758-444-2
Copyright
c
 2020 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved
13

aging a heterogeneous environment, better QoS and
these requirements can’t be served by the traditional
network due to its limitation (Vora et al., 2018).
SDN is such a developing technology in which
management of the network abstraction through the
lower level of functionality is handled by the network
administrator. SDN framework mainly comprises of
two sections, First is control plane and second one
is information plane(Liu et al., 2017) (Bhatia et al.,
2018). The control plane is generally implemented
in the server as a software component. The data for-
warding mechanism is executed by the data plane
available on the network device like switch and router.
An Open Flow is a standard used for communication
between infrastructure and control plane. Northbound
and Southbound APIs can be used for communica-
tion and those APIs can also be changed according to
requirements. The Road Side Unit (RSU) fully exer-
cises the logically centralized controller (Bhatia et al.,
2020). SDN concept was not fully fledge explored
in VANET so far. The closest studies of SDN are a
couple of implementations in VANET (Tanwar et al.,
2018c).
Some changes in the traditional VANET architec-
ture will help to integrate VANET with SDN. SDN-
based VANET example is shown in Fig.2. In place
of a base station, SDN supported open-flow switches
that can be used for communicating with the SDN
controller directly or through the Internet. Based on
the requirements, RSU may communicate with a base
station and a base station in further, may communi-
cate with Controller or RSU can directly communi-
cate with SDN controller. It is also possible to deploy
the SDN controller on the cloud or on fog if the area,
which is covered by the SDN controller is beyond the
coverage (Tanwar et al., 2018a).
Figure 2: SDN-based VANET.
This work is dedicated to the efficient routing of
vehicular application data traffic under the I2V com-
munication environment. Communication efficiency
between vehicles and RSU is the fundamental func-
tion of such services.. The principle ideas of the
whole paper are as follows:
• Vehicles are likely to publish or subscribe to var-
ious infotainment and emergy messages. Moti-
vated by this observation, the routing of those
messages on RSU in such an I2V communication
environment has been investigated. This is the
study based on SDN for efficient routing of net-
work traffic in VANETs with consideration of ap-
plication requirement under communication con-
straints.
• Based on a global view of traffic information in
the network, a simulation model has been devel-
oped. Under different traffic loads, performance
evaluation of the proposed algorithm has been
done. The efficiency of simulation is discussed
in the result section.
An overview of the paper is as follows. First, the main
notion on which our approach is based on is intro-
duced. Then various network routing strategies and
related works in VANETs have been reviewed by au-
thors (Tanwar et al., 2019) (Tanwar et al., 2014). Fur-
thermore, the author proposed the SDN enabled ar-
chitecture for the network traffic routing for better re-
sponse time and higher throughput. After that, based
on the precisely described algorithm, problem anal-
ysis has been conducted. At the end, the simulation
model and the experimental results are discussed by
the authors. Finally, it is concluded with open issues
that can be carried as future work.
2 RELATED WORKS
Past studies on data dissemination in VANETs em-
phasizes on improving communication reliability,
Quality of Service (QoS) and the issues commonly
dwell in PHY and MAC layers (Chaturvedi and Sri-
vastava, 2017)(Bhatia et al., 2019). One of the prime
issues is to manage the network traffic efficiently un-
der communication constraints. To answer this ques-
tion, lots of mechanisms came into existence (Wang
et al., 2018) (K
¨
uhlmorgen et al., 2019) (Tanwar et al.,
2018b) (Wu et al., 2019). Using SDN instead of the
traditional network provides many benefits like better
efficiency, congestion control, better PDR. The main
idea after introducing the SDN in the field of VANET
is that the traditional network is not able to provide
the required QoS to VANET (Li et al., 2018) (Dave
and Bhatia, 2013) (Wu et al., 2018).
Ku et al.(Ku et al., 2014b), proposed the idea of
SIMULTECH 2020 - 10th International Conference on Simulation and Modeling Methodologies, Technologies and Applications
14

using SDN with VANET to make it available with
flexibility and programmability. In the proposed idea,
the OpenFlow controller communicates with Open-
Flow switches. Two different communication chan-
nels are also used, Wi-Fi for the data plane and LTE
for the Control Plane. The architecture was classi-
fied into three operation modes, one of them is cen-
tralized control mode, second is hierarchical control
mode and the last one is distributed control mode (He
et al., 2016b) (Bhatia et al., ).
Duan et al.(Duan et al., 2014) (Jindal et al., 2018)
proposed a special architecture, Vehicular Cyber-
Physical Systems (VCPS). The proposed architecture
reduces packet delay time by 20%. The key compo-
nents of the system were vehicles, RSU, OpenFlow
switches, and global controller. Proposed approach
works on the location-based routing protocol. Liu et
al.(Liu et al., 2015), discussed the idea of GeoBroad-
cast in VANET. It supports periodic broadcast mes-
sage. Controller overhead is reduced by the proposed
approach. It uses less amount of bandwidth and la-
tency is also reduced. In proposed architecture flood-
light was used as a controller.
He et al.(He et al., 2016d), presented the idea of
using multicast in VANET based on SDN on the basis
of trajectory prediction, which reduces the burden of
SDN control and data plane and also provides better
multicast scheduling decisions. Dong et al. (Dong
et al., 2016), presented SDN-based on-demand rout-
ing protocol SVAO. In Map based protocol, traffic is
forwarded according to road topology. Road informa-
tion is taken into the account in Map-Based protocols.
Ji et al.(Ji et al., 2016), proposed the approach
of using geographic routing in SDN based VANET.
Some, typical geographic routing protocols only use
local information to make decision, which may lead
to local maxima. Kumar et al.(Sahoo and Yunhas-
nawa, 2016)(Bhatia, 2015), proposed the idea of us-
ing cloud service in VANET based on SDN for better
Packet Delivery Ratio (PDR) and less RTT. In SDN
based VANET, many vehicle requests for data, cloud
service can be used to meet these requirements. Luo
et al. (Luo et al., 2016), introduced hybrid architec-
ture in VANET for managing physical resources in it,
to provide pre-warning collision and topology change
in VANET. The proposed sdnMAC protocol is TDMA
based. In proposed algorithm time is divided into
equal frame sizes. Each RSU should at least acquire
a one-time slot in one frame.
Liu et al.(Liu et al., 2016), idea of using Coop-
erative Data Scheduling (CDS) in hybrid SDN based
VANET is proposed by authors. The ultimate goal is
to serve the maximum number of request. In the pro-
posed approach two service channels are used, one
is for I2V communication and one is for V2V com-
munication. At a time vehicles can only send or re-
ceive data. Each vehicle has its own record of re-
quested items. At a time vehicle can transmit or re-
ceive only one data item in one scheduling period.
He et al.(He et al., 2016c) proposed the idea of us-
ing SDN in VANET to schedule resources and re-
duce cost. Heterogeneity of the current VANET in-
troduces two problems: one is interoperability and the
other one is resource allocation. To solve these issues
researchers design resource scheduling solutions for
VANET.
Wang et al.(Wang et al., 2017) claims that using
the traditional network in VANET introduces com-
plexity. To overcome this issue they proposed SDN
based IoV (SDIV). Using IoV (Internet of Vehicles)
is also a great challenge because it has limited size
flow table and it will be difficult to configure hetero-
geneous switches. He et al.(He et al., 2016a) pro-
posed the idea of using fog computing in SDN based
VANET to provide less latency and support mobility
and location awareness. Modified Constrained Op-
timization Particle Swarm Optimization (MPSO-CO)
algorithm is proposed by authors. To reduce the bur-
den of cloud computing, idea of using cloud comput-
ing with fog computing and SDN is introduced (Bha-
tia et al., 2018)(Yao et al., 2018)(Shah et al., 2019).
In contrast to prior work, we have designed an
SDN based network traffic rerouting under the over-
load condition. SDN leverages the global view of net-
work traffic load on RSU, which helps to minimize
the response time to request when network traffic load
tends to higher.
3 PROPOSED APPROACH
In this section, the proposed algorithm is briefly ex-
plained. The basic idea is to exploit the SDN capa-
bilities to analyze the network and reroute the traf-
fic based on the network overhead. SDN controller
has a global view of the network, which comprises
of the parameters like a number of vehicular nodes in
the network, bandwidth and total requirement of the
nodes. The load factor is calculated by dividing band-
width by requirement of a single application running
on an application layer. Thus, if there are N nodes, it
can be directly calculated as nodes requirement. Now
the value of available nodes and load for an intermedi-
ate node is compared where the load is the maximum
number of nodes which can be served by an interme-
diate device at a time. If the value of the load factor is
less than number of available nodes, it indicates that
the number of available nodes is more and the sys-
SDN based Network Traffic Routing in Vehicular Networks: A Scheme and Simulation Analysis
15

tem is being overloaded. For SDN based vehicular
network, the SDN controller has all the information
about the network. So, we can get the nearest under-
loaded node called as a proxy node for the interme-
diate device which is overloaded. After getting proxy
node, start rerouting traffic from that node. Now, the
load factor is calculated for a proxy node and com-
pared with load factor. If load factor is less then avail-
able nodes for proxy node then repeat above process
for proxy node, else continue routing from that node.
After fixing time of interval, the value of load
factor is recalculated because many applications run
on the application layer. On the basis of the newly
calculated load value, again conditions are checked
whether current traffic can be served by an interme-
diate node or not. If traffic is more, then find proxy
node and reroute traffic from that node. After check-
ing conditions for the load factor of an intermedi-
ate node and for the proxy node, if both conditions
are not matched, it signifies that traffic is decreased
and the proxy node can be relieved, but if all the
nodes are heavily overloaded, the performance of the
system may be decreased and a warning message is
displayed. In the result section, it is clear that our
Algorithm 1: SDN based Traffic Rerouting protocol.
Step 1: Get bandwidth, Nodes, Requirement . Parameters
required to calculate load factor
Step 2: Calculate load f actor for intermediate node through
which traffic is being passed using the following function:
load f actor = (bandwidth)/Requirement
. Load factor is maximum number of nodes that can be served by
an intermediate node
Step 3: Compare Load Factor with number of available nodes to
intermediate device.
if (Load<= AvailNodes) then
Go to Step 4.
else
System is not overloaded, so keep observing.
end if
Step 4: Get nearest node of intermediate device which is under-
loaded and called as proxy node.
step 5: Find load factor for proxy node.
Step 6: Compare values of load factor and available nodes for the
proxy node.
if (Load <= AvailNodes) then
Repeat step 4, step 5 and step 6 for proxy node.
else
System is safe and Continue routing from the proxy node.
end if
Step 7: Repeat Step 3
. At regular time interval value of load factor is recalculated.
Step 8:If all the nodes are heavily loaded then give warning that
performance of the system may decrease.
Step 9:Exit
proposed algorithm outperforms traditional compet-
ing approach.
Computational Complexity. However, small over-
head is incurred to find out the total number of nodes
and requirement of the system at the fixed interval
of time. At regular interval of time, the load value
is calculated and if traffic is more, proxy node is
started. After some time if load decreases, it relieves
the proxy node and resets the counter value. If the
counter value is increased more then load, it means
now proxy node is also overloaded and performance
of system decreases. This overhead is minor and can
be ignored, but it still affects the performance. Proto-
col overhead is lower than before and comparatively
achieved efficiency is higher which can be seen in the
result section.
4 EXPERIMENTAL SETUP
4.1 Simulation Scenario
Two networks were taken to perform the simulation
viz., simulation area of 1 KM and 2 KM is shown in
Fig.3. In both networks, block size is of 200m*200m
and four different scenarios were taken into consider-
ation where number of nodes varies as 50,100,150 and
200. The details of simulation parameter are shown in
the Table 1.
Table 1: Simulation Parameters.
Parameter Value
Tools Used NS-3,SUMO(Song
et al., 2014)
Simulation Area 1KM, 2KM
Simulation Time 250 Second
Chanel bandwidth 300kbps
Proxy Node capacity 300kbps
Number of Vehicles 50-200
Vehicle Max Speed 14m/s
Open flow module OFSwitch 1.3
For both networks, simulation was performed for
different number of vehicles varying from 50 to 200.
All vehicles moves randomly in network with random
speed and maximum vehicle speed is set to 14m/s. All
the nodes pings to a server from one route and at one
stage traditional switch is unable to serve all nodes.
However, in case of SDN, due to the global view of
the network re-routing the traffic when network traffic
increases is needed.
SIMULTECH 2020 - 10th International Conference on Simulation and Modeling Methodologies, Technologies and Applications
16

Figure 3: Grid Road Network 1 and 2 KM.
Figure 4: Packet Delivery Ratio for 1KM.
4.2 Evaluation Metrics
Fig.4 shows the Packet Delivery Ration (PDR)for
1KM area and for 50,110,150 and 200 nodes. As
the number of node increases, it can be seen that in
TP (Traditional protocol), PDR decreases and same is
applicable for 2 KM area, which can be seen in Fig.5.
But in proposed NP (New Protocol), PDR is almost
100%. Only few packets are lost. SDN conroller has
an idea about the traffic load on particular node and it
reroutes the traffic.
RTT is time taken by a packet to reach at the des-
tination and come back to the source node. We have
used ping application which uses ICMP protocol. We
can see that as number of node increases, RTT in-
creases too. Fig.6 shows the maximum RTT time
for simulation area of 1 KM and for 50,100,150 and
200 nodes. where Fig.7 shows the maximum RTT
time for simulation area 2 KM for 50,100,150 and 200
nodes. We can see that when number of node reaches
at 150 and 200 in TP(Traditional Protocol), RTT also
increases whereas in our proposed NP(New protocol)
RTT remains same because of SDN and traffic rerout-
ing paradigm.
Figure 5: Packet Delivery Ratio for 2KM.
Figure 6: Max RTT Time for 1KM.
5 DISCUSSION
Two performance parameters were taken into consid-
eration, i.e., PDR and RTT which can be seen in Fig.4
and Fig:5 respectively. In Fig.4 PDR for 1 KM area
is shown. It is visible that for 50 and 100 nodes, Tra-
ditional Protocol serves well, but as number of nodes
increases, PDR decreases whereas our proposed pro-
tocol yields almost the same results. Fig.5 shows PDR
for 2 KM area in which, if we compare with 1 KM
area it can be seen that PDR in 2 KM is slightly less
then 1 KM because of the distance. Second parame-
ter is RTT for which we have taken 10 sample nodes
from all simulations and compared them. Fig.8 shows
SDN based Network Traffic Routing in Vehicular Networks: A Scheme and Simulation Analysis
17

Figure 7: Max RTT Time for 2KM.
Figure 8: RTT Time for 1KM.
RTT for 1 KM for 2 different scenarios where num-
ber of nodes are 50 and 100 and RTT of Tradition
protocol is compared with the newly proposed proto-
col. Fig.9 shows RTT for 1 Km area for 150 and 200
nodes. Fig.10 shows RTT for 2 KM area for number
of nodes of 50, and 100 and Fig.11 shows RTT for 2
km area for 150 and 200 nodes.
It can be seen that both in 1 KM area and 2 KM
area for 50 and 100 nodes almost both protocols gives
the same result, but when traffic increases in case of
150 and 200 nodes, PDR decreases in traditional net-
work and because of heavy traffic RTT time increases
in traditional network. RTT is almost 3.8 seconds for
2 KM area and 200 nodes. Definitely all networks
have some limitations like number of nodes which
can be served or limited bandwidth, but When we use
SDN openflow approach, controller comes to know
Figure 9: RTT Time for 1KM.
Figure 10: RTT Time for 2KM.
when node is being heavily congested and it reroutes
the traffic accordingly. SDN reroutes the traffic and
another node behaves as proxy and much more bet-
ter result can be gained by this paradigm. It can be
seen in Fig:11 that traditional network’s RTT is much
more, but the maximum RTT time of the new pro-
posed protocol is approximately 800 ms, which is
much lower then second one. The results also de-
pends on the capacity of proxy node. Actually, we
can say that it depends on the node which is behaving
as proxy for all the traffic.
In our simulation, we inferred that at rate of 300
Kbps bandwidth, traditional protocol is capable of
serving around 100 to 105 nodes and after that, there
is drastic performance drop. When traffic increases,
SDN performs rerouting and uses another path. But if
node count increases to a certain limit N or if traffic
increases to certain limit T, it is possible that proposed
new protocol is unable to serve all and performance is
dropped.
SIMULTECH 2020 - 10th International Conference on Simulation and Modeling Methodologies, Technologies and Applications
18

Figure 11: RTT Time for 2KM.
6 CONCLUSION AND FUTURE
DIRECTIONS
SDN plays a very important role in VANET for net-
work traffic routing. The issues of the traditional net-
works have been solved by SDN. Due to its flexibil-
ity and programmability new scope of technology has
been risen. The proposed strategy yields low latency
and high throughput. The impact of the different
number of vehicular nodes has been analyzed in I2V
networks by considering a grid network. The authors
give a substantial analysis of both the requirement and
constraint on network traffic routing. On this basis,
an algorithm has been described. A dynamic algo-
rithm for the same is intended to enhance throughtput
and reducing response time. SDN enabled VANET
architecture leverages the benefits of a global view
of traffic information which assists our approach for
handling the network traffic efficiently. Simulation
results under different traffic loads demonstrate the
superiority of the proposed approach. Designing a
backup mechanism for the SDN controller in case of
failure can be a motivation for future direction.
REFERENCES
Bhatia, J., Dave, R., Bhayani, H., Tanwar, S., and Nayyar,
A. (2020). Sdn-based real-time urban traffic analy-
sis in vanet environment. Computer Communications,
149:162–175.
Bhatia, J., Govani, R., and Bhavsar, M. (2018). Soft-
ware defined networking: From theory to practice.
In 2018 Fifth International Conference on Parallel,
Distributed and Grid Computing (PDGC), pages 789–
794.
Bhatia, J., Kakadia, P., Bhavsar, M., and Tanwar, S. (2019).
Sdn-enabled network coding based secure data dis-
semination in vanet environment. IEEE Internet of
Things Journal, pages 1–1.
Bhatia, J., Mehta, R., and Bhavsar, M. (2018). Variants of
software defined network (sdn) based load balancing
in cloud computing: A quick review. In Future Inter-
net Technologies and Trends, pages 164–173, Cham.
Springer International Publishing.
Bhatia, J., Modi, Y., Tanwar, S., and Bhavsar, M. Soft-
ware defined vehicular networks: A comprehensive
review. International Journal of Communication Sys-
tems, page e4005.
Bhatia, J. B. (2015). A dynamic model for load balancing
in cloud infrastructure. Nirma University Journal of
Engineering and Technology (NUJET), 4(1):15.
Chahal, M., Harit, S., Mishra, K. K., Sangaiah, A. K., and
Zheng, Z. (2017). A survey on software-defined net-
working in vehicular ad hoc networks: Challenges,
applications and use cases. Sustainable Cities and So-
ciety.
Chaturvedi, M. and Srivastava, S. (2017). Multi-modal de-
sign of an intelligent transportation system. IEEE
Transactions on Intelligent Transportation Systems,
18(8).
Dave, J. R. and Bhatia, J. (2013). Issues in static periodic
broadcast in vanet. International Journal of Advances
in Engineering & Technology, 6(4):1712.
Dong, B., Wu, W., Yang, Z., and Li, J. (2016). Software
defined networking based on-demand routing proto-
col in vehicle ad hoc networks. In Mobile Ad-Hoc
and Sensor Networks (MSN), 2016 12th International
Conference on, pages 207–213. IEEE.
Duan, P., Peng, C., Zhu, Q., Shi, J., and Cai, H. (2014). De-
sign and analysis of software defined vehicular cyber
physical systems. In Parallel and Distributed Systems
(ICPADS), 2014 20th IEEE International Conference
on, pages 412–417. IEEE.
He, X., Ren, Z., Shi, C., and Fang, J. (2016a). A novel
load balancing strategy of software-defined cloud/fog
networking in the internet of vehicles. China Commu-
nications, 13(2):140–149.
He, Z., Cao, J., and Liu, X. (2016b). Sdvn: enabling rapid
network innovation for heterogeneous vehicular com-
munication. IEEE network, 30(4):10–15.
He, Z., Zhang, D., and Liang, J. (2016c). Cost-efficient
sensory data transmission in heterogeneous software-
defined vehicular networks. IEEE Sensors Journal,
16(20):7342–7354.
He, Z., Zhang, D., Zhu, S., Cao, J., and Liu, X. (2016d).
Sdn enabled high performance multicast in vehicular
networks. In Vehicular Technology Conference (VTC-
Fall), 2016 IEEE 84th, pages 1–5. IEEE.
Ji, X., Yu, H., Fan, G., and Fu, W. (2016). Sdgr: An sdn-
based geographic routing protocol for vanet. In In-
ternet of Things (iThings) and IEEE Green Comput-
ing and Communications (GreenCom) and IEEE Cy-
ber, Physical and Social Computing (CPSCom) and
SDN based Network Traffic Routing in Vehicular Networks: A Scheme and Simulation Analysis
19

IEEE Smart Data (SmartData), 2016 IEEE Interna-
tional Conference on, pages 276–281. IEEE.
Jindal, A., Aujla, G. S., Kumar, N., Chaudhary, R., Obai-
dat, M. S., and You, I. (2018). Sedative: Sdn-enabled
deep learning architecture for network traffic control
in vehicular cyber-physical systems. IEEE Network,
32(6):66–73.
Ku, I., Lu, Y., Gerla, M., Gomes, R. L., Ongaro, F.,
and Cerqueira, E. (2014a). Towards software-defined
vanet: Architecture and services. 2014 13th Annual
Mediterranean Ad Hoc Networking Workshop (MED-
HOC-NET), pages 103–110.
Ku, I., Lu, Y., Gerla, M., Ongaro, F., Gomes, R. L.,
and Cerqueira, E. (2014b). Towards software-defined
vanet: Architecture and services. In Ad Hoc Network-
ing Workshop (MED-HOC-NET), 2014 13th Annual
Mediterranean, pages 103–110. IEEE.
K
¨
uhlmorgen, S., Lu, H., Festag, A., Kenney, J., Gemsheim,
S., and Fettweis, G. (2019). Evaluation of congestion-
enabled forwarding with mixed data traffic in vehicu-
lar communications. IEEE Transactions on Intelligent
Transportation Systems.
Li, F., Song, X., Chen, H., Li, X., and Wang, Y. (2018). Hi-
erarchical routing for vehicular ad hoc networks via
reinforcement learning. IEEE Transactions on Vehic-
ular Technology, 68(2):1852–1865.
Liu, K., Feng, L., Dai, P., Lee, V. C., Son, S. H., and Cao,
J. (2017). Coding-assisted broadcast scheduling via
memetic computing in sdn-based vehicular networks.
IEEE Transactions on Intelligent Transportation Sys-
tems, 19(8):2420–2431.
Liu, K., Ng, J. K., Lee, V. C., Son, S. H., and Stojmen-
ovic, I. (2016). Cooperative data scheduling in hybrid
vehicular ad hoc networks: Vanet as a software de-
fined network. IEEE/ACM transactions on network-
ing, 24(17):1759–1773.
Liu, Y.-C., Chen, C., and Chakraborty, S. (2015). A soft-
ware defined network architecture for geobroadcast in
vanets. In Communications (ICC), 2015 IEEE Inter-
national Conference on, pages 6559–6564. IEEE.
Luo, G., Jia, S., Liu, Z., Zhu, K., and Zhang, L. (2016). sd-
nmac: a software defined networking based mac pro-
tocol in vanets. In Quality of Service (IWQoS), 2016
IEEE/ACM 24th International Symposium on, pages
1–2. IEEE.
Sahoo, P. K. and Yunhasnawa, Y. (2016). Ferrying vehicu-
lar data in cloud through software defined networking.
In Wireless and Mobile Computing, Networking and
Communications (WiMob), 2016 IEEE 12th Interna-
tional Conference on, pages 1–8. IEEE.
Shah, N. B., Shah, N. D., Bhatia, J., and Trivedi, H.
(2019). Profiling-based effective resource utiliza-
tion in cloud environment using divide and conquer
method. In Fong, S., Akashe, S., and Mahalle, P. N.,
editors, Information and Communication Technology
for Competitive Strategies, pages 495–508, Singa-
pore. Springer Singapore.
Song, J., Wu, Y., Xu, Z., and Lin, X. (2014). Research
on car-following model based on sumo. In Advanced
Infocomm Technology (ICAIT), 2014 IEEE 7th Inter-
national Conference on, pages 47–55. IEEE.
Tanwar, S., Kumar, N., and Niu, J.-W. (2014). Eemhr:
Energy-efficient multilevel heterogeneous routing
protocol for wireless sensor networks. International
Journal of Communication Systems, 27(9):1289–
1318.
Tanwar, S., Trivedi, H., and Priyank, T. (2018a). Soft-
ware defined network-based vehicular adhoc net-
works for intelligent transportation system: Recent
advances and future challenges. Futuristic Trends in
Network and Communication Technologies (FTNCT),
858:325–337.
Tanwar, S., Tyagi, S., Kumar, N., and Obaidat, M. S.
(2019). La-mhr: Learning automata based multilevel
heterogeneous routing for opportunistic shared spec-
trum access to enhance lifetime of wsn. IEEE Systems
Journal, 13(1):313–323.
Tanwar, S., Tyagi, S., and Kumar, S. (2018b). The role of in-
ternet of things and smart grid for the development of
a smart city. In Intelligent Communication and Com-
putational Technologies, pages 23–33. Springer.
Tanwar, S., Vora, J., Tyagi, S., Kumar, N., and Obaidat,
M. S. (2018c). A systematic review on security issues
in vehicular ad hoc network. Security and Privacy,
1(5):1–23.
Vora, J., Kaneriya, S., Tanwar, S., and Tyagi, S. (2018).
Performance evaluation of sdn based virtualization for
data center networks. In 2018 3rd International Con-
ference On Internet of Things: Smart Innovation and
Usages (IoT-SIU), pages 1–5.
Wang, X., Ning, Z., Hu, X., Wang, L., Hu, B., Cheng, J.,
and Leung, V. C. (2018). Optimizing content dissem-
ination for real-time traffic management in large-scale
internet of vehicle systems. IEEE Transactions on Ve-
hicular Technology, 68(2):1093–1105.
Wang, X., Wang, C., Zhang, J., Zhou, M., and Jiang, C.
(2017). Improved rule installation for real-time query
service in software-defined internet of vehicles. IEEE
Transactions on Intelligent Transportation Systems,
18(2):225–235.
Wu, G., Wang, J., Obaidat, M. S., Yao, L., and Hsiao,
K.-F. (2019). Dynamic switch migration with non-
cooperative game towards control plane scalability in
sdn. International Journal of Communication Sys-
tems, 32(7):e3927.
Wu, G., Wang, L., Xu, Z., Yao, L., and Obaidat, M. S.
(2018). A qos and cost aware fault tolerant scheme
insult-controller sdns. In 2018 IEEE Global Com-
munications Conference (GLOBECOM), pages 1–6.
IEEE.
Yao, L., Zhao, X., Wu, G., and Obaidat, M. S. (2018).
The community characteristic based controller de-
ployment strategy for sdns. In 2018 IEEE Global
Communications Conference (GLOBECOM), pages
1–6. IEEE.
SIMULTECH 2020 - 10th International Conference on Simulation and Modeling Methodologies, Technologies and Applications
20