Instrumental Environment of Multi-protocol Cloud-oriented
Vehicular Mesh Network
Glazunov Vadim, Kurochkin Leonid, Kurochkin Mihail and Popov Sergey
Telematics Department, Saint-Petersburg State Polytechnic University, St. Petersburg, Russia
Keywords: Digital Communication Protocol, MESH Network, Traffic Network, Interface, Route Protocol, Cloud
Service, Interface Network Capacity, Network Traffic.
Abstract: The article deals with issues of message transfer adequacy in the cloud-oriented vehicular MESH network,
specification of message delivery time with different traffic network configuration, vehicular traffic density,
network traffic density.
1 INTRODUCTION
Nowadays a greater attention has been paid to
researches of issues of cloud-oriented vehicular
MESH network construction. Constant enhancement
of communications transmission media, network
hardware, methods of inter-network interaction and
access to cloud services allow to set new tasks of
providing information services to information
network users. Special interest is paid to issues of
quality increase of message exchange between
transport networks users in the process of movement
in unstable connection areas (Zaborovsky, 2011).
Messages exchange between a vehicle and a
cloud is done through a communication channel with
the cloud detached to the vehicle organized by
means of automotive telematics devices, or user’s
personal devices such as smart phones, laptops and
public cellular networks in communication mode.
Service quality in the cloud-oriented environment is
determined by third-party equipment reliability and
cellular networks coverage quality (Mulukha, 2011).
Organization of a mobile self-organized local
vehicular network with exit points to the cloud
environment has been considered as a perspective
direction of information network development in
unstable coverage areas. In such model message
exchange between the vehicle and the cloud can be
fulfilled through several alternative paths, and a
vehicular aggregate on the route can be regarded as a
mobile local network with altering topology and
composition with a variable number of
communication transmission points with the cloud
network. MESH network is a perspective technology
of communication transmission between vehicles
and the cloud environment. Most of the authors who
work in this direction study models of networks with
stationery signal relays.
The paper (Ghamri-Doudane, 2012) deals with
issues of the cloud-oriented MESH environment
construction with a regular structure of stationery
network points. It proves a necessity of development
of a new multi-address communication protocol in
the MESH environment based on data of vehicles
and stationary points location. Authors consider
issues of structure of the MESH-environment
information control system that is oriented at
vehicles.
The paper (Aoyama, 2011) offers a structure of
automotive information system, conception of
access sceneries to cloud environment and customer
sceneries of interaction with the cloud.
The paper (Olariu, 2012) is dedicated to the
description of technology environment of wireless
vehicular networks traffic modeling with application
of the NS-3 simulator libraries.
The paper (Rémy, 2012) offers the network with
a one-step signal relay through a vehicle equipped
with LTE transmitter.
The paper (Benslimane, 2011) describes
algorithms of forming and simulating of dynamic
VANET clusters for access to global services.
568
Vadim G., Leonid K., Mihail K. and Sergey P. (2013).
Instrumental Environment of Multi-protocol Cloud-oriented Vehicular Mesh Network.
In Proceedings of the 10th International Conference on Informatics in Control, Automation and Robotics, pages 568-574
DOI: 10.5220/0004589805680574
Copyright
c
SciTePress
Figure 1: Scheme of interaction of single-rank heterogeneous network objects with the cloud environment.
2 INTERACTION OF NETWORK
OBJECTS WITH THE CLOUD
ENVIRONMENT
Most of the researchers pay no attention to the
situation of continuous communication lack with a
stationary transmitter when immediate
communication transfer is impossible. Meantime if
each vehicle is regarded as a mobile transmitter it is
possible to provide message transfer with the cloud
through communication transfer via them. The
choice of communication transfer point to the cloud
environment appears if numerous access points to
the cloud exist.Present day classical algorithms of
MESH networks route search are used to build
access routes to the cloud environment. They offer
the search of single route to a unique pre-known
user. In the case of the cloud-oriented service it is
required to make a choice of the most perspective
point from several alternatives. It is necessary to find
an available point with the cloud access and to
evaluate perspective of communication transfer and
receipt through them. (Gramaglia, 2011)
Figure 1 presents the chart of interaction of
single-rank heterogeneous network objects with the
cloud environment.
Communication with the cloud can be
implemented in two ways: by the vehicle with
communication equipment and by the stationery
point. The vehicle situated out of communication
area can access the cloud through a vehicle
transmitter chain. The network provides a
bidirectional message transfer – from the vehicle to
the cloud and inversely.
The most important task of such network
building is the alternative choice of communication
protocol usage for increasing message transfer
InstrumentalEnvironmentofMulti-protocolCloud-orientedVehicularMeshNetwork
569
adequacy. Estimation of message time delivery to
cloud services and the MESH network users depends
on vehicle traffic intensity, communication network
download traffic, interface availability and its
composition in vehicles.
This paper studies the problem of message
transfer adequacy increase between the cloud
environment and the MESH (802.11s) network.
A set of interfaces that allows being a user of
LTE and MESH networks concurrently or MESH
only (communication between vehicles only) is
installed on vehicles (network points). The double-
interface point (LTE, 802.11s) serves as a gateway
providing communication between the MESH
network and the cloud environment through LTE.
Dependencies of message transfer adequacy,
design and actual communication speed relations,
mean lag of message delivery during interaction of
the point with the cloud environment depending on
used route protocol and intensity of communication
streams are determined in the frame of the research.
Features of the communication network in
question given in fig.1 are following:
lack of a stable communication channel between
the vehicle and the cloud;
necessity of message transfer via MESH to the
point that has an access to the cloud environment
(LTE).
3 NS-3 NETWORK CORE
MODULES AND EXTERNAL
MODELS
Basic research method is a simulation modeling of
the network function in NS-3 environment (discrete-
event simulator of telecommunication systems
(Nsnam.org, 2012)).
NS-3 simulator is a free software licensed under
the GPL license. It is targeted for research and
educational use. NS-3 source codes are opened for
research, modification and usage and available at the
project’s site. C++ and Python are used as an
embedded language for model description.
Models of wire and wireless network that allow
simulating mixed networks with various complexity
topologies are developed in the NS-3. Special
interest in this paper is given to the MESH networks
realization on the base of 802.11s stack protocol.
Intelligent Driver Model (IDM) was used to
build a vehicular network and model highway
mobility (Treiber, 2000).
IDM describes movement of the vehicle on the
highway network taking into account the vehicle
location on traffic lanes, vehicle’s overall
dimensions, distance between vehicles, vehicle’s
behavior at the lane change, lane traffic direction,
average speed and acceleration in the given lane,
traffic lights signals, traffic allocation on cross-roads
(Arbabi 2010).
Simulation was carried out in the NS-3 simulator
(v.3.16). The IDM model realization redesign
connected with adaptation of its data structures to
the latest version of the NS-3 (v.3.16) was fulfilled
to coordinate the NS-3 and IDM versions.
A model of mobile point with a set of network
interfaces was developed to perform experiments.
To synthesize a multi-protocol point a new class
was added to the IDM model. It enlarged a list of
point available interfaces with a following set:
802.11abg, 802.11s, LTE. Primarily the interface set
was limited with a 802.11s interface.
In the process of simulation the movement of the
multi-protocol mobile point is made with setting the
parameters of mobile traffic scheme. Change of
vehicle coordinates leads to change of multi-
protocol mobile point coordinates. The vehicle is
moved discretely in the interval of 0.1 sec during all
the period of modeling.
Method of variable parameters initialization is
developed to conduct the research. It includes an
“access to cloud” point feature, a route protocol, a
communication speed, number of points in the
traffic, number of points with simultaneous
transmission, size of transmitted packages, vehicular
communication protocol. Measurable parameters are
time of package sending, time of package delivery,
number of lost packages, package size, IP addresses
of sources and message recipients.
4 IMPLEMENTATION
4.1 Creation of Multi-protocol Point
The NS-3 simulator v. 3.16 does not give a ready
solution for creation of a multi-protocol point that
serves as a gateway between 802.11s and LTE
networks (fig.2). Building of simulation model of
mobile communication network required realization
of a multi-protocol point model (fig.3) functioned as
message routing between 802.11s and LTE
communication networks. Realization of a multi-
protocol point is made on the base of a “spot-to-
spot” virtual point enabled intermediate interaction
between 802.11s and LTE interfaces.
To implement a model with the multi-protocol
ICINCO2013-10thInternationalConferenceonInformaticsinControl,AutomationandRobotics
570
point that provides interaction of LTE and MESH
networks it is required to carry out following
actions:
1. to create points with a 802.11s interface
(NodeContainer class);
2. to program movement model of LTE points (it is
important to make this before installation of UE
and eNB interfaces) (LTE, 2012);
3. to install LTE base station points (eNB,
InstallEnbDevice method, LteHelper class);
4. to install LTE clients points (UE,
InstallUeDevice method, LteHelper class);
5. to create a “spot-to-spot” virtual point that serves
as an intermediate link to connect LTE and
MESH interface;
6. to add LTE interface in MESH point (Add
method NodeContainer class);
7. to create a gateway to connect LTE with extranet
(GetPgwNode method EpcHelper class);
8. to assign IP-addresses to network points
including virtual point (Assign method
Ipv4InterfaceContainer class);
9. to add a route from a chosen point to LTE
network via intermediate multi-interface point
with MESH and LTE (AddNetworkRouteTo
method Ipv4StaticRouting class);
10. to set route on default on LTE interfaces
(through a gateway in this network, a gateway
address is got by GetUeDefaultGatewayAddress
method EpcHelper class and set by
SetDefaultRoute method Ipv4StaticRouting
class).
Figure 2: Basic implementation of gateway 802.11s-LTE.
The chart given at fig.3 allows message
delivering from any MESH network point to the
cloud.
Figure 3: Implementation of the multi-protocol point.
4.2 Special Features of the Model
Program Implementation
Modules NS-3 used in the process of model
implementation:
802.11s interface model (
Andreev, 2010).
Implementation enables by default to use the
HWMP route protocol in proactive and reactive
modes, in addition to that route protocols for
wireless MESH networks will be used: OLSR,
AODV, DSDV.
implementation of route protocol models in
wireless networks HWMP, OLSR, AODV,
DSDV (Narra, 2011);
FlowMonitor is a module of network traffic
statistics collecting and processing that provides
various methods of modeling network devices
and communication channels specification
collecting;
WireShark is an analyzer of computer network
traffic that gives broad opportunities to filter and
sort traffic data of various network protocols;
PyViz is a module of model visualization that
allows to show topology of a modeling network,
data traffic, specifications of interfaces and
channels and their alteration during simulation
process as well;
NS-3-highway-mobility is a model of vehicular
traffic.
The given chart (fig.4) of modules interaction allows
combining various route protocol, network
interfaces, and models of network point mobility
within the NS-3 simulator.
Parameters of simulation are sent to
NetworkNodes class where choice and route protocol
setting is implemented, several network interfaces
are created, and speed of communication, number of
transmitting points and type of network traffic are
set up. Simulation is resulted in a set of xml-files
generated by the FlowMonitor module.
InstrumentalEnvironmentofMulti-protocolCloud-orientedVehicularMeshNetwork
571
Figure 4: Presents the environment structure of heterogeneous network model.
Table 1: Experiments initial data.
Parameter Value
Network 802.11s, LTE
Quantity of network points 4 -16
Highway fragment
200m x 200m with two-
way traffic
Communication intensity from the
point to the cloud
8-2048 Kbit/s
Maximal speed of the point 50 km/h
Message size 1024 bytes
Route protocols
HWMP, OLSR, AODV,
DSDV
Traffic type TCP, UDP
One network points is equipped with LTE
interface (it sends data to the cloud environment),
data from other network users are sent to it.
Valid values of simulation parameters prescribed
by such limits allows researching the most dynamic
periods of the MESH network existence (short time
of the network life, wide range of network traffic
intensity, high intensity of route relocation).
5 ANALYSIS OF RESULTS
5.1 Estimation of Design and Actual
Communication Speed Relations
Research of actual communication speed was carried
out for different routing protocol and short time of
network life. Routing protocols in wireless MESH
networks OLSR, DSDV, AODV and the HWMP
protocol developed specially for 802.11s were
studied for comparison. The UPD traffic sent with
speeds of 8, 32, 64, 128, 512, 1024, 2048 Kbit/s was
used in the experiment. Relation of actual data
transfer rate to network communication speed is
ICINCO2013-10thInternationalConferenceonInformaticsinControl,AutomationandRobotics
572
given in fig.5 and fig.6 shows relation of design data
transfer rate to actual one.
Figure 5: Actual data transfer rate for different routing
protocols: for 8-64 (a), 128-2048 (b) Kbit/s transfer speed.
Figure 6: Relation of design and actual data transfer rates.
Actual data transfer rate in the mobile network is
lower than design one due to generation of service
traffic required by routing protocols to support of
actual data of network topology state. Interval of
actual speed decrease makes up from 5 to 37%.
Substantial decrease of actual communication speed
(up to 35%) is observed at speeds over 1024 Kbit/s
required for transfer media data. AODV and DSDV
protocol shows best results with high intensity
streams. HWMP protocol demonstrates the highest
speed of short messages transfer with low intensity.
5.2 Estimation of Delivered Messages
Number
Number of delivered messages was defined for the
most difficult modes of the network existence when
the time of the vehicle staying in the network is less
than 1 second. Constant work of routing protocol
and a significant percentage of lost messages are
typical for them. Following factors were taken into
account in a simulation process with fixed number
of mobile transmitters of a mobile point – 16.
Figure 7: Percentage of lost messages.
5.3 Estimation of Average Time
of Message Transfer through
the Route
Average time of message transfer from the user to
the mobile transmitter that has a communication
channel with the cloud for different number of
points (8 and 16), for the HWMP routing protocol.
Following parameter points were given during
simulation process with 512 Kbit/s average data
transfer rate and TCP traffic.
At this parameters average time of message
delivery from the user to the cloud environment
made 70,41 ms for 8 points and 4.04 ms for 16
points. Reduction of message delivery time for 16
points is connected with the increase of
communication routes number.
6 CONCLUSIONS AND FUTURE
WORK
The research has led to the following results:
statement of urgent problem of studying behavior
of mobile wireless communication network on the
base of 802.11s, LTE directed at usage of the cloud
InstrumentalEnvironmentofMulti-protocolCloud-orientedVehicularMeshNetwork
573
environment services;
development of wireless communication model
based on the IDM that differs from analogues
with multi-protocol points provided interaction
of 802.11s, LTE networks and the cloud
environment;
selection of the set of parameters and parameter
domain: routing protocol, communication traffic
intensity, number of network points, etc. that
allows studying periods of the greatest alteration
of the structure and state of mobile wireless
communication network;
estimation of security level of data transfer,
actual speed of data transfer supported by the
network and average time of message exchange
of the network point with the cloud environment.
It is planned to make a detailed research of temporal
features, to analyze possibility of building optimal
routes of message re-sending between the network
point and the cloud environment, to develop a
universal routing protocol for a multi-protocol
cloud-oriented vehicular MESH network based on
the current research.
REFERENCES
Ghamri-Doudane, 2012, Yacine Ghamri-Doudane, Piotr
Szczechowiak, Sean Murphy, Vehicular Mesh
Networks for Infotainment Content Delivery: the
Carmesh Perspective. In Global Information
Infrastructure and Networking Symposium (GIIS).
IEEE Computer Society, Choroni, Venezuela, 2012, 1-
6.
Aoyama, 2011, A. Iwai and M. Aoyama, Automotive
cloud service systems based on service-oriented
architecture and its evaluation. In Proceedings of the
2011 IEEE 4th International Conference on Cloud
Computing. Cloud Computing (CLOUD),
Washington, DC, USA. IEEE Computer Society,
2011, 638–645.
Olariu, 2012, Cristian Olariu, Adriana Hava, Piotr
Szczechowiak, Yacine Ghamri-Doudane, Prototyping
Telematic Services in a Wireless Vehicular Mesh
Network Environment. In Wireless Communications
and Networking Conference Workshops (WCNCW),
2012, Paris, France, 371-376.
Rémy, 2012, Guillaume Rémy, Sidi-Mohammed Senouci,
François Jan, Yvon Gourhant LTE4V2X: LTE for a
Centralized VANET Organization. In Global
Telecommunications Conference (GLOBECOM 2011),
IEEE Computer Society, Houston, TX, USA, Houston,
TX, USA 2011, 1-6.
Benslimane, 2011, A. Benslimane, T. Taleb, and R.
Sivaraj, Dynamic clustering-based adaptive mobile
gateway management in integrated vanet-3g
heterogeneous wireless networks. In Selected Areas in
Communications, IEEE Journal on (Volume:29 ,
Issue: 3), Mar. 2011, 559-570.
Arbabi 2010, Highway mobility and vehicular ad-hoc
networks in ns-3. In Proceedings of the Winter
Simulation Conference (WSC '10), Björn Johansson,
Sanjay Jain, and Jairo Montoya-Torres (Eds.). Winter
Simulation Conference 2010, 2991-3003.
Nsnam.org, 2012, NS-3.[online]. Available at:
<http://www.nsnam.org>[Accessed 10 March 2013].
Treiber, 2000, Martin Treiber , Ansgar Hennecke, Dirk
Helbing. Congested Traffic States in Empirical
Observations and Microscopic Simulations, In Rev. E
62, Issue 2, 1805-1824.
Andreev, 2010, Kirill Andreev, Pavel Boyko. IEEE
802.11s Mesh Networking NS-3 Model.[online].
Available at: <http://www.nsnam.org/workshops/wns3
-2010/dot11s.pdf>[Accessed 10 March 2013].
LTE, 2012, LTE Module.[online] Available at:
<http://www.nsnam.org/docs/release/3.16/models/html
/lte.html>[Accessed 10 March 2013].
Narra, 2011, Hemanth Narra, Yufei Cheng, Egemen K.
Çetinkaya, Justin P. Rohrer and James P.G.
Destination-Sequenced Distance Vector (DSDV)
Routing Protocol Implementation in ns-3. In
Proceedings of the 4th International ICST Conference
on Simulation Tools and Techniques (SIMUTools '11),
Brussels, Belgium, 2011, 439-446.
Gramaglia, 2011, Marco Gramaglia, Carlos J Bernardos
and Maria Calderon, Seamless internet 3G and
opportunistic WLAN vehicular connectivity.[online].
Available at: <http://link.springer.com/content/pdf/
10.1186%2F1687-1499-2011-183.pdf>[Accessed 10
March 2013].
Mulukha, 2011, Vladimir Zaborovsky, Alexey Lukashin,
Sergey Kupreenko, Vladimir Mulukha. Dynamic
Access Control in Cloud Services. Proceedings of The
2011 IEEE International Conference on Systems, Man
and Cybernetics (IEEE SMC 2011), Anchorage,
Alaska, USA, October 9-12, 2011, 1400-1406
Zaborovsky, 2011, Vladimir.Zaborovsky, Vladimir
Mulukha. Access Control in a Form of Active
Queuing Management in Congested Network
Environment. Proceedings of the Tenth International
Conference on Networks (ICN 2011; GlobeNet 2011),
January 23-28, 2011 - St. Maarten, the Netherlands
Antilles. Published by IARIA, 2011, 12-17.
ICINCO2013-10thInternationalConferenceonInformaticsinControl,AutomationandRobotics
574
