IMPORTANCE OF REALISTIC MOBILITY MODELS FOR VANET NETWORK SIMULATION

International Journal of Computer Networks & Communications (IJCNC) Vol.6, No.5, September 2014
IMPORTANCE OF REALISTIC MOBILITY
MODELS FOR VANET NETWORK
SIMULATION
Bahidja Boukenadil¹
¹Department of Telecommunication, Tlemcen University, Tlemcen,Algeria
ABSTRACT
In the performance evaluation of a protocol for a vehicular ad hoc network, the protocol should be tested
under a realistic conditions including, representative data traffic models, and realistic movements of the
mobile nodes which are the vehicles (i.e., a mobility model). This work is a comparative study between two
mobility models that are used in the simulations of vehicular networks, i.e., MOVE (MObility model
generator for VEhicular networks) and CityMob, a mobility pattern generator for VANET. We describe
several mobility models for VANET simulations.
In this paper we aim to show that the mobility models can significantly affect the simulation results in
VANET networks. The results presented in this article prove the importance of choosing a suitable real
world scenario for performances studies of routing protocols in this kind of network.
KEYWORDS
VANET; Mobility Model; Simulations; Real World; MOVE; SUMO; CityMob; NS-2 etc.
1. INTRODUCTION
The mobility is the principal constraint met in the vehicular networks, therefore the performance
investigation of VANET network routing require the exact prediction of mobility of nodes
forming the network, and this is realized by the good choice of the mobility model.
Generally, the traces and the synthetic models [1] are two kinds of mobility patterns used for the
simulation of networks .for simulating the complex and large scale networks that require more
real world scenarios for a real modeling, the traces are used. But in several situations, these traces
are created with difficulty, particularly for the networks of high mobility like the vehicular
networks. In this case, the mobility of nodes (i.e., the vehicles nodes) are modeled with the
synthetic patterns, these models try to really present the behavior of these environments of
vehicles.
A variety of synthetic mobility models for vehicular networks are discussed in Literature, some of
them are presented in Section II. Section III illustrate that a mobility model has a large effect on
the performance evaluation in simulation of VANET network. Finally, Section IV presents some
concluding remarks.
DOI : 10.5121/ijcnc.2014.6513 175

176
2. MOBILITY MODEL FOR VEHICULAR NETWORKS
Many models of mobility can be used to generate the movement of the vehicles instead of the
traces which are difficult to create in the VANET environments .
However, the behavior of these networks is very complex; the drivers must react to the high
change of the road conditions like the congestion, traffic jam, works in the roads. The state of the
roads in turn depends on the behaviors of the drivers. Thus, the choice of the mobility model
affects the relevance and viability of the obtained results.
In general, vehicular traffic simulators or mobility models can be classified in two great kinds, the
microscopic and macroscopic simulators. The microscopic simulator is interested in the
movement of each vehicle taking part in the road traffic. In contrast the macroscopic simulators
are interested in all the vehicles forming the networks; it compute road capacity and the
distribution of the traffic in the road net, by determining a certain parameters like traffic density
(number of vehicles per km per lane) or traffic flow (number of vehicles per hour crossing some
points of interest, usually an areas with a great density of nodes) .
The literature present a large variety of mobility models for VANET networks.
The work of Saha and Johnson [2] model vehicular traffic in real road topologies of the maps of
USA available from TIGER (Topologically Integrated Geographic Encoding) [3] database. The
mobility of nodes is random.Vehicles compute the shortest path to get their destination over the
graph by weighting the cost of displacement on each road on its speed limit and the traffic jam.
Huang et al. [4] studied taxi behavior. In this work, a Manhattan style grid with a uniform block
size is modeled. The streets of simulation area are two-way, with one lane in each direction.
These lanes constrain vehicle movements. A set of parameters characterize the vehicle behavior
which are preferred velocity ,maximum acceleration and deceleration, a velocity variation
associated with the preferred speed at steady state and a list of preferred destinations, i.e., the taxi
stands. . The taxis assigne randomly one of three preferred speeds.
In the work of Choffnes et al. [5] , an integrated mobility and traffic model is conceived, named
STRAW.For modeling real traffic conditions ,this model incorporates a simple car-following
model with traffic control. The road map of this pattern is constructed by street plans with one
lane in each direction .The vehicles move in these lanes according to a random street placement
model.Each vehicle enters the map is placed behind the existing one.
In [6] Haerri et al. proposed a vehicular mobility simulator for VANET networks, called
VanetMobiSim [7] .The speed of vehicles for this model is determined by the use of the
Intelligent Driver Model (IDM).
Three different models were presented in the work of Mahajan et al. [8]: Stop Sign Model (SSM),
Traffic Light Model (TLM) and Probabilistic Traffic Sign Model (PTSM).In this models All
roads are two-way, with one lane in each direction for the SSM and PTSM, whereas TLM model
streets with multiple lanes. An algorithm is employed to reproduce stop signs for each model.
Martinez et al [9] present CityMob [10] .This mobility model generate different mobility patterns
in VANETs, Simple Model (SM), Manhattan Model (MM) and Downtown Model (DM).Each
model has its own characteristics,For the SM,the plan of the road is very simple, neither
semaphores nor direction changes.In constract, MM and DM simulates semaphores at crossing

and random positions in zones with different vehicle density. CityMob allows to simulate
damaged cars and tries to prevent congestions. Figure 1 present the interface of this model.
Karnadi et al. [12, 13, 14] develop a tool MOVE (MObility model generator for VEhicular
networks) [11] .This model provide facility for the users to generate real world scénarios in
VANET environments. This model contains two editors, one for road map and the second for
vehicle movement .The user can manually and automatically create the road topology or import it
from existing real world maps such as Google maps. For vehicle movement, the pattern can be
manually created, generated automatically or specified based on a bus time table to simulate the
movements of public transportations. Figure2 shows the interface of this model.
177
Figure 1. Interface of CityMob

178
Figure 2: Mobility Model Generator
3. INFLUENCE OF MOBILITY MODEL IN VANET
SIMULATIONS
.
The simulation results are greatly affected by the mobility model selected .The results presented
in this section illustrate the importance of choosing a suitable mobility model for performances
evaluation of routing protocols in VANET network.
We use ns-2 [34] to compare the performance of the CityMob Model and the MOVE tool via a
simulation.The routing of packets is accomplished with the Ad hoc On Demand Distance Vector)
(AODV) [20]. We have chosen the parameters for both mobility models in a manner to simulate
the same path as possible.
AODV (Ad hoc On Demand Distance Vector) [1] is a reactive protocol that does route discovery
when needed by a node. The route discovery process is initiated only when a source node has data
traffic to send to a destination node, that makes AODV a truly On- Demand routing protocol.
In our simulations, we chose AODV since it behaves well in several of the performance
evaluations of the routing protocols (e.g. [21, 22, 23]).

The ns-2 code used in our simulations of AODV was obtained from [24].
Each simulation run lasted for 300 seconds with a uniform block size of 500 x 500 meters; the
maximum speed of vehicles is of 40 m/s.
The number of vehicles nodes from 15 to 100. We chose the traffic sources at a constant rate
CBR (Constant Bit Rate). Traffic between nodes is generated using a traffic generator which is
characterized by the following parameters: Data packets Size is 512 bytes, Interval between
packets is 0.25s , maximum of packets transmitted 1000. All nodes use IEEE 802.11 MAC
operating at 2Mbps. The propagation model employed in the simulation is TwoRayGround
reflextion.
In our comparison of the two mobility models, we consider the following performance metrics
obtained from the AODV protocol: throughput, end-to-end delay and protocol overhead.
179
0 20 40 60 80 100
0,023
0,022
0,021
0,020
0,019
0,018
0,017
0,016
0,015
0,014
0,013
0,012
0,011
0,010
0,009
Move
CityMob
Throughput (bits/s)
Number of vehicles
Figure 3 : Throughput vs. number of vehicles.
0 20 40 60 80 100
0,50
0,45
0,40
0,35
0,30
0,25
0,20
MOVE
CityMob
Average End-to-End Delay (sec)
Number of vehicles
Figure 4 : End-to-end delay vs. number of vehicles.

180
0 20 40 60 80 100
110
100
90
80
70
60
50
40
MOVE
CityMob
Overhead
Number of vehicles
Figure 5 : overhead vs. number of vehicles.
Figures 3, 4 and 5 illustrate the performance (i.e., Throughput ratio, end-to-end delay and
overhead) of AODV with the two mobility models chosen.The Throughput (in figure 3) of
AODV when using MOVE mobility model is higher and more stable than when using CityMob
model.The trace 4 shows that AODV causes a low stable delay with MOVE because the roads are
more defined compared to CityMob.
Figures 5 illustrate the overhead AODV required with each of the chosen mobility models. The
vehicles moving with CityMob have a higher overhead, as a result this model requires a higher
amount of overhead compared to MOVE.These results confirm the suitability of MOVE tool for
simulating VANET.
4.CONCLUSIONS
In this paper, we compared the performance of two mobility models for VANET Simulation i.e.
MOVE (MObility model generator for VEhicular networks) and CityMob (City
Mobility).Simulation analysis using realistic mobility model for VANET environment show that
the performance of the protocol is greatly affected by the mobility model. The performance of an
ad hoc network protocol can vary significantly with different mobility models then, the choice of
mobility model in simulating VANET is very important.The mobility models for VANET should
be most closely match the expected real-world scenario. In fact, the realistic scenarios can aid the
development of the routing protocols significantly.
As future work we plan to compare other mobility models discussed above. Results obtained from
these studies would certainly facilitate in meeting the challenges associated with future
development and evaluation of suitable routing protocols in vehicular networks.

181
REFERENCES
[1] M. Sanchez and P. Manzoni. “A java-based ad hoc networks simulator. In Proc. of the SCS
Western Multiconference Web-based Simulation Track, Jan. 1999
[2] A.K. Saha and D.B. Johnson, “Modeling mobility for vehicular ad hoc networks,” in ACM
Workshop on Vehicular Ad Hoc Networks (VANET 2004), Philadelphia PA, October 2004.
[3] U.S. Census Bureau - Topologically Integrated Geographic Encoding and Referencing (TIGER)
system, http://www.census.gov/geo/www/tiger.
[4] E. Huang,W. Hu, J. Crowcroft, and I.Wassell, “Towards commercial mobile ad hoc network
applications: A radio dispatch system,” in Sixth ACM International Symposium on Mobile Ad
Hoc Networking and Computing (MobiHoc 2005), Urbana-Champaign, Illinois, May 2005
[5] D.R. Choffnes and F.E. Bustamante, “An integrated mobility and traffic model for vehicular
wireless networks,” in ACM Workshop on Vehicular Ad Hoc Networks (VANET 2005),
Cologne, Germany, September 2005.
[6] J. Haerri, M. Fiore, F. Filali, and C. Bonnet, “Vanetmobisim: generating realistic mobility
patterns for vanets,” in ACM Workshop on Vehicular Ad Hoc Networks (VANET 2006), Los
Angeles, California, September 2006.
[7] VanetMobiSim.http://en.pudn.com/downloads160/sourcecode/app/detail720214_en.html
[8] A. Mahajan, N. Potnis, K. Gopalan, and A.Wang, “Evaluation of mobility models for vehicular
ad-hoc network simulations,” in IEEE International Workshop on Next Generation Wireless
Networks (WoNGeN 2006), Bangalore, India, December 2006.
[9] Francisco J. Martinez, Juan-Carlos Cano, Carlos T. Calafate, Pietro Manzoni, “CityMob: a
mobility model pattern generator for VANETs,” in the ICC 2008 workshop proceedings.
[10] CityMob’s source code is available at http://www.grc.upv.es/
[11] F. Karnadi, Z. Mo, K.-C. Lan, .Rapid Generation of Realistic Mobility Models for VANET.,
Poster Session, 11th Annual International Conference on Mobile Computing and Networking
(MobiCom 2005), Cologne, Germany, August 2005.
[12] F. K. Karnadi, Z. H. Mo, and K. c. Lan, “Rapid generation of realistic mobility models for
vanet,” in IEEE WCNC,2007, pp. 2506–2511.
[13] Karnadi, F.K.; Zhi Hai Mo; Kun-chan Lan; Sch. of Comput. Sci. & Eng., New South Wales
Univ., Sydney, NSW, “Rapid Generation of Realistic Mobility Models for VANET”. IEEE
Xplore
[14] MOVE http://www.cs.unsw.edu.au/klan/move/.
[15] SUMO Simulation of Urban MObility. http://sumo.sourceforge.net/.
[16] A. Uchiyama, .Mobile Ad-hoc Network Simulator based on Realistic Behavior Model, 6th ACM
International Symposium on Mobile Ad Hoc Networking and Computing (MobiHoc 2005),
Urbana Champaign, IL, USA, May 2005.
[17] Deutsches Zentrum f¨ur Luft-und Raumfahrt e.V. (DLR). Sumo– simulation of urban mobility.
http://sumo.sourceforge.net/.
[18] The Network Simulator ns 2. http://www.isi.edu/nsnam/ns/index.html.
[19] QualNet Network Simulator. http://www.scalable-networks.com/.

[20] Perkins, C. and Royer, E. (1999). Ad hoc On Demand Distance Vector (AODV) Routing. In
Proceedings 2nd Workshop on Mobile Computing Systems and Applications. New Orlean s, LA,
USA: IEEE, February 1999, pp. 90–100.
[21] J. Broch, D. Maltz, D. Johnson, Y. Hu, and J. Jetcheva. Multi-hop wireless ad hoc network
routing protocols. In Proceedings of the ACM/IEEE International Conference on Mobile
Computing and Networking (MOBICOM), pages 85–97, 1998.
[22] B.Boukenadil, M.Feham, “Comparison between DSR, AODV and DSDV in VANET using
CityMob”, in Proc. the 1th International Conference on New Technologies and Communication
(ICNTC’2012), Chlef, Algeria, 05-06 December 2012.
[23]P. Johansson, T. Larsson, N. Hedman, B. Mielczarek, and M. Degermark. Routing protocols for
mobile ad-hoc networks - a comparative performance analysis. In Proceedings of the ACM/IEEE
International Conference on Mobile Computing and Networking (MOBICOM), pages 195–206,
1999.
[24]The Rice University Monarch Project. Monarch extensions to the ns simulator. URL:
182
http://www.monarch.cs.rice.edu/. Page accessed on May 30th, 2002

IMPORTANCE OF REALISTIC MOBILITY MODELS FOR VANET NETWORK SIMULATION

More Related Content

What's hot (19)

Viewers also liked (19)

Similar to IMPORTANCE OF REALISTIC MOBILITY MODELS FOR VANET NETWORK SIMULATION (20)

More from IJCNCJournal (20)

Recently uploaded (20)

IMPORTANCE OF REALISTIC MOBILITY MODELS FOR VANET NETWORK SIMULATION