Towards More Realistic Mobility Model in Vehicular Ad Hoc Network

Description

International Journal of Computer Science and Information Security (IJCSIS) provide a forum for publishing empirical results relevant to both researchers and practitioners, and also promotes the publication of industry-relevant research, to address the significant gap between research and practice. Being a fully open access scholarly journal, original research works and review articles are published in all areas of the computer science including emerging topics like cloud computing, software development etc. It continues promote insight and understanding of the state of the art and trends in technology. To a large extent, the credit for high quality, visibility and recognition of the journal goes to the editorial board and the technical review committee. Authors are solicited to contribute to the journal by submitting articles that illustrate research results, projects, surveying works and industrial experiences. The topics covered by this journal are diversed. (See monthly Call for Papers) For complete details about IJCSIS archives publications, abstracting/indexing, editorial board and other important information, please refer to IJCSIS homepage. IJCSIS appreciates all the insights and advice from authors/readers and reviewers. Indexed by the following International Agencies and institutions: EI, Scopus, DBLP, DOI, ProQuest, ISI Thomson Reuters. Average acceptance for the period January-March 2012 is 31%. We look forward to receive your valuable papers. If you have further questions please do not hesitate to contact us at ijcsiseditor@gmail.com. Our team is committed to provide a quick and supportive service throughout the publication process. A complete list of journals can be found at: http://sites.google.com/site/ijcsis/ IJCSIS Vol. 10, No. 3, March 2012 Edition ISSN 1947-5500 � IJCSIS, USA & UK.

Shared by: ijcsiseditor

Stats

views:: 96
posted:: 5/15/2012
language:: English
pages:: 8

Public Domain

Document Sample

(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 3, March 2012

Towards More Realistic Mobility Model
in Vehicular Ad Hoc Network
Dhananjay S. Gaikwad, MTech, Mahesh Lagad, MTech(Appear)
Assistant professor, dhananjayg63@g mail.com Assistant professor, maheshlag ad@gmail.com

Prashant Suryawanshi, MTech(Appear) Vaibhav Maske, MTech(Appear)
Assistant professor, s prashant1234@gmail.com . Assistant professor, vamaske@gmail.com

Co mputer Engineering Depart ment,
HSBPVT‟S GOI, Co llege of Engineering,
Kashti. India- 414701.

Abstract— Mobility models or the movement patterns of packets need to pass through several nodes to reach
nodes communicating wirelessely, play a vital role in the destination [2]. Veh icular Ad Hoc Network
simulation-based evaluation of vehicular Ad Hoc (VANETs) are a special case of Mobile Ad Hoc
Networks (VAN ETs). Even though recent research has Network (MANETs) and consist of a nu mber of
developed models that better corresponds to real world
mobility, we still have a limited understanding of the vehicles traveling on urban streets capable of
level of the required level of mobility details for communicat ing with each other without fixed
modeling and simulating VANETs. In this paper, we infrastructures. VA NETs are expected to benefit
propose a new mobility model for VANETs that works safety applications, gathering and disseminating real-
on the city area and map the topology of streets and time traffic congestion and routing information,
behavior of vehicles at the intersection of roads. Our informat ion services Such as transparent connection
model change the speed of nodes after some specific to internet etc [3].
distance in accordance to neighboring nodes that is One critical aspect of VA NETs simulat ion is the
according to a desity of nodes, so that this will lead to a
realistic situation on the roads. Our model accounts the
movement pattern of vehicles, also called mobility
various characteristics of VANETs such as traffic lights, models. Mobility model determine the location of
acceleration/deceleration due to nearby vehicles, nodes in the topology at any given instant, which
attraction points where maximum numbers of vehicle strongly affects network connectivity and throughput
tends to go. Using the real and controlled map of street, [5]. There are several mobility models such as
we compare our mobility model with the random random pattern, graph constrained commonly used in
direction mobility model. Our result demonstrates that popular wireless simu lators such as ns2 [16] by
probability of link availability in VAN ETs is more VA NET researchers [4]. But one problem with these
sensitive to the vehicles waiting at intersections and models is that they ignores some critical aspects of
acceleration/deceleration of vehicles. We also found that
probability of link availability suffers at the interse ction
the real world traffic such as queuing of vehicles at
of the roads; because of some nodes cross the signal road intersection, traffic lights and traffic signs,
continue movement in horizontal direction while some acceleration and deceleration according to neighbor
nodes change the direction of traveling to vertical. vehicles. Mobility models should reflect as possible
as the real behavior of vehicular t raffic on the road
Keywords- Vehicular Ad Hoc Network (VANETs), [1]. In this paper, we propose a new mobility model,
Mobile Ad Hoc Network (MANETs). which in corporates important features of mobility
model on the road, such as presence of traffic light on
I. INTRODUCTION the road, node movement is restricted to the road
structure and speed changes in accordance to the
Ad hoc network is a collection of wireless mobile neighboring vehicle.
nodes without any fixed base station infrastructure The rest of paper is organized as follows:
and centralized management. Each node acts as both section 2 describes some currently used mobility
host and router, which moves arbitrarily and models and some tools for generation of mobility
communicates with each other via mu ltiple wireless models. Section 3, describes our proposed mobility
lin ks. It is a mult i-hop wireless network, where model and simu lation of our model. Finally Section 4

83 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 3, March 2012

concludes the paper. difficult to co llect the real t ime traces of all the
nodes. Follo wing section describe mobility models
II. M OBILITY MODEL OVERVIEW that generate the trace file, contain the traces of
Mobility model reflects the behavior of the nodes vehicles movement.
throughout the simu lation time. It shows how the 2.1.1 Mobility model generator for Vehicular Network
nodes change their speed and direction in (MOVE): Mobility model generator for Vehicular
accountancy to the neighboring vehicles and Network (MOVE) [12] facilitates users to rapidly
according to the traffic rule. Fo llo wing are some generate realistic mobility models for VANET
important factors those affect the mobility of nodes in simulation with a visualization property. This model
VA NETs. works with another micro-simulator traffic model,
 Street structure: Streets force nodes to called SUMO [13]. MOVE model consists of two main
confine their movements to well-defined components: Map Editor and Vehicle Movement Editor.
paths. This constrained movement pattern Map Editor is used to create the road topology, which is
determines the spatial distribution of nodes either created by manually, automatically or by
and their connectivity. Streets can have importing the maps from databases such as TIGER
either single or mu ltip le lanes and can allow
((Topologically Integrated Geographic Encoding and
 either one-way or t wo-way t raffic. 
Referencing). The Vehicle Movement Editor used for
 Block size: A city block can be considered
the generation of vehicle movement. The output of
the smallest area surrounded by streets. The
MOVE is a mobility trace file which contains the
block size determines the number of
intersections in the area, wh ich in turn information on vehicle movement that can be used by
determines the frequency with which a network simulator. All the parameter configuration of
vehicle stops.  vehicle movement is done in a static way. This model
 Traffic control mechanisms: The most does not consider micro – mobility features.
common t raffic control mechanisms at 2.1.2 Street Random Waypoint (STRAW): Street
intersections are stop signs and traffic lights. Random Way point (STRAW) is a tool [14] that
These mechanisms result in the format ion of generates the mobility patterns with extraction of urban
clusters and queues of vehicles at topologies from the TIGER database. It supports for the
intersections and subsequent reduction of micro – mobility features of models. STRAW
 their average speed of movement.  implements a complex intersection management using
 Interdependent vehicular motion: Movement traffic lights and traffic signs. Due to this characteristic,
of every vehicle is influenced by the vehicle shows a more realistic behavior when reaching
movement pattern of its surrounding at intersection. It includes a traffic control mechanisms
 vehicles.  that force drivers to follow deterministic admission
 Average speed: The speed of the vehicle control protocol when encountering intersection.
determines how quickly its position changes,
Drawback of STRAW model is it does not give details
which in turn determines the rate 
of network topology changes. about the traffic flows. Also it does not specify the lane
Whenever a mobility model is designed for changing behavior.
VA NETs, that model should consider the factors 2.2 Entity mobility model
that affect the mobility of node. These types Entity mobility model represents mobile
based on the number of nodes considered while node as a random entity wh ich moves randomly over
designing the model and the way mobility the observed area, where speed and directions of
informat ion is stored. node independent with the neighboring nodes.
2.1 Trace based mobility model 2.2.1 Random Walk mobility mode: Random Walk
This type of model [6] is suitable to emulate mobility model is Entity mobility model, in which
the real scenarios in MANET and VA NET. Traces mobile node moves from its current location to a new
describe the movement of vehicles throughout the location by randomly choosing a direction and speed
simu lation. Traces are the best information to find the in which to travel [4]. The new speed and direction
mobility patterns of node, if we have traces of long are both chosen from pre-defined ranges, respectively
period and involvement of many participants. Traces [min-speed,max-speed] and [0,2*pi] respectively.
reflect the movement histories of the nodes in the Each movement in the Random Walk Mobility
network. We can expect mobility patterns provided Model occurs in either a constant time interval T or a
by them lead to realistic mobility modeling. But the constant traveled distance, at the end of which a new
VA NET applications are not widely deployed; there direction and speed are calculated.
are fewer traces for evaluation. Another issue related 2.2.2 Random way point model: In Random way point
to traces is, the nature of network is decentralized and model [7], mobile nodes move randomly and freely

84 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 3, March 2012

without any restrictions. In this model, the destination, phase, node independently selects its new direction
speed and direction all are chosen randomly and and speed of movement. Speed and d irection are kept
independent of other nodes. The fraction of nodes in constant for the whole duration of node move phase.
network remains static for the entire simulation time. 2.3 Group Mobility model
The velocity of node is uniformly chosen at random Entity mob ility models represent mult iple
from the interval [Vmin, Vmax]. The node moves mobile nodes whose actions are completely
towards destination with a velocity v. When it reaches independent of each other. In an ad hoc network,
to destination, it remains static for the predefined pause however, there are many situations where it is
time and moving again according to same rule. The necessary to model the behavior of mobile nodes as
mobility behavior of nodes very much depends on the
they move together. For example, a group of soldiers
pause time and maximum speed of nodes. The
in a military scenario may be assigned the task of
parameters to describe a simulation setup of model are
searching a particular plot of land in order to destroy
Size and shape of the deployment region Q, initial
land mines, capture enemy attackers. In order to
spatial node distribution Fint (x), static parameter ps,
model such situations, a group mobility model is
with 0< ps<1, Probability density function fTp (tp) of
pause time, Minimum and maximum speed : 0 < Vmin needed to simulate this cooperative characteristic. In
≤ Vmax. this section, we present reference group mobility
The components of node distribution fx(x) is models.
composed of three distinct components as shown in Reference Point Group Mobility model: The
equation. (fx(x)=fs(x)+fp(x)+fm(x) …(1) Reference Point Group Mobility model represents the
2.2.3 Gauss Markov model: Random way-point random mot ion of a group of mobile nodes as well as
model generates speed and direction of nodes the random motion of each indiv idual mob ile node
independent on previous history. It directly selects within the group. Group movements are based upon
speed and direction fro m its predefined range, so this the path traveled by a logical center for the group. It
can create a sudden stop and sharp turn problem. is used to calculate group motion via a group motion
Gauss Markov model [8] first calculates the speed vector, GM. The motion of the group center
and direction of movement for each node. Then completely characterizes the movement of this
nodes move with the calculated speed and direction corresponding group of mobile nodes, including their
direction and speed. Individual mobile nodes
for a period. After that period similar movements
randomly move about their own pre-defined
begins again. The time that is used in the movement
reference points whose movements depend on the
in each interval before the change in speed and
group movement [11]. As the indiv idual reference
direction, is constant. The current speed and direction
point move fro m time t to t+1, their locations are
related to the previous speed and direction shown by
updated according to the group's logical center. Once
equation (2) and (3). the updated reference group points, RP (t+1) are
 s n = α sn -1 + ( 1- α) * s + (1 -α 2) * s xn- 1 ½ (2)calculated, they are co mb ined with a random vector,
 d n = α d n-1 + (1 -α )* d + ( 1- α2 ) * s dn -1 ½ (3)RM, to represent the random motion of each mob ile
node for mo ve individual pe riod time n, s n an d d n -1 ar
Wh er e, s n an d d n ar e the valu es o f sp ee d a nd direction about its m ent in the reference point.-1The lengthe the
values of speed and direction for movement in the of RM is uniform distributed within a specified
period time n-1, α is the constant value in the range
[0, 1], s and d are constants representing the mean radius centered at RP (t+1) and its direction is
speed and direction, α sn-1 and αdn-1 are variables fro m uniformly distributed between 0 and Pi.
a Gaussian distribution. Gauss Markov model
overcomes sudden stop and sharp turn problems of 3. PROPOSED MOBILITY M ODEL
Random way point model [7].
2.2.4 Random direction mobility model: Random
3.1 Our proposed model
direction mobility model is, bes ides the random
We develop our model by considering the
waypoint model, p robably the most widely used real scenarios on the roads. We consider several
model. This model considers individuals moving on parameters of real t raffic situation on the roads of
straight walk segments with constant speed and city. The parameters attraction point, speed variation,
optional pauses between the walk segments. In this traffic light are considered. We change the speed of
model each node alternates periods of movement nodes after some specific distance according to the
(move phase) to periods during which it pauses density of the nodes. We model the signal based on
(pause phase) [10]; at the beginning of each move the horizontal traffic time stamps and vertical traffic
time stamps.

85 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 3, March 2012

Attraction point: In real scenarios, vehicles do not intersection point, to handle the traffic in both the
move randomly fro m one point to another. They used direction.
to set some fixed destination to reach. We consider Node movement: At the beginning, the nodes are
this fact in our proposed model. Vehicles generally distributed along the starting points of the horizontal
move in deterministic way. Whenever vehicles enter in and vertical lane. Nodes are allowed to move only
the city, they not always move in straight line. They through the predefined paths. Node knows the
may change their direction of movement to a specific distance from the origination point or arriving point
point according to the importance of that point. Suppose of traffic to the signaling point. The nodes have
if the vehicle rider is a student, then he will definitely assigned probability values according to their point
move to the college road. Like this if vehicle rider want of interest. Each node calculates the time required to
to go restaurant, he will move to the restaurant road. reach at signal. The traffic signal is modeled on the
Users move in group‟s towards the attraction points. time stamp basis. At the signal, each node checks its
More number of nodes will be around the most next direction of movement according to the
attractive point compare to less attractive.
probability value and attraction point, i.e. destination
In our model we have modeled the attraction point.
point as the function of probability value at signaling
3.2 Operation of proposed mobility model
point. In our model, each vehicle has assigned some
probability values whenever they enter in the
In this section, we describe algorith m of movement
simu lation area according to their attraction behavior.
of mobile nodes of proposed mobility model.
The checking of the probability value is done at the
Initially, all nodes start from the init ial point on the
signaling point.
road and moves up to the boundary of the simulat ion
Speed variation : Generally whenever vehicles enter
area.
in the simulation area as city area in our proposed
model, their speed do not remains constant 1. Define all the necessary variables signal
throughout the simulation t ime and area. The speed time, signal coordinates, number of nodes,
of vehicle is changes in according to the neighboring area of simu lation, TURN.
vehicles, traffic lights, street layout, and pedestrian 2. Do for all nodes, fro m 1 to n, where n is
movement.
number of nodes in the simulation.
In our proposed model, we change the speed
3. Node movement starts fro m in itial point
of the vehicle after some specific distance. So it gives towards the boundary of simulat ion area in
a scenarios that a vehicle changes its speed in both directions horizontal and vertical.
according to the neighboring vehicles. So when any
vehicle co mes closer to already moving vehicle 4. Set distance after which the node speed will
for the overtaking, the moving vehicle will decrease change to 15m and speed of node= rando m
the speed for some time and overtaking node will
increase the speed at that time. Th is speed variation value between {5m/s and 25m/s}.
parameter will increase the realis m of our proposed
a. Calculate the time required to
model. travel given dist by the formu la,
Traffic light: Traffic light is for the management of time=d ist/speed.
intersection on the roads. In our proposed model, we b. Add time and dist to the total time
have modeled the traffic light as a coordinated traffic and total distance of the node
light. For that first consider the single horizontal lane.
respectively.
The light turns green in such a manner that only 5. If node_distance==signal, then
traffic along the single lane cross the intersection
Check the time of signal and the probability
simu ltaneously. Veh icles that need to turn left will go
value of node.
directly. We have modeled the signal in such that
If signal_time==node_time (that means
when the traffic light turns red, the vehicles that need
signal is red) and p roba bility valu e! = TU R N.
to cross the signal will wait other vehicle directly
Then wait for the remain ing time till the
take turn.
signal turns green and go to step 8.
Simulation area: In our proposed model, we have
Otherwise, cross the signal without wait ing.
considered the area of city for the deployment of our
A nd t ak e th e tur n a n d go in up w a r d dir ectio n
model. In that vehicles enter the simulat ion area on
with out sto ppi ng at sig n al a nd g o to ste p 8 .
the left side of the road. A vehicle then moves 6. Continues the nodes movement and go to
towards the right direction with speed range fro m step 5.
5m/s to 25m/s. There are traffic lights at the 7. While (end of simu lation area).

86 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 3, March 2012

8. Take a movement for other nodes and go to node moves towards the end of simu lation area and
time for the rando m d irection mob ility model. This is
step 2.
because of each node takes random direction
9. While (nu mber o f nodes)
independent on the previous node and that direction is
10. End not restricted to the particular road topology. But in our
11. Now d istance matrix contains the time and
proposed model, the probability of link availability
distance values for nodes. decreases only at the signaling point and probability
12. Take the values fro m the distance matrix. value remains high for rest time of simulation. This
13 Calculate the probability of link availability Table 2
and average number of neighbors per node.
14 Co mpare these values with random Parameter Values
waypoint, gauss Markov, random direction Simulation time 100 seconds
and city section models. Number of nodes 10,20,30,40,50,75
MATLAB version MATLAB 7.8.0.347
15 Show the results.
Mobility model characteristics Horizontal and vertical lanes.

3.3 Co mparison of mobility model Randomly assigned between 0
popularity of attraction points
to 1.
Table 1 MAXSPEED 25 m/s.
MINSPEED 5 m/s.
Fea tures Random waypoint Our model
model Dist anc e aft er that spe ed
10m, 15m.
Horizontal No horizontal and Horizontal and chang es
and vertical vertical lane present,
vertical lane, nodes are
lane which shows the road Our proposed model and
move randomly around Mobility model
of the city. Random direction model.
simulation area.
Cross point No cross point Cross points are
present, show the road would give the actual situation on the roads.
intersection.
When we increase the number of nodes fro m 20 to
Att ra ction No attra ction point Check the probability
point of the nodes. 30, the figure 2 shows the results. The results indicate
Speed Const ant for som e Change after some that, there is improvement in the probability of link
variation time interval specific distance.
availability. This is due to the increase in the density
Table 1 shows broadly the comparison between our of nodes.
proposed mobility model and random waypoint
B. Behavior of our mobility model with varying
model.
number of nodes.
3.5 Simulation results Case 1: This section shows the behavior of our
We performed simulation using the proposed mobility model, by varying the number of
MATLAB. All nodes position is shown by their x
nodes, while keeping the speed constant between
and y coordinate values. We are not taking into
5m/s to 25m/s. The figure 3 shows the results. The
account the third dimension (z- direct ion) of position.
So nodes are assumed to move in t wo dimensional results indicate that the probability value increases
planes all the time. All nodes initialized by their with the increase in the nu mber of nodes in the
initial position and make them travel to a specified simu lation. This is because the density of nodes
destination point. Simu lation parameters are shown in increases number of connections between the nodes.
the table 2. Case 2: This section shows the behavior of our
A. Varying number of nodes model with varying nu mber of nodes and increasing
This section compares the mobility models with the speed of nodes. The figure 4 shows the results
different number of nodes in 800mX800 road with the increase in the speed of nodes. The speed of
topology. Figure 1 co mpares our proposed mobility nodes is in between 0m/s to 25m/s. The result
model with random d irect ion models with 20 indicates that there is imp rovement in probability of
numbers of nodes.
The results indicate that probability of link lin k availability value for 10 and 20 nodes. But the
availability is h igher at the init ial time of simulat ion probability values decrease for 30 and 40 nodes and
and it is gradually decreases as time passes i.e. as again increases for 50 nodes. So by comparing results

87 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 3, March 2012

in figure 3 and figure 4, we can show that the speed The figure 5 and 6 show the average number of
of the nodes play an important ro le during the neighbors per node throughout the simu lation t ime.
communicat ion. If we increase the speed of nodes The number of neighbors is being observed for
that does not mean that we get the better results. So increased level of density. The average number of
neighbors per nodes varies smoothly fro m start to end
the performance of routing protocols very much
of the simu lation time. At some point in the
depends on the speed and density of the nodes. simu lation, the value decreases,

Figure 4: probability of link availability versus
number of simul ated nodes.
Figure 1: Probability of Link Availability versus Time. this is due to the intersection of the road. Our model
shows that the number of neighbors decreases only at
the road intersections, as some nodes change the
direction of traveling. W ith the increase in the density
of the nodes, there is increase in the numbers of
neighbors per node. This is due to that more nu mber
of neighbors per node increase as there are more
nodes on the road. Our proposed model gives the
better results in terms of average number of
neighbors per node. This will give mo re stability to
our model. Our model shows better results in all the
experiments, while showing the real situation on the
road.

Figure 2: Probability of Link Availability Versus Time.

Figure 5: Average numbers of neighbors per node for 50 nodes
Figure 3: Probability of Link Availability versus Number
of Simulate d Nodes.

Case 3: Average number of neighbors per node :-

88 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 3, March 2012

REFERENCES
[1]. Harri, J.; Filali, F.; Bonnet, C.; "Mobility models for vehicular
ad hoc networks: a survey and taxonomy," Communications
Surveys & T utorials, IEEE, vol.11, no.4, pp.19-41, Fourth Quarter
2009
[2]. Azarmi, M.; Sabaei, M.; Pedram, H., "Adaptive routing
protocols for vehicular ad hoc networks," Telecommunications,
2008. IST , 2008. International Symposium on, vol., no., pp.825-
830, 27-28 Aug. 2008.
[3]. Yufeng Chen; Zhengtao Xiang; Wei Jian; Weirong Jiang; ,
"An improved AOMDV routing protocol for V2V
communication," Intelligent Vehicles Symposium, 2009 IEEE ,
vol., no., pp.1115-1120, 3-5 June 2009.
[4]. http://en.wikipedia.org/wiki/Mobility_model.
[5]. Mouzna, J.; Uppoor, S.; Boussedjra, M.; Pai, M.M.M.,
Figure 6: Ave rage numbers of neighbors per node for 75 "Density aware routing using road hierarchy for vehicular
nodes. networks," Service Operations, Logistics and Informatics, 2009.
SOLI „09 IEEE/INFORMS International Conference, vol., no.,
pp.443-448, 22-24 July 2009.
[6]. Vetriselvi, V. and Parthasarathi, R., “Trace based mobility
model for ad hoc networks,” In Proceedings of the Third IEEE
4. Conclusion and Future Work international Conference on Wireless and Mobile Computing,
In this research, we have proposed the new Networking and Communications.WIMOB, IEEE Computer
Society, Washington, DC, pp.81-81, 2007.
mobility model that covers the city area. In our [7] Christian Bettstetter, Giovanni Rasta and Paolo Santi,; “The
proposed model we change the speed of nodes after Node Distribution of the Random Waypoint Mobility Model for
some particular distance, in accordance to Wireless Ad Hoc Networks”, IEEE TRANSACT IONS ON
MOBILE COMPUTING, VOL. 2, NO. 3, JULY-SEPTEMBER
neighboring nodes. We assigned probability values to 2003.
the nodes based on the attraction point, where nodes [8] Ariyakhajorn, Jinthana; Wannawilai, Pattana;
Sathitwiriyawong, Chanboon; , "A Comparative Study of Random
most likely to move. This would leads to the actual Waypoint and Gauss-Markov Mobility Models in the Performance
scenarios on the road. Evaluation of MANET," Communications and Information
Through simu lation we have shown that our Technologies, 2006. ISCIT '06. International Symposium on , vol.,
no., pp.894-899, Oct. 18 2006-Sept. 20 2006.
model performs better than random direction model [9] Boundless mobility model, http://www-
in terms probability of lin k availab ility. We public.itsudparis.eu/~gauthier/MobilityModel/mobilitymodel.ht
compared our proposed model with rando m direction ml #Boundless.
[10] Zhi Ruxin; Gao Fei; Yang Jie; "Nonuniform Property of
model through the simu lation. Fro m the simulation Random Direction Mobility Model for MANET, " Wireless
we got the result that shows that probability of lin k Communications, Networking and Mobile Computing, 2009.
WiCom '09. 5th International Conference on , vol., no., pp.1-4, 24-
availability decreases at the traffic signal point. Th is 26 Sept. 2009.
is because of the vehicle either waits or takes turn [11] Ng, J.M.; Yan Zhang; , "Reference region group mobility
and change the direction of traveling. Our result model for ad hoc networks," Wireless and Optical
Communications Networks, 2005.
demonstrates that probability of link availab ility in [12] F. Karnadi, Z. Mo, K.-C. Lan,”Rapid Generation of Realistic
VA NETs is more sensitive to the vehicles wait ing at Mobility Models for VANET ”, Poster Session, 11th Annual
intersections and acceleration/deceleration of International Conference on Mobile Computing and Networking
(MobiCom 2005), August 2005.
vehicles. We have tested our model by increasing [13]SUMO,
speed of the nodes. We found that the connectivity http://sourceforge.net/apps/mediawiki/sumo/index.php?title=Main
among the nodes is very much depends on the speed _Page.
of the nodes. We also found that performance of [14] D. Choffnes, F. Bustamante, “An Integrated Mobility and
Traffic Model for Vehicular Wireless Networks”, 2nd ACM
mobility model is depends on both the speed and Workshop on Vehicular Ad Hoc Networks (VANET 2005),
density of the nodes. Thus our model tries to depict September 2005.
the realistic scenarios on the real road. [15] CANU Project Home Page, http://canu.informatik.uni-
In future, we will try to extend our model by stuttgart.de.
[16] http://www.isi.edu/nsnam/ns/index.html.
considering the overtaking parameter into account.
[17] http://pcl.cs.ucla.edu/projects/glomosim/.
We will try to run our model on a two lane of the [18] Gainaru, A.; Dobre, C.; Cristea, V.;"A Realistic Mobility
road. Model Based on Social Networks for the Simulation of VANET s,"
Vehicular Technology Conference, 2009. VT C Spring 2009.IEEE
th
69 ,vol. no., pp.1-5, 26-29 April 2009.
[19] M. T reiber, A. Hennecke, D. Helbing,;”Congested traffic
states in empirical observations and microscopic simulations”,
Phys. Rev. E62, Issue 2, August 2000.
[20] Bhandari, Shiddhartha Raj; Lee, Gyu Moung; Crespi,
Noel; , "Mobility Model for User's Realistic Behavior in Mobile
Ad Hoc Network," Communication Networks and Services
Research Conference (CNSR), 2010 EighthAnnual , vol., no.,
pp.102-107, 11-14 May 2010.

89 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 3, March 2012

AUTHORS PROFILE

My self Dhananjay Gaikwad,
completed Mtech from National Institute of
Technology, surat (India) 2011. Since july 2011, I am
working as Assistant professor in Parikram college of
engineering, kashti. My papers have been published
in Springer and IEEE explorer.

My self Mahesh Lagad, pursuing
Mtech (MS) from University of Pune (India). I have
two year industrial experience, joined the college in
july 2011. My paper has been published in
International conference.

I, Prashant suryawanshi currenty
doing Mtech, from Hyderabad (India). I have total six
year Industrial plus academic experience.

I, Vaibhav maske currenty doing
Mtech, from Hyderabad (India). I have total four
year teaching experience.

90 http://sites.google.com/site/ijcsis/
ISSN 1947-5500