Message Propagation in Vehicular Ad hoc Networks under free flow and congested flow Scenarios

International Journal of Ad hoc, Sensor & Ubiquitous Computing (IJASUC) Vol.2, No.1, March 2011
DOI : 10.5121/ijasuc.2011.2101 1
Message Propagation in Vehicular Ad hoc
Networks under free flow and congested flow
Scenarios
Sara Najafzadeh, Norafida Ithnin, Shukor Abd Razak, Ramin Karimi
Faculty of Computer Science and Information System
University Technology Malaysia
Email :{ sara60, nithnin, sabdrazak, rakarimi1}@gmail.com
Abstract. Vehicular Ad hoc Network (VANET), a subclass of mobile ad hoc networks (MANETs), is a
promising approach for the intelligent transportation system (ITS). The key difference of VANET and
MANET is the special mobility pattern and rapidly changeable topology. Temporary network
fragmentation problem and the broadcast storm problem are further considered for designing Message
propagation in VANETs. The temporary network fragmentation problem which happens in free flow caused
by rapidly changeable topology influence on the performance of data transmissions. The broadcast storm
problem in congested flow network seriously affects the successful rate of message delivery in VANETs.
This paper presents a connectivity analysis under free flow and congested flow scenario in Vehicular Ad
hoc Networks.
.
Key words: Vehicular Ad hoc Network, Message Propagation, free flow, congested flow
1. Introduction
Vehicular Ad hoc Networks (VANET), a new technology to build a wireless network between
vehicles (V2V). VANETs are based on short-range wireless communication (e.g., IEEE 802.11)
between vehicles. The Federal Communication Commission (FCC) has allocated 75 MHz in 5.9
GHz band for Dedicated Short Range Communication (DSRC). DSRC was conceived to provide
architecture for vehicles in Vehicular Network to communicate with each other and with
infrastructure. In DSRC, subsequently specialized as Wireless Access in Vehicular Environment
(WAVE), GPS-enabled vehicles that are equipped on-board units can communicate with each
other.
VANETs are special class of Mobile Ad hoc Networks (MANETs).The major characteristics as
compared to MANETs are following: components building the network are vehicles, dynamic
topology, geographically constrained topology, vehicle mobility, frequently disconnected
network and time-varying vehicle density [1]. VANETs could be playing an important role in
future of vehicle communications. VANETs provide a variety of interesting applications like
cooperative forward collisions, sharing emergency warning messages, weather and traffic data
among vehicles. The past few years have seen increasing technological advances in Vehicular Ad
hoc Network. There are some significant challenges in VANETs for broadcast protocol.
First, high mobility and dynamic topology of vehicles is one of the issues should be consider in
VANETs. In order to high speed of vehicles, topology of VANET is sometimes changing. Due to
the same reason the connection between nodes frequently could be changed rapidly and
unpredictably [2]. So there is also a high degree of change in the number and distribution of the

2
node in the network. Since vehicle mobility depends heavily on environment scenarios, the
movement direction sometimes is predictable.
Second, high network density related to traffic density in Vehicular Network is the other
challenge. Traffic density is high during peak hour in urban areas. Because of large number of
vehicles , blindly flooding the packets which each node re-broadcast messages to its entire
neighbor except the one it got the message from may lead to serious contention and collisions in
transmission among neighboring nodes [3]. This problem is mostly referred to as broadcast storm
problem [4].
Third, disconnection is another challenging issue in designing a data dissemination in Vehicular
ad hoc network. Apart from the studies conducted in dense networks, there is a growing need to
deal with disconnection in free flow conditions in Vehicular Ad hoc Networks. Where nodes are
distributed sparsely, most of the time path between source and destinations may not exist.
Traditional routing protocol can only transmit packet over end-to-end connected path so they
would fail in such conditions. Although it is very hard to find end-to-end connection for free flow
, the high mobility of vehicular networks introduces opportunities for mobile vehicles to connect
each other during the movements. Connectivity has been studied for information dissemination in
multihop networks [5].
2. Free Flow Scenarios
One of the most important limitations of V2V networks is end to end delay due to partitioning of
networks. It is observed that carry-and-forward is the new and key consideration for designing all
routing protocols in VANETs. With the consideration of multi-hop forwarding and carry and-
forward techniques, min-delay and delay-bounded routing protocols for VANETs are discussed.
In this section, a connectivity analysis model has been employed. In this section vehicles
categorized in two sections.
1) Informed vehicle:
This refers to the nodes which carries the information along a road.
2) Uninformed Vehicle:
Uninformed vehicles refers to the nodes which have not received information .whenever an
uninformed vehicles enters to the radio range of an informed one, it becomes informed.
A distributed vehicle traffic model is used with the following assumptions:
• Poisson Arrival: Traffic moves on the road follows a Poisson process with an average
rate equal to traffic flow rate( vehicles/time)
• Independent vehicle mobility: vehicle travels with a speed that follows a random
distribution between [Vmin,Vmax].
Packets forward across a platoon via a single hop or multihop route, the number of middle hops
depends on the number of vehicles in a platoon space of two adjacent vehicle named intra platoon
spacing which calculate the Probability Distribution Function(PDF) of Sintra as follow:
݂ܵ݅݊‫ܽݎݐ‬ ൌ ‫ݎ݌‬ሺ‫ݍ|ݍ‬ ൏ ܴሻ ൌ
λୣషλୗ୧୬୲୰ୟ
ଵି௘λோ
(1)

3
which R is the transmission range of equipped vehicles and λ is probability distribution of
distance between equipped vehicles. By the assumption of undisturbed vehicle traffic model,
vehicles passing a location (A) follows a Poisson process [6].
Assume n vehicles pass location A during (0, t].
P[X(t)<XIN(t)=n]=P[V*t<x]*P[v*(t-T)<X)n
(2)
Where T and Ti has same distribution and V and has same distribution . Based on the
assumption of poisson process .N(t) has Poisson distribution
(3)
Then we have
P[X(t)<X I N(t)=n]=
P[V0*x, Vi *(t-Ti)<x for each i=1,2,…,n] (4)
By the assumption of undistributed node traffic model, vehicle passing A follow Poisson process.
Under this situation that n vehicles pass A during (0, t], the times at which these vehicles pass
location A consider unordered random variables distributed both independently and uniformly in
the interval (0,t],
Since T1, T2,…, Tn and V0, V1,…,Vn are (5)
P[X(t)< x I N(t)=n] = P[V*t<x] * P[V*(t-T)<x]n
(6)
Where T has the same distribution as Ti and V has the same distribution as Vi. Based on the
assumption of Poisson process, N(t) has the Poisson distribution
(7)
Where λ is the traffic flow rate.
FX(t) (x)= (8)
Let V(t) be the message speed
(9)
E[V(t)], the average message speed is given in the above equation.( Figure 1)
Figure 1- Average message speed

4
3. Congested Flow Scenarios
Besides, the temporary network fragmentation problem and the broadcast storm problem are
further considered for designing routing protocols in VANETs. The temporary network
fragmentation problem caused by rapidly changeable topology influence on the performance of
data transmissions. The broadcast storm problem seriously affects the successful rate of message
delivery in VANETs [7].
In this section a congested flow network will be modeled. For a given time t, we define Vp(t) to
be the average message propagation speed during the interval[0, t] , p=X(t)/t. we define the long-
term average message propagation speed vp to be
Vp= lim୲՜∞ Vሺtሻ = lim୲՜∞ x/t (10)
So
Vp=
୉ሾଡ଼୲ሿା୉ሾ୶୤ሿ
୉ሾ୘ሿା୉ሾ୘୤ሿ
(11)
Where Xt (Xf) is the average distance traveled during forward phase and Tf is the time spent
during forward phase. Figure 2 indicate numerical result related to average speed. we compare
E[Xt] and E[Xf]. If the source and destination of a message are both within the same partition,
there is an end-to-end (E2E) connection.
Figure 2- Numerical result
4. Conclusions
Besides, the temporary network fragmentation problem and the broadcast storm problem are
further considered for designing routing protocols in VANETs. The temporary network
fragmentation problem caused by rapidly changeable topology influence on the performance of
data transmissions. The broadcast storm problem seriously affects the successful rate of message
delivery in VANETs. To understand the characteristics of VANETs in different network densities
we presented models for vehicle traffic where information is propagating in the direction of
vehicle traffic flow. Then we develop a collection of models to study information propagation
using V2V communications. These models can provide expected information propagation
speed/distance over time.

5
References
1.Blum, J., Eskandarian, A., & Hoffman, L.:Challenges of inter-vehicle ad hoc networks,.IEEE
Transactions on Intelligent TransportationvSystems, 5(4), 347–351. doi:10.1109/TITS.2004.838218 (2006)
2. Chen, Y. S., Lin, Y. W., & Lee, S. L.: A Mobicast Routing Protocol for Vehicular Ad HocNetworks.
Mobile Networks and Applications(MONET).New York: ACM/Springer. (2009)
3. Fukuhara, T., Warabino, T., Ohseki, T., Saito, K., Sugiyama, K., Nishida, T., & Eguchi, K. :Broadcast
Methods for Inter-vehicle Communications System. IEEE Wireless Communications and Networking
Conference, 4, 2252-2257. (2005)
4. Tseng, Y. C., Ni, S. Y., Chen, Y. S., & Sheu,J. P. :The Broadcast Storm Problem in a Mobile Ad Hoc
Network. [WINET].ACM Wireless Networks, 8(2), 153–167.doi:10.1023/A:1013763825347.(2002)
5. Naumov, V., Baumann, R., & Gross, T.:An Evaluation Of Inter-Vehicle Ad Hoc Networks Based On
Realistic Vehicular Traces. In ACM International Symposium on Mobile Ad Hoc Networking and
Computing, (pp. 108-119). . (2006).
6.Sara Najafzadeh, Norafida Ithnin, Ramin Karimi, “Analytical Analysis of sparse and Dense Vehicular Ad
hoc Networks”, International Conference on Computer Science and Information Technology, Bangalore,
India(2011)
7. Tonguz, O., Wisitpongphan, N., Bai, F., Mudalige, P., & Sadekar, V.: Broadcasting in VANET.Mobile
Networking for Vehicular Environments,(pp. 7-12). (2007)

Message Propagation in Vehicular Ad hoc Networks under free flow and congested flow Scenarios

More Related Content

What's hot (18)

Similar to Message Propagation in Vehicular Ad hoc Networks under free flow and congested flow Scenarios (20)

Recently uploaded (20)

Message Propagation in Vehicular Ad hoc Networks under free flow and congested flow Scenarios