Content uploaded by Sanjoy Das
Author content
All content in this area was uploaded by Sanjoy Das on Mar 11, 2016
Content may be subject to copyright.
Wireless Pers Commun (2013) 68:65–78
DOI 10.1007/s11277-011-0439-8
Performance Analysis of P-GEDIR Protocol for Vehicular
Ad Hoc Network in Urban Traffic Environments
Ram Shringar Raw · Sanjoy Das
Published online: 5 November 2011
© Springer Science+Business Media, LLC. 2011
Abstract A Vehicular Ad hoc Network (VANET) is a wireless ad hoc network that is
formed between vehicles on an on demand basis. In VANETs all the vehicles (nodes) are
used as routers and these routers are free to move randomly and organized themselves arbi-
trarily. A lot of research work around the world is being conducted to design an efficient
routing protocol for VANETs. In this paper, we propose a new routing method known as
Peripheral node based GEographic DIstance Routing (P-GEDIR), a position-based routing
protocol that takes advantage of GEographic DIstance Routing (GEDIR). It may not be pos-
sible to find node at the extreme end of the transmission range. Therefore, we have considered
an area around the extreme end of the transmission range. Further a mathematical model for
the protocol has been designed to determine expected number of successful hops, expected
distance to the next-hop node, and expected one-hop progress. The protocol has been sim-
ulated using MATLAB. In this work, results clearly show that using the peripheral node is
an advantage to maximize the performance of routing protocol in terms of average number
of successful hops and expected one-hop progress. The result of P-GEDIR is compared with
the existing GEDIR protocol.
Keywords MANET · VANET · Routing protocols · GEDIR · P-GEDIR ·
Urban environment
1 Introduction
According to World Health Organization (WHO), millions of people around the world die
every year because of vehicular traffic accidents and one fourth of all deaths caused by
injury. Also about 50 millions of people are injured in vehicular traffic accidents. Take the
metropolitan city Delhi in India for example, where plenty of vehicles like car, truck, buses,
R. S. Raw (
B
) · S. Das
School of Computer and Systems Sciences, Jawaharlal Nehru University, New Delhi 110067, India
e-mail: rsrao08@yahoo.in
S. Das
e-mail: sdas.jnu@gmail.com
123
66 R. S. Raw, S. Das
motorcycles etc. runs on the road at any given time. Delhi has huge population and with
respect to population, vehicles population in Delhi is large among all metropolitan cities in
India. On an average about 600 new vehicles are added in Delhi every day [1]. Department
of transportation annual reports says, thousands of people around the Delhi city die every
year because of the vehicular traffic accidents and many more are injured.
Road safety is a major factor of vehicular traffic management. The exceptional growth in
the number of vehicles in the city with limited road space, careless driving and violation of
traffic rules caused large number of traffic accidents. Increasing parking demand with limited
parking space and unfamiliar with travel related information is an obstruction to the smooth
flow of vehicular traffic, especially in crowded and major commercial areas.
Therefore, to reduce large number of vehicular traffic accidents, improve safety, manage
traffic control system, and provide important facilities to drivers and passengers with high
and reliable efficiency, computer networking researchers proposed a new wireless networking
concept called Vehicular Ad hoc Network (VANET) [2].
In this paper, we present the significant role of routing protocol; especially the position-
based routing protocol for VANET. Routing is the process of finding optimal path between
source and destination node and then sending message in a timely manner. Routes between
source and destination node may contain multiple hops, this condition is more complex than
the one-hop communication. Intermediate vehicles can be used as routers to determine the
traffic’s path along the way. Since the network topology is frequently changing, finding and
maintaining routes is very challenging task in VANETs. Traditional topology-based routing
protocols [3] are not suitable for VANETs. Position-based routing protocols such as GPSR,
GPCR, A-STAR, MFR, GEDIR, etc. are more suitable than other routing protocols. Routing
protocols in VANET can be classified into four significant categories as given in the Fig. 1.
In this work, we propose a new position-based routing protocol; Peripheral node based
GEographic DIstance Routing (P-GEDIR) which takes advantage of GEographic DIs-
tance Routing (GEDIR) protocol. P-GEDIR improves data delivery in various scenarios of
VANETs. Specially, P-GEDIR is designed to efficiently route the packet with less number
of hops in city or urban vehicular environments. This routing scheme uses the concepts
of peripheral node of the sender’s communication range to minimize the number of hops
between source and destination.
The rest of this paper is organized as follows. In Sect. 2, VANET is briefly described. In
Sect. 3, we describe the routing protocols and data dissemination issues. Sect. 4,presents
the related works. In Sect. 5, we introduce the design of P-GEDIR protocol method. Sect. 6
presents the mathematical analysis of the proposed protocol. Simulation results and perfor-
mance analysis are discussed in Sect. 7. Finally, we conclude this paper in Sect. 8.
Fig. 1 Classification of routing protocols
123
Urban Traffic Environments 67
2 Vehicular Ad Hoc Network
Vehicular Ad hoc Network (VANET0 is a budding and challenging subclass of Mobile Ad
Hoc Networks (MANETs) [4]. A VANET has emerged as one of the most recent research
areas for the last one decade. In VANET, vehicles can communicate to each other through
multiple paths using intermediate nodes to forward the packets from the source to destina-
tion. VANETs are designed to make available drivers with immediate information in two
ways: Vehicle-to-Vehicle (V2V) or inter-vehicle communication and Vehicle-to-Roadside
infrastructure (base station) (V2R) communication [5](showninFig.2).
Therefore, VANETs [6] will provide safer and well-organized road in future by commu-
nicating information in timely manner to drivers and concerned authorities. VANET is based
on short range wireless communication. IEEE 802.11p (modified version of IEEE 802.11a
standard protocol) [7,8] is a wireless communication protocol. This protocol is specially
designed for VANETs to support safety and non-safety applications. This standard provides
wireless devices that are able to communicate between highly mobile vehicles and fixed road
side infrastructure units. This mode of operation is known as Wireless Access in Vehicular
Environment (WAVE). It will operate in 5.9 GHz frequency band and provides an enhance-
ment to the physical layer (PHY). Medium Access Control (MAC) layer is based on IEEE
802.11 Distributed Coordination Function (DCF) for the Dedicated Short Range Communi-
cation (DSRC) standard and was adapted by ASTM and IEEE. DSRC has 75 MHz licensed
frequency band divided into 7 different channels with 10 MHz channel bandwidth each. It
supports line of sight distance with a range of 1 km and vehicle speed of up to 150 km/h.
VANET support GPS enable vehicles that are equipped with computing mechanism, short
range wireless interface and a GPS receiver. A GPS receiver is a device that capable to receive
the information sent by the satellites. GPS receiver uses this information to calculate its dis-
tance and finally compute its position in the geographical area on the earth in terms of
latitude, longitude and altitude. VANET have some important characteristics such as nodes
forming the networks are vehicles, restricted vehicle movements on the road, highly mobil-
ity of vehicles, rapid change in network topology, and time-varying vehicle density. These
characteristics make the VANET a special type of network. The network behavior is greatly
affected by these characteristics and many challenges have to be address while deploying the
vehicular networks to provide safety and comfort services for the passengers on the roads. In
highways, vehicles can move at high speeds and they can communicate with other vehicles
Fig. 2 VANET communication
scenarios. a V2V
communication. b V2B
communication
123
68 R. S. Raw, S. Das
within the communication range. But in city or urban areas vehicles are slow and there may
be radio obstacles because of buildings and trees. In VANETs, vehicle may join and leave the
network much more frequently than other networks. Because of the different traffic density,
sometimes it is very difficult to find an end-to-end connectivity when there is no vehicle
present that can forward the packet to the destination.
3 Routing and Data Dissemination Issues in VANETs
Since the topology frequently changes due to high mobility of nodes causes short commu-
nication connections lifetime especially with multi-hop paths. These characteristics degrade
the performance of routing protocols. The topology based routing needs to maintain a path
from source to destination, but due to rapid movements of nodes, the lifetime of path become
very short. In vehicular networks, on highways or urban areas traffic density is high during
day time and in rural areas or late night hour’s traffic density is less and create a sparsely con-
nected network [9]. As vehicular network support variety of applications, so that routing and
data dissemination techniques should support the characteristics and applications of vehicu-
lar networks. While in disseminating message there must be categorization of messages (i.e.,
safety, non safety, casual etc.) and according to priority message should be disseminated. The
development and deployment dissemination algorithms should consider the traffic pattern
(dense or sparse) over the network and type of applications customer want to access. For
example, the dissemination of safety messages should broadcasted in the network on priority
over non safety messages. In case, non safety message disseminated through unicast or mul-
ticast transmission. Broadcasting in densely populated network introduce broadcast storm
problem. Several mechanisms already proposed to mitigate the effect of broadcast storm
problem. A robust and efficient routing and dissemination mechanisms need to develop for
future deployment of VANETs. While developing such mechanisms, the topology structure,
traffic density, interfering objects, latency, etc. should be considered.
4 Related Work
Greedy routing scheme is a loop free and memoryless routing scheme. In the greedy position-
based routing scheme, a source node finds the position information of its direct neighbors
and selects that neighbor which is nearest to the destination node as the next-hop node.
AGEDIR[10] is a loop free position-based routing algorithm. In GEDIR, a source node
forwards packets to its neighbor node that is closest to the destination node. In the Fig. 3,
source node S has two neighbors A and B. When source node S wants to send a message to
destination node D, it uses the location information of D and for all its direct neighbors to
Fig. 3 GEDIR forwarding
method
123
Urban Traffic Environments 69
determine the neighbor B which is closest to D.NowthemessageisforwardedtoB and the
same procedure is repeated until the packet reached to D. From the Fig. 3 we can see that
path chosen by GEDIR routing methods is S → B → C → D.
5ProposedWork
5.1 Assumptions
The P-GEDIR protocol design is based on the following assumptions [11,12]:
• Peripheral nodes for forwarding packets
• Hello (beacon) control message for next-hop neighbors
• Nodes are equipped with GPS receiver
• Vehicles equipped with digital maps and sensors
• Communication between vehicles using wireless ad hoc network
• No other communication infrastructure
• Maximum forwarding distance is fixed
• Forwarding direction towards destination
5.2 Neighbor Node
A node has a set of one-hop nodes in its transmission range. These one-hop nodes are called
neighbor nodes. In dynamic mobile ad hoc network, source node and its neighbors are mov-
ing randomly and changing their positions frequently. Each neighbor node updates their
information like current location, current time, speed and direction by exchanging the Hello
message.
5.3 Peripheral Node
Generally, the neighboring nodes are found one-hop away within the transmission range of
the source node. The one-hop nodes are divided into interior nodes and peripheral nodes. A
peripheral node is defined as a border node, whose distance from the source node (central
node) is exactly R
o
, which is equal to the radius of the maximum transmission range R of the
source node. Therefore the peripheral node lies furthest away within the transmission range
of the source node (shown in Fig. 4).
5.4 Peripheral Node Based GEographic DIstance Routing Protocol
VANET can be improved for better and efficient routing decisions in greedy forwarding
method for heterogeneous unevenly random vehicular environment [13]. In this paper, we
propose an alternative novel routing protocol; Peripheral Node Based GEographic DIstance
Routing Protocol (P-GEDIR) protocol that improves the packet delivery in city vehicular
environment where vehicles are distributed unevenly. The P-GEDIR utilizes the peripheral
node to avoid sending packet to an interior node within the transmission range of source node.
P-GEDIR protocol selects the only peripheral node that is closest to the destination node as
the next-hop node for forwarding packet from source to destination, as shown in Fig. 5.
In Fig. 5, node A is a peripheral node of source node S, since node A is positioned at
maximum transmission range and has the greatest progress towards destination. Therefore
123
70 R. S. Raw, S. Das
Fig. 4 Peripheral node
architecture
Fig. 5 P-GEDIR packet
forwarding method
A is selected as the next-hop forwarding node. Node A when receives the message from S
uses the same method, to find the next-hop forwarding node with greatest progress towards
destination. In this case, node B is selected as a peripheral node of A for forwarding packets
to destination. Finally node B directly delivers the message to destination node D. The whole
P-GEDIR method is summarized through data flow diagram in Fig. 6.
6 Mathematical Analysis of the Proposed Protocol
In VANET, a node sends information to all its neighbors that are located within its transmis-
sion range. Because of the limited transmission range, the routes between nodes are usually
created through several hops in VANETs. Two nodes A and B in the network are direct neigh-
bors if the distance between them is at most R,whereR is the transmission range which is
equal for all nodes in the network. In highly dynamic networks such as vehicular networks,
the average number of hops and one-hop progress are considered the important parameters.
These key metric are used for performance comparison between different routing protocols.
6.1 Distribution of Nodes in the Shaded Area
In this paper, we study general properties of position-based routing in VANETs. We use
a circular region to place the neighboring nodes of vehicular nodes and simulate routing
performance for different node densities. For our analysis, the vehicular nodes are consid-
ered uniformly distributed over the entire two-dimensional area. All nodes have a maximum
transmission range R. This also indicates the radius of the circular region for packet trans-
mission. Assume A is the area of the transmission range (circular region), λ is the vehicle
123
Urban Traffic Environments 71
Fig. 6 Flow diagram of P-GEDIR protocol
Fig. 7 Shaded area having
peripheral nodes
density, and N (N = λπ R
2
) the number of nodes in the transmission range. For simplicity
of our analysis, a node is considered to be a possible forwarder if it is in the half circle of the
transmission range of source node towards the destination (the entire shaded area in Fig. 7).
Figure 7 shows that neighbor nodes (peripheral nodes) are placed in the given shaded area
nearest to the border or on the border of the transmission range of the source node.
123
72 R. S. Raw, S. Das
According to the practical situation, we use the transmission range R and geometrical
(shaded) area of the communication circle can be formulated as:
S
A
=
π R
2
2
−
πr
2
2
=
π
2
R
2
− r
2
(1)
We assume that availability of nodes in a given region follows Poisson distribution. If X is
the random variable representing the number of nodes in the shaded area then the probability
of n nodes present in shaded area is:
P
S
A
(X = n) =
(
λS
A
)
n
· e
−λS
A
n!
=
λπ(R
2
−r
2
)
2
n
· e
−
λπ(R
2
−r
2
)
2
n!
(2)
where λ is the node density. The probability of selecting k nodes out of n nodes is:
P
(
Y = k
)
=
n
k
p
k
q
n−k
=
n
k
(
p
)
k
(
1 − p
)
n−k
(3)
where p is probability of selecting a node and q = (1 − p) is the probability of not selecting
a node. Now probability of selecting exactly k nodes in the given shaded area is [13,14]:
P
(
k
)
=
∞
n=k
⎡
⎢
⎣
n
k
(
p
)
k
(
1 − p
)
n−k
.
λπ(R
2
−r
2
)
2
n
· e
−
λπ(R
2
−r
2
)
2
n!
⎤
⎥
⎦
=
pλπ(R
2
−r
2
)
2
k!
k
· e
−
pλπ( R
2
−r
2
)
2
(4)
Therefore the probability to select at least k nodes in the shaded area of the communication
range:
P(k) = 1 −
k−1
i=0
⎡
⎢
⎢
⎢
⎣
pλπ
R
2
−r
2
2
i!
i
· e
−
pλπ
(
R
2
−r
2
)
2
⎤
⎥
⎥
⎥
⎦
(5)
From the Eq. (5), we can easily obtain the probability P to select at least one node within
the shaded area with radius R.
P = 1 − P
(
X = 0
)
= 1 − e
−
pλπ( R
2
−r
2
)
2
(6)
Figure 8 shows the probability of finding at least k nodes in the area S
A
when λ are 0.0003
and 0.0005 nodes/km
2
. From the Fig. 8, it can be seen that the probability of finding one
or more vehicles within the transmission range is close to 1. Therefore, a vehicle should be
within the range of at least one other vehicle to maintain connectivity and support multihop
routing in the network.
123
Urban Traffic Environments 73
0 5 10 15 20 25 30 35 40 45 50
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Number of Nodes
Probability (k)
= 0.0003
= 0.0005
Fig. 8 Probability of at least k nodes in the shaded area
6.2 Expected Number of Successful Hops
Packet transmission fails if packet not arrives in the shaded area, which happens with proba-
bility q = 1 − p. We consider the topology effect on the route to the destination. If a packet
arrives in the shaded area, it will be forwarded towards the destination by using the neighbor
node closest to the destination. Let H is a random variable representing the number of hops
upto which link is available between source and destination. Therefore, the probability that
there are n links available between source and destination is given by:
P
(
H = n
)
=
(
1 − p
)
p
n
Then with Eq. (6), the expectation E(H) for the number of successful hops is given as
follows:
E
(
H
)
=
p
1 − p
=
1 − P
(
X = 0
)
P
(
X = 0
)
=
1 − e
−
pλπ( R
2
−r
2
)
2
e
−
pλπ( R
2
−r
2
)
2
= e
pλπ( R
2
−r
2
)
2
− 1(7)
6.3 Expected Distance Between Source and Next-Hop Node
Assume a source node S has n neighbors in the direction of destination node. Let A is the
farthest node (peripheral node) of the transmission range R of source node S (as shown in
Fig. 6). Let d
1
, d
2
, d
3
...,d
n
denotes the distances between source node and its neighbors
[15]. x is the distance between source node and its farthest node, i.e.
x = Max
n
i=1
d
i
Then we can calculate the expected value of distance x for the peripheral node to send the
packet to the destination as follows:
123
74 R. S. Raw, S. Das
Let F(x) and f (x) be the CDF and PDF of x. Then,
F(x) = P
[
d
1
≤ x, d
2
≤ x,...,d
n
≤ x
]
=
n
i=1
P
[
d
i
≤ x
]
=
x
R
n
Similarly,
f (x) =
d
dx
F(x)
=
d
dx
x
R
n
=
n
R
x
R
n−1
The expected value of x is,
E(x) =
R
r
x. f (x)dx
=
n
R
n
R
r
x · x
n−1
dx =
n
R
n
x
n+1
n + 1
R
r
=
n
R
n
R
n+1
(n + 1)
−
r
n+1
(n + 1)
E(x) =
n ·
R
n+1
− r
n+1
(n + 1) · R
n
(8)
6.4 Expected One-Hop Progress
The actual distribution of number of nodes located in a shaded area towards destination as
derived in Eq. (4). From Eqs. (4)and(8), we can obtain the expected one-hop progress EHP
depending on the transmission range R and node density λ.
EHP =
∞
k=1
pλπ(R
2
−r
2
)
2
k!
k
· e
−
pλπ( R
2
−r
2
)
2
.
k.
R
k+1
− r
k+1
(
k + 1
)
· R
k
= e
−
pλπ( R
2
−r
2
)
2
∞
k=1
pλπ(R
2
−r
2
)
2
k!
k
.
k.
R
k+1
− r
k+1
(
k + 1
)
· R
k
(9)
In Eq. (6), we obtained the probability P to select at least one node within the shaded area
with radius R. Therefore, the expected one-hop progress (EHP) for a node present in the
shaded area can be obtained dividing by (6).
EHP =
e
−
pλπ( R
2
−r
2
)
2
1 − e
−
pλπ( R
2
−r
2
)
2
.
∞
k=1
pλπ(R
2
−r
2
)
2
k!
k
.
k.
R
k+1
− r
k+1
(
k + 1
)
· R
k
=
1
e
pλπ( R
2
−r
2
)
2
− 1
.
∞
k=1
pλπ(R
2
−r
2
)
2
k!
k
.
k.
R
k+1
− r
k+1
(
k + 1
)
· R
k
(10)
123
Urban Traffic Environments 75
Table 1 Parameter setup
Parameter Value
Simulation area 2000 m × 2000 m
Transmission range 200 m
Number of nodes 0–200
7 Results and Performance Analysis
VANETs basically employ multi-hop communications, where message is forwarded from
source to destination through multiple paths using intermediate nodes. In city vehicular traf-
fic environment, there are many intersections with traffic signs. To communicate with other
vehicles, a packet is passed from one intersection to another intersection. In this section,
we have evaluated the performance of our proposed routing protocol for vehicular networks
where results obtained through simulation. To simulate an unbounded area, only nodes located
at a distance larger than the transmission range R away from any peripheral of the area are
considered for packet transmissions. We use GEDIR routing protocol for comparison with
P-GEDIR.
In this section, some results obtained through MATLAB simulator are presented. The
performance of P-GEDIR is computed analytically and numerically both. Based on the sim-
ulation parameters given below, we have simulated the protocol with a variable number of
nodes from 0 to 200 and node densities. We use a 2000 m × 2000 m square area and a
transmission range of 200 m for simulation. In the simulations, results have been computed
in terms of one-hop progress and average number of successful hops between source and
destination Table 1.
7.1 One-Hop Progress
Figure 9 shows the corresponding result for one-hop progress. From the Fig. 9 we can observe
that as the number of nodes increases, the one hop progress initially increase rapidly. After the
0 10 20 30 40 50 60
20
40
60
80
100
120
140
160
180
200
Nnmber of Nodes
One Hop Progress
r = 170
r = 175
Fig. 9 Expected one-hop progress
123
76 R. S. Raw, S. Das
20 40 60 80 100 120 140 160 180 200
0
0.5
1
1.5
2
2.5
3
3.5
4
Number of Nodes
Aver age Nnmber of Successful Hops
P-GEDIR
GEDIR
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
x 10
-4
0
5
10
15
20
25
Node Densit
y
Ave r age Number of Successful Hops
P-GEDIR
GEDIR
(a)
(b)
Fig. 10 Average number of successful hops
number of nodes reaches 10, the one-hop progress remains constant at about 196.1016 and
174.9050 m and then gradually reaches 196.1026 and 174.9056 m for two different values
of radius r, 170 and 175 m respectively. As the radius r is decreases, the number of nodes
in the shaded area is increases. Therefore, the expected one-hop progress attains quickly the
maximum transmission range R.
7.2 Average Number of Successful Hops
In this section, we have shown the performance comparison between GEDIR and P-GEDIR.
We considered the average number of successful hops is an essential performance measure
for VANET and the results for which are shown in Fig. 10. At first, we notice that the average
number of successful hops for both the protocols clearly increase as the number of nodes and
node density increases. But for P-GEDIR, number of successful hops is significantly lower
than GEDIR due to using the peripheral node for packet transmission only. This difference is
123
Urban Traffic Environments 77
clearly evident from the Fig. 10a, when the number of nodes is 200, the number of successful
hops for P-GEDIR is 1.2280 and for GEDIR it is 3.8105. Similarly, from the Fig. 10bwhen
node density is 0.0001, the number of successful hops for P-GEDIR is 2.9529 and for GEDIR
it is 22.1407.
8 Conclusion and Future Works
In this work, we have proposed a new position-based routing protocol that we call Peripheral
node Geographic Distance Routing (P-GEDIR). The main design goal of P-GEDIR method
is to select the appropriate peripheral node to route data packet in VANETs. P-GEDIR opti-
mizes the forwarding behavior based on the one-hop neighbor information received in Hello
packet exchange process. It reduced the forwarding delay and considers constant forwarding
distance between two neighboring nodes to achieve the high reliability. Simulation results
shows that P-GEDIR gives better performance than GEDIR in terms of average number of
successful hops and one-hop progress. As for future works, VANETs needs more research
which could lead to further improvements in vehicular ad hoc routing.
References
1. Rites Ltd (1994). Urban environmental engineering household surveys. Delhi.
2. Moustafa, H., & Zhang, Y. (2009). Vehicular networks: Techniques, standards, and applications.Boca
Raton: CRC Press India.
3. Li, F., & Wang, Y. (2007). Routing in vehicular ad hoc networks: A survey. IEEE Vehicular Technology
Magazine, 2(2), 12–22.
4. Mohapatra, P., & Krishnamurthy, S. V. (2009). Ad hoc networks: Technologies and protocols.
Berlin: Springer.
5. Raw, R. S., Das, S., & Agarwal, A. (2011). Thoughts on Vehicular Ad Hoc Networks (VANETs) in
the real world traffic scenarios. International Journal of Computer Science and Management System
(IJCSMS), 3(1), 19–26.
6. Manvi, S. S., & Kakkasageri, M. S. (2008). Issue in mobile ad hoc networks for vehicular commu-
nication. IETE Technical Review, 25(2), 59–72.
7. Task Group p,. IEEE 802.11 p wireless access for vehicular environment draft standard. http://grouper.
ieee.org/groups/802/11/. Accessed 25 June 2010.
8. Chen, Q., Jiang, D., Taliwal, V., & Delgrossi, L. (2006). IEEE 802.11 based vehicular communication
simulation design for NS-2. Los Angeles, California, USA: VANET’ 06.
9. Das, S., & Raw, R. S. (2010). Vehicular ad-hoc networks deployments: Challenges and issues. National
conference & workshop on High Performance Computing & Applications (HPCA), February 08–10,
2010, Banaras Hindu University, India.
10. Stojmenovic, I., Ruhil, A. P., & Lobiyal, D. K. (2006). Voronoi diagram and convex hull based
Geocasting and routing in wireless networks. Wireless Communications and Mobile Computing, 6(2),
247–258.
11. Raw, R. S., & Lobiyal, D. K. (2010). B-MFR routing protocol for vehicular ad hoc networks (pp. 420–
423). Manila, Philippines: IEEE ICNIT 2010.
12. Zhang, M., & Wolff, R. S. (2010). A border node based routing protocol for partially connected
vehicular ad hoc networks. Journal of Communications, 5(2), 130–143.
13. Raw, R. S., & Lobiyal, D. K. (2011). E-DIR: A directional routing protocol for VANETs in a city
traffic environment. International Journal of Information and Communication Technology (IJICT), 3(3),
242–257.
14. Heissenbuttel, M., & Braun, T. (2003). A novel position-based and beacon-less routing algorithm for
mobile ad-hoc networks (pp. 197–210). Bern, Switzerland: ASWN.
15. Yi, C., Chuang Y., Yeh, H., Tseng, Y., & Liu, P. (2010). Streetcast: An urban broadcast protocol for
vehicular ad-hoc networks. In 71st IEEE vehicular technology conference (pp. 1–5).
123
78 R. S. Raw, S. Das
Author Biographies
Ram Shringar Raw received his B.E. (Computer Science and Engi-
neering) from G. B. Pant Engineering College, Pauri-Garhwal, UK,
India and M. Tech (Information Technology) from Sam Higginbot-
tom Institute of Agriculture, Technology and Sciences, Allahabad
(UP), India in 2000 and 2005, respectively. He is pursuing Ph.D.
(Computer Science) from School of Computer and Systems Sciences,
Jawaharlal Nehru University, New Delhi, India. He is currently work-
ing as Assistant Professor at Computer Science and Engineering
Department, G. B. Pant Engineering College, Uttarakhand Technical
University, since 2001. His current research interest includes Mobile
Ad hoc Networks and Vehicular Ad hoc Networks. Mr. Raw has
published papers in International Journals and Conferences including
IEEE, Springer, and Inder Science.
Sanjoy Das received his B.E. (Computer Science and Engineering)
from G. B. Pant Engineering College, Pauri-Garhwal, UK, India and
M. Tech (Computer Sc. & Engg.) from Sam Higginbottom Institute
of Agriculture, Technology and Sciences, Allahabad (UP), India in
2001 and 2006, respectively. He is full time research scholar at School
of Computer and Systems Sciences, Jawaharlal Nehru University,
New Delhi, India. He has worked as Assistant Professor at Computer
Science and Engineering Department, G. B. Pant Engineering College,
Uttarakhand Technical University, from 2001–2008. His current
research interest includes Mobile Ad hoc Networks and Vehicular Ad
hoc Networks.
123
