International Journal of Engineering Trends and Technology (IJETT) - Volume4Issue5- May 2013
Cluster based destination driven on demand
Multicast routing protocol
Gourav Karwar
Department of Electronics and Communication, Lovely professional university Phagwara Punjab, India
Phagwara Punjab, India
Abstract: Multicasting is the ability of a communication
network to accept a single message from an application
and to deliver copies of the message to multiple
recipients at different locations. Multicast routing in
wireless networking is the newest technology that works
with network groups. Multicast routing plays an
important role in point-to-point or multipoint-tomultipoint communications. There are many multicast
routing protocol exist in wired network as well as
mobile ad-hoc network. In this paper we propose a
cluster based Destination driven on demand multicast
routing protocol that work with the mobile ad-hoc
networks. Simulation/animation results shows that
clustering on this protocol enhances its performance in
terms of network communication cost and looping
effect will be permanently reduced.
Keywords: MANET, multicasting, multicast routing
protocols.
Introduction:
Mobile ad hoc network is a wireless multi hop network
which can be conveniently deployed without any fixed
infrastructure in the hostile environment .MANET is rooted
in DARPA packet radio network. It find application in
many fields such as military deployment , electronic class
room etc .In the ad hoc network, communication between
two nodes which are outside the range of communication
ISSN: 2231-5381
with each other are realized by packet relay of intermediate
node . MANET is a collection of mobile routers which are
interacted with each other via wireless link or over the
radio channel and are free to move about arbitrarily. i.e due
to limited radio propagation range of wireless devices
communication between the couple of mobile nodes out of
transmission range of each other require a relay of
intermediate nodes between them . In other words, each
node operates as a router and collaborates on delivery
packet.
Ad hoc wireless networks inherit the traditional problems
of wireless communications and wireless networking:
• The wireless medium has neither absolute, nor readily
observable boundaries outside of which stations are known
to be unable to receive network frames;
• The channel is unprotected from outside signals.
• The wireless medium is significantly less reliable than
wired media
• The channel has time-varying and asymmetric
propagation properties
• Hidden-terminal and exposed-terminal phenomena may
occur.
There are three most important issue in the mobile ad hoc
networks are first and for most issue is its radio interface ,
second one is mobility of nodes and the third and important
issue is the limited equipment resources including storage
capacity and power.
Wireless application such as military battlefields,
emergency searches, rescues where sending of information
is mandatory require rapid deployable and quick
reconfigurable protocols. So , just because of these
situations there is a need of multicast routing protocols .
Multicasting [6] is the ability of a communication network
http://www.ijettjournal.org
Page 1506
International Journal of Engineering Trends and Technology (IJETT) - Volume4Issue5- May 2013
to accept a single message from an application and to
deliver copies of the message to multiple recipients at
different locations. One of the major challenges is to
reduce the amount of network resources employed by
multicasting.
Any multicast network should have three main
properties:
1) High probability of delivery of information.
2) Low delay between source and receiver which mean
network should be real time.
3) Hiding information from intermediate nodes which data
should be safe and secure from eavesdroppers.
These three properties are obtained with the help of optical
signal only electrical signal is unable to provide these
properties.
There are two types of multicast routing protocol in
wireless ad hoc network. These are Tree based multicast
routing protocol and Mesh based multicast routing
protocols.
Tree based routing protocols
The tree based routing protocol are highly unstable in the
mobile ad hoc network as it need frequent re-configuration
in the dynamic network ( mobility of nodes ).There is only
a single path between multicast source and receiver. There
mesh based protocol .example of hybrid protocol are (AM
Route) ad hoc multicast routing protocol
The rest of the paper is organized as follows.
Section1 gives the overview of clustering technique
.Section 2 describes the operation of D-ODMRP.and
cluster based D-ODMRP. Simulation results of cluster
based Destination driven on demand multicast routing
protocol are reported in Section 3. At the last in section 4
we will give concluding remarks of this paper.
Section 1
Clustering
A clustering is nothing but a division of network into an
interconnected sub networks. These sub networks are
known as clusters. Each cluster has a cluster head. The
cluster head is responsible for the routing inside the cluster
or outside the cluster. Each cluster head is act as a
temporary based station within its cluster or zone. It also
communicates with the other cluster in the network. The
ordinary nodes have a direct access only to cluster head
and gateway. A gateway is nothing but a single node
responsible for communicate within two or more cluster
heads. The Cluster based routing provides an answer to
address nodes heterogeneity, and to limit the amount of
routing information that propagates inside the network. The
grouping of network nodes into a number of overlapping
clusters is the main idea behind clustering. A hierarchical
routing is possible by clustering in which paths are
recorded between clusters instead of between nodes. A
clustering increases the route life time, thus decreases
routing control overhead.
are various tree based protocols like multicast extension of
ad hoc on demand distance vector (MAODV) and Adaptive
demand driven multicast routing protocol (ADMRP).
Mesh based routing protocols
K-means
K-means clustering algorithm is used to divide the network
into k number of clusters. K-means clustering uses a two
phase iterative algorithm to minimize the sum of point to
centroid distances, summed over all k clusters.
Section2: protocol description
Mesh based protocol is highly stable in mobile ad hoc
network .There may exist more than one path between
Destination driven on demand multicast routing
source and receiver. There are many mesh based multicast
protocol (D-ODMRP)[1]:
routing protocols exist in MANET. Examples of mesh
based protocols are on demand multicast routing protocol
(ODMRP), Destination driven on demand multicast routing
protocol (D-ODMRP) and core assisted mesh protocol
It is the extension in on demand multicast routing protocol
(ODMRP). The design objective of D-ODMRP is just to
improve the multicast forwarding efficiency. To achieve
this goal the path to reach a multicast destination (source to
(CAMP).
receiver) is biased towards those path passing through
There is one more types of routing protocol hybrid protocol
another multicast destination. If number of paths are
which is a combination of both tree based protocol and
available it select path which is of least extra cost. In other
ISSN: 2231-5381
http://www.ijettjournal.org
Page 1507
International Journal of Engineering Trends and Technology (IJETT) - Volume4Issue5- May 2013
words, D-ODMRP is used to reduce the number of nodes
1.
Join query/route request
to be added in the forwarding group. D-ODMRP introduces
2.
Join reply/route reply
the destination driven features in to the existing on demand
3.
Data forwarding
process of multicast forwarding structure. D-ODMRP
When a source has to send data but it has no route then
modifies the regular route request flooding process for
it send join query to establish a route between source
building a multicast tree by intensely add a deferring time
and destination. D-ODMRP also work on the
at each intermediate node to forward the every route
forwarding group concept which means only a subset
request based on how far this intermediate node is away
of nodes will forward data from source to destination.
from the last multicast group member visited by the
Only a multicast member will respond to the query and
received route request.
if the join query is the simple request then all nodes
will respond to them. Finally when the query reaches a
receiver it creates a join_ reply and broadcast it to the
neighbors. Node receiving the join -reply checks
whether the next node ID in the join-reply matches its
own. If yes, it is a part of the forwarding group then set
its FG-flag and broadcast its join-reply built upon the
matched entries.
Now the join reply is propagated by each forwarding group
member until it reaches the source via shortest path. Routes
from source to receiver build a multicast mesh of nodes is
called forwarding group.
Cluster based destination driven on demand multicast
routing protocol (CH-D-ODMRP):
Figure1. DODMRP
Basically we divide the network into number of
The larger the distance is more will be the deferring time
.so, on considering this deferring time the route request
introduce the least extra cost are travel faster than the
others which have larger deferring time . As a result cost
overlapping clusters. Finding cluster head is the main issue
in the mobile ad hoc network. For that purpose there are
many cluster head election algorithm are there. These are
listed as below:
effective forwarding structure can be created. No extra
overhead is created as compare with the existing on
1.
Weight based cluster head algorithm.
demand routing protocol. D-ODMRP is the advancement
2.
Highest degree cluster head algorithm.
in the existing on-demand multicast routing protocol [3]
3.
Lowest identifiers cluster head algorithm.
and greatly improves the multicast forwarding efficiency.
4.
Distributed clustering algorithm.
D-ODMRP is consists of three phase. These three phases
Instead of these there is one new technique for
repeat themselves periodically as the multicast source
dividing the network into number of overlapping
keeps sending multicast packet into the network.
clusters and then election of cluster head on the
basis of shortest distance from the centroid point.
ISSN: 2231-5381
http://www.ijettjournal.org
Page 1508
International Journal of Engineering Trends and Technology (IJETT) - Volume4Issue5- May 2013
The technique is known as k-mean clustering
algorithm. It is highly used for obtaining clusters
from the network. It obtained cluster on the
density basis.
Algorithm specified cluster based destination
driven on demand multicast routing protocol:
1.
We first create a network in which number of
nodes is placed at random position.
2.
Applying
k-means clustering algorithm and
finding cluster head on the basis of shortest
distance.
3.
Applying D-ODMRP on to cluster heads in the
network.
Finding the parameters that will show proposed
Figure3.CH-D-ODMRP
implementation is better in terms of network
Section 3:
communication cost.
Performance evaluation:
In this section we conducted simulation and animation to
evaluate the performance of cluster based destination
driven on demand multicast routing protocol using the
MATLAB (R2012a).
A. Simulation environment:
In our simulation’s 25 nodes are there in the
150x150 m2 square area. All nodes are movable at
run time. All nodes places are random in nature.
These 25 nodes are group into 5 numbers of
clusters. K-means clustering algorithm is used to
group the total network into 5 numbers of clusters.
The selection of cluster head is on density basis.
Figure2. Network with cluster head selected
We have used the following metrics:
1.
Network communication cost: The network
communication cost is on the basis of how
much distance it cover going from source to
multicast destination.
2.
Load: cluster load is new parameter in the
clustered networks. It is defined as a how
ISSN: 2231-5381
http://www.ijettjournal.org
Page 1509
International Journal of Engineering Trends and Technology (IJETT) - Volume4Issue5- May 2013
much load a cluster attained every time a
Section4:
network is simulated.
Conclusion: In this paper we have designed a cluster based
B. Simulation results:
We simulate the network into 10 times. After
taking simulation we have obtained a different
graphs related to the load obtained at each cluster.
destination driven on demand multicast routing protocol
(CH-D-ODMRP) for mobile ad-hoc network. CH-DODMRP is advancement in the existing D-ODMRP. On
introducing a clustering algorithm into existing D-ODMRP
The graph is shown as below:
a path to reach a multicast destination is greatly biased
towards those paths passing through another multicast
destination.
Simulation
implementation
is
results
shows
better in terms
that
of
our
network
communication cost as well as our clustering algorithm is
sufficient enough for the selection of cluster head in the
mobile ad hoc network.
REFERENCES
1. Ke Tian, Baoxian Zhang, Hussein Mouftah, Zhuang Zhax and Jian
Ma(2009) “Destination-Driven On-Demand Multicast
Routing
Protocol
Ad
for
Wireless
Hoc
Networks” IEEE ICC 2009.
2. Anees Shaikh and Kang G. Shin () “Destination-Driven Routing for
Low-Cost Multicast” Real-Time Computing Laboratory Department of
Figure4. Load obtained at cluster1
The following figure number 7 shows the Total
cost obtained at after 10 numbers of simulations:
Electrical Engineering and Computer Science The University of Michigan
Ann Arbor.
3. Sung-Ju
Lee, Mario
Gerla, and
DemandMulticast Routing Protocol”
AdaptiveMobility Laboratory
Computer
Ching-Chuan
Chiang “On-
Wireless
Science
Department
University of California Los Angeles.
4. Mohammed R. BAKER u, M. Ali AKCAYOL(2011) “A Survey
of
Multicast Routing Protocols in Ad-Hoc Networks ” Gazi University,
Faculty of Engineering, Department of Computer Engineering, 06500
Ankara, Turkey.
5. Book: WDM optical network concepts, design and algorithm c.siva ram
murthy and Mohan gurusamy.
6.C.-K. Toh, “Ad Hoc Mobile Wireless Networks:
Protocols and Systems”, Prentice Hall, 2002.
Figure5.Total cost obtained after simulation
ISSN: 2231-5381
http://www.ijettjournal.org
Page 1510
International Journal of Engineering Trends and Technology (IJETT) - Volume4Issue5- May 2013
7. Waseem Shahzad, Farrukh Aslam Khan, and Abdul Basit Siddiqui,’’
Weighted Clustering using Comprehensive Learning Particle Swarm
Optimization for Mobile Ad Hoc Networks’’ International Journal of
Future Generation Communication and Networking Vol. 3, No. 1, March
, 2010
ISSN: 2231-5381
http://www.ijettjournal.org
Page 1511