Source routing in Mobile Ad hoc NETworks (MANETs)

RESEARCH PAPER
International Journal of Recent Trends in Engineering, Issue. 1, Vol. 1, May 2009
105
Source routing in Mobile Ad hoc NETworks
(MANETs)
Narendra Singh Yadav 1
and R.P.Yadav3
Department of Electronics & Communication Engineering,
Malaviya National Institute of Technology,
Jaipur, India
Email: {narensinghyadav@yahoo.com, rp_yadav@yahoo.com}
Abstract— Mobile ad hoc network is a reconfigurable
network of mobile nodes connected by multi-hop wireless
links and capable of operating without any fixed
infrastructure support. In order to facilitate communication
within such self-creating, self-organizing and self-
administrating network, a dynamic routing protocol is
needed. The primary goal of such an ad hoc network routing
protocol is to discover and establish a correct and efficient
route between a pair of nodes so that messages may be
delivered in a timely manner. Route construction should be
done with a minimum of overhead and bandwidth
consumption. This paper examines two routing protocols,
both on-demand source routing, for mobile ad hoc
networks– the Dynamic Source Routing (DSR), an flat
architecture based and the Cluster Based Routing Protocol
(CBRP), a cluster architecture based and evaluates both
routing protocols in terms of packet delivery fraction,
normalized routing load, average end to end delay,
throughput by varying number of nodes per sq. km, traffic
sources and mobility. Simulation results show that in high
mobility (pause time 0s) scenarios, CBRP outperforms DSR.
CBRP scales well with increasing number of nodes.
Index Terms—MANET, CBRP, DSR, routing, performance
I. INTRODUCTION
Mobile ad hoc networks are formed by autonomous
system of mobile nodes that utilize multi-hop radio
relaying and connected by wireless links without any
preexisting communication infrastructure or centralized
administration. Such networks are also known as
infrastructureless or multi-hop wireless networks.
Communication is directly between nodes or through
intermediate nodes acting as routers. Such network
requires each mobile host to be more intelligent so that it
can perform both functions of transmitting and receiving
of data as a host and of forwarding packets for other
mobile host as a router. The advantages of such a network
are rapid deployment, robustness, flexibility and inherent
support for mobility. Ad hoc networks, due to their quick
and economically less demanding deployment, find
applications in military operations, collaborative and
distributed computing, emergency operations, wireless
mesh networks, wireless sensor networks and hybrid
networks.
Due to dynamic topology, lack of fixed infrastructure
and frequent link failure the traditional routing for wired
networks cannot be directly applied to mobile ad hoc
networks because wired routing methods assume the
network to be stable and routing overhead to be almost
negligible. They either lack the ability to adapt to the
dynamic topology change of mobile ad hoc networks or
may cause large overhead that degrades network
performance. Therefore, routing is the most studied
problem in mobile ad hoc networks and a number of
routing protocols have been proposed [1-13], derived
from either distance-vector [14] or link-state [15] based
on classical routing algorithms.
Routing protocols for Mobile ad hoc networks can be
classified into two main categories: Proactive or table
driven routing protocols and Reactive or on-demand
routing protocols. In proactive protocols, every node
maintains the network topology information in the form
of routing tables by periodically exchanging routing
information. They include the Destination Sequenced
Distance Vector (DSDV) [2], the Wireless Routing
Protocol (WRP) [3], Source-Tree Adaptive Routing
(STAR) [5] and Cluster-head Gateway Switch Routing
protocol (CGSR) [4]. On the other hand, reactive
protocols obtain routes only on demand, which include
the Dynamic Source Routing (DSR) protocol [6], the Ad
hoc On-demand Distance Vector (AODV) protocol [7],
the Temporally Ordered Routing Algorithm (TORA) [8],
and the Associativity Based Routing (ABR) protocol
[10].
On-Demand routing protocols can be classified into
two categories: source routing and hop-by-hop routing. In
Source routed on-demand protocols each data packets
carry the complete path from source to destination.
Therefore, each intermediate node forwards these packets
according to the information in the header of each packet.
The major drawback with source routing protocols is that
in large networks they do not perform well. This is due to
two main reasons; firstly as the number of intermediate
nodes in each route grows, then so does the probability of
route failure. Secondly, as the number of intermediate
nodes in each route grows, then the amount of overhead
carried in each header of each data packet will grow as
well.
In hop-by-hop routing each data packet only carries
the destination address and the next hop address.
Therefore, each intermediate node in the path to the
destination uses its routing table to forward each data
packet towards the destination. The advantage of this
strategy is that routes are adaptable to the dynamically
changing environment of MANETs, since each node can
© 2009 ACADEMY PUBLISHER

RESEARCH PAPER
106
update its routing table when they receiver fresher
topology information and hence forward the data packets
over fresher and better routes. Using fresher routes also
means that fewer route recalculations are required during
data transmission. The disadvantage of this strategy is
that each intermediate node must store and maintain
routing information for each active route and each node
may require being aware of their surrounding neighbors
through the use of beaconing messages.
A crucial issue for mobile ad hoc networks is the
handling of a large number of nodes. As more nodes join
a mobile ad hoc network, contention is more likely and
the open nature of a mobile ad hoc network makes it
important that a network continues to operate even if
there are more nodes involved Several researchers have
done the qualitative and quantitative analysis of Ad Hoc
Routing Protocols in terms of different performance
metrics under different scenarios and is summarized in
brief in the following table 1.
TABLE 1
RELATED WORKS
Author [ ]
Parameters
Performance
parameter
Protocol Area No.
of
Node
Connections Packet
rate
J.Broch
[16]
DSDV,
DSR,
TORA,
AODV
1500 m
x 300 m
50 10-30 1,4,8 per
sec
Pause time
S. Das
[17,18 ]
DSR,
AODV
1500 m
x 300 m
50 10,20,30,
40
4/ sec
3/sec
Pause time
2200 m
x 600m
100 10, 20,
40
4/sec
2/sec
Pause time,
Network load
A.
Boukerche
[19 - 21]
AODV,
CBRP,
DSR
1500 m
x 300 m
50 10,20,30,
40
4/ sec
3/sec
Pause time
2200 m
x 600m
100 10, 20, 30
40
4/sec
2/sec
Pause time
A.
Boukerche
[22]
AODV,
CBRP,
DSR,
PAODV,
DSDV
1500 m
x 300 m
50 10,20,30,
40
4/ sec
3/sec
Pause time
2200 m
x 600m
100 10, 20, 30
40
4/sec
2/sec
Pause time
Similar simulations were also carried out in [23-26]
and all of them have used pause time as a performance
parameter. Most of the above work is limited on
performing simulations for ad hoc networks with a
limited number of nodes (either 50 or 100 nodes)
deployed in fixed geographical area maintaining node
density constant. But none of these works consider the
scalability issue and carry out the simulations with
number of nodes as a performance parameter. Nowadays,
the ad hoc network technology becomes more and more
popular and, as a result, large-scale ad hoc networks may
be deployed in battlefields, regions of disaster and large
towns. In fact, this was the motivation of this paper: to
make observations about the behavior of these protocols
in large-scale environment, to show which protocol scales
well in large scale mobile ad hoc network and to decide
which topology either flat or cluster is best suited in low
and high mobility.
The rest of the paper is organized as follows: Section II
provides an overview and general comparison of the
routing protocols used in the study. The simulation
environment and performance metrics are described in
Section III and then the results are presented in Section
IV. Finally Section V concludes the paper.
II. OVERVIEW OF DSR AND CBRP
As each protocol has its own merits and demerits, none
of them can be claimed as absolutely better than others.
Two mobile ad hoc routing protocols – the Dynamic
Source Routing (DSR), the flat architecture based On-
Demand source routing protocol and the Cluster Based
Routing Protocol (CBRP), the cluster architecture based
On-Demand source routing protocol are selected for
study. The reasons behind comparing of CBRP with DSR
are as follows:
Among flat architecture AODV and DSR protocols are
very much popular. AODV makes an assumption that
there are only bi-directional links (and ignores any links
that are not bi-directional) and that they all incur the same
cost to use. The assumption can decrease the overall
performance of the network if a large number of the links
do not meet the assumption. Also, whenever an
established route is broken the source must be notified
and this way the source gets interrupted whereas DSR
protocol can successfully discover and forward packets
over paths that contain unidirectional links. The DSR
protocol operates entirely on demand. It does not use any
periodic routing advertisement, link status sensing, or
neighbor detection packets; nor does it rely on these
functions from any underlying protocols in the network.
In hierarchical approach Cluster Based Routing
Protocol is suggested in [28] which seem to give answer
to the scalability issue and further DSR and CBRP both
use source routing. The key advantage of a source routing
design is that intermediate hops do not need to maintain
fresh routing information in order to route packets they
receive, since the packets themselves already contain all
the routing decisions. This advantage eliminates the need
for the periodic route advertisement and neighbor
detection packets present in other protocols. So in this
study performance of two routing protocols DSR and
CBRP is analyzed and compared. .
A. Dynamic Source Routing protocol (DSR)
The Dynamic Source Routing Protocol [6,27] is an on-
demand routing protocol designed to restrict the
bandwidth consumed by control packets by eliminating
the periodic table-update messages required in the table-
driven approach. It is beacon-less and hence does not
require periodic hello packet transmissions, which are
used by a node to inform its neighbors of its presence.
DSR is in a different class than most other routing
protocols for multi-hop ad hoc networks in that it uses
source routes supplied by a packet’s originator to
determine the packet’s path through the network, rather
than using independent hop-by-hop routing decisions
made by each node that receives the packet. In a source
routing design, each packet to be routed through the
network carries in its header the complete, ordered list of
nodes through which the packet must pass.

RESEARCH PAPER
107
The key distinguishing feature of DSR is the use of
source routing. That is, the sender knows the complete
hop-by-hop route to the destination. These routes are
stored in a route cache. The data packets carry the source
route in the packet header. The basic operations of the
DSR protocol are divided into two mechanisms: Route
Discovery and Route Maintenance.
Route Discovery is the mechanism by which a node
which has a data packet to send to a destination obtains a
source route to the desired destination. The route
discovery process is carried out by flooding the route
request throughout the network in controlled manner as
follows.
Node having a data packet to send to some desired
destination it
Checks the route cache
If (there exists a fresh enough path for
desired destination)
{Use it}
Otherwise
{Broadcast a route request (RREQ) packet}
Intermediate or destination node receiving the RREQ
checks
If (Itself is a destination)
{Send a route reply (RREP)}
Otherwise
{
If (there exists a fresh enough path for
desired destination)
{Send RREP}
Otherwise
{Broadcast a route request (RREQ) packet) }
}
The RREQ builds up the path traversed across the
network. The RREP routes back to the source by
traversing this path backward. The route carried back by
the RREP packet is cached at the source for future use.
Route Maintenance is the mechanism by which a
sending node is able to detect if the network topology has
changed such that it can no longer use its route to the
desired destination. For example, a route becomes broken
if two nodes listed as neighbors on the route have moved
out of range of each other. If any link on a source route is
broken, the source node is notified using a route error
(RERR) packet. The source removes any route using this
link from its cache. A new route discovery process must
be initiated by the source if this route is still needed. DSR
makes very aggressive use of source routing and route
caching. No special mechanism to detect routing loops is
needed. Also, any forwarding node caches the source
route in a packet it forwards for possible future use.
B. Cluster Based Routing Protocol (CBRP)
In CBRP the nodes of a wireless network are divided
into clusters. The diameter of a cluster is only two hops
and clusters can be disjoint or overlapping. Each cluster
elects one node as the clusterhead that is unique to each
cluster, for maintaining information about its members
and is responsible for the routing process. Flooding traffic
during route discovery gets minimized and route
discovery process speeds up due to grouping of nodes
into clusters.
The head of a cluster knows the addresses of its
members and each member has a bi-directional link to it.
Clusterheads communicate with each other through
gateway nodes. A gateway is a node that has two or more
clusterheads as its neighbors when the clusters are
overlapping or at least one clusterhead and another
gateway node when the clusters are disjoint.
The CBRP is characterized by its fully distributed
operation involving less flooding traffic during the
dynamic route discovery process with additional local
repair and route shortening mechanism that explicitly
utilize uni-directional links which would otherwise be
unutilized.
The operation of CBRP can be divided in three phases:
Cluster formation, routing process and Cluster
maintenance
(a) Cluster Formation: At any time, a node is in one of
the three states: a cluster member (CM), a cluster
head (CH), or undecided (UNDECIDED) (meaning
still searching for its host cluster). Every node
broadcast a hello message to its neighbors
periodically. When a node comes up, it enters the
undecided state and broadcasts a Hello message.
When a cluster-head gets this hello message it
responds with a triggered hello message immediately.
When the undecided node gets this message it sets its
state to member. If the undecided node times out, then
it makes itself the cluster-head if it has bi-directional
link to some neighbor otherwise it remains in
undecided state and repeats the procedure again.
(b) Routing process: The routing process works in two
steps. First, it discovers a route from a source node S
to a destination node D, afterwards it routes the
packets. When a source has to send data to
destination, it floods route request packets (but only
to the neighboring cluster-heads). On receiving the
request a clusterhead checks to see if the destination
is in its cluster. If yes, then it sends the request
directly to the destination else it sends it to all its
adjacent cluster-heads. The cluster-heads address is
recorded in the packet so a cluster-head discards a
request packet that it has already seen. When the
destination receives the request packet, it replies back
with the route that had been recorded in the request
packet. If the source does not receive a reply within a
time period, it backs off exponentially before trying to
send route request again. In CBRP, routing is done
using source routing. It also uses route shortening that
is on receiving a source route packet, the node tries to
find the farthest node in the route that is its neighbor
and sends the packet to that node thus reducing the
route. While forwarding the packet if a node detects a
broken link it sends back an error message to the
source and then uses local repair mechanism. In local
repair mechanism, when a node finds the next hop is
unreachable, it checks to see if the next hop can be
reached through any of its neighbor or if hop after

RESEARCH PAPER
108
next hop can be reached through any other neighbor.
If any of the two works, the packet can be sent out
over the repaired path.
(c) Cluster maintenance: Clusterheads are elected by the
individual clusters. It is expected that the clusterheads
will not change frequently. For avoiding frequent
change of clusterheads, a non-clusterhead never
challenges the status of the existing clusterhead even
though it satisfies the preliminary requirements of
becoming a clusterhead and the clusterhead change
takes place only when two clusterheads move next to
each other in which case one of them leaves the role
of clusterhead depending on certain rules for node
removal, node addition and cluster head contention.
C. Comparison of DSR and CBRP
The routing protocols used in this study are compared
with respect to different routing parameters are compared
below in table 2
TABLE 2
COMPARISON OF DSR AND CBRP ROUTING PROTOCOLS
Parameter DSR CBRP
Routing structure Flat Cluster
Hello messages No Yes
Multiple routes Yes No
Critical nodes No Yes
Route maintained Route Cache Route table at
cluster-Head
Routing metric Shortest path or
next available in
route cache
First available
route
Time complexity (route
discovery)
O (2D) O (2D)
Time complexity (route
maintenance)
O (2D) O (2d)
Communication complexity
(route discovery)
O (2N) O (2C)
Communication complexity
(route maintenance)
O (2N) O (2n)
Advantages Multiple routes,
beacon less
cluster heads
exchange
routing
information
Disadvantages Scalability
problem, large
delays
Cluster
maintenance
Abbreviations:
D = Diameter of the network
N = Number of nodes in the Network
d = Diameter of affected area
C = Number of Clusters
n = Number of affected nodes
III. SIMULATION PARAMETERS AND PERFORMANCE
METRIC
A. Simulation model
Network Simulator2 (NS-2) [29] a object-oriented,
discrete event driven network simulator developed at UC
Berkely written in C++ and OTcl, particularly popular in
the ad hoc networking research community is use for the
simulations. The traffic sources are CBR (continuous bit
– rate). The source-destination pairs are spread randomly
over the network. The node movement generator of ns-2
is used to generate node movement scenarios. The
movement generator takes the number of nodes, pause
time, maximum speed, field configuration and simulation
time as input parameters. The parameter, which is of
primary importance, is pause time. Pause time basically
determines the mobility rate of the model, as pause time
increases the mobility rate decreases. At the start of the
simulations nodes are assigned some random position
within the specified field configuration, for pause time
seconds nodes stay at that position and after that they
make a random movement to some other position. The
propagation model is the Two way ground model [30].
Each data point in the following figures is the average of
5 runs each lasting for 500s of the simulated time with the
same scenario configuration but different random seeds.
Simulation parameters are listed in table 3.
TABLE 3
SIMULATION PARAMETERS
B. Performance metrics
The following performance metrics are considered for
evaluation:
Packet Delivery Fraction: The ratio of the data packets
delivered to the destinations to those generated by the
sources.
Average end-to-end delay: This includes all possible
delays caused by buffering during route discovery
latency, queuing at the interface queue, retransmission
delays at the MAC, and propagation and transfer times.
Normalized routing load: The number of routing
packets transmitted per data packet delivered at the
destination.
Throughput: The amount of data transmitted by the
network divided by time period.
IV. RESULTS AND DISCUSSION
The simulation results are shown in the following
section in the form of line graphs. Graphs show
comparison between the two protocols on the basis of the
above-mentioned metrics as a function of node density in
high mobility (pause time = 0s) and stationary (pause
time = 100s) scenarios with 30%, 70% traffic sources.
Parameter Value
Simulator ns-2
Protocols simulated DSR and CBRP
Number of nodes 50,100,150,200,250
Traffic type CBR (UDP)
Traffic sources 30% of nodes
70% of nodes
Simulation time 100 seconds
Simulation area 2000 m x 500 m
Transmission range 250 m
Node movement model Random waypoint
Propagation model Two way ground
Data payload 512 bytes/packet
Packet rate 4 packets/sec
Node pause time 0s highest mobility
100 s lowest mobility
Speed 20 m/s
Bandwidth 2 Mb/s

RESEARCH PAPER
109
A. Packet delivery fraction
Figure 1 shows the packet delivery fraction as a
function of both nodes density and different number of
traffic sources. Each graph shows the PDF of both the
protocols, DSR and CBRP, in a high mobility (pause time
= 0s) and stationary (pause time = 100s) scenarios.
(a) 30% sources
(b) 70% sources
Figure1 Packet delivery fraction as a function of 50, 100, 150, 200, 250
nodes per sq. km and 30%, 70% traffic sources
In static scenarios (p = 100s) with 30% CBR sources,
PDF is nearly 100% for both the protocols. For DSR,
PDF remains 100% upto number of nodes, N, = 100 per
sq. km after that PDF slowly degrades to 99.80% at N =
250 with an overall decrease of 0.2% from N = 50 to N =
250, whereas for CBRP the PDF remains 100% upto N =
200 and degrades to 99.96% from N = 200 to N = 250
with an decrease of just 0.04%. In high mobility (p = 0s)
scenarios, the packet delivery fraction of DSR degrades
from 95.32% at N = 50 to 69.14% at N = 250 with an
overall decrease of 26.18% as the density of nodes
increases, whereas in CBRP packet delivery improves
from 96.98% at N = 50 to 100% at N = 200 after that
PDF degrades as the density increases.
However in stationary (p = 100s) scenarios with 70%
CBR sources, the PDF of DSR increases from 96.56% at
N = 50 to 99.77% at N = 150 after that it degrades to
97.75% at N = 250, whereas for CBRP the PDF increases
from 96.82% at N = 50 to 100% at N = 150 after that
PDF degrades to 98.97% at N = 250. Both the protocols
scale well with increasing number of nodes per sq. km.
However this is not true for high mobility (p = 0s)
scenarios. For DSR the PDF degrades from 95.79% at N
= 50 to 31.73% at N = 250, whereas for CBRP the PDF
first increases from 95.62% at N = 50 to 99.98% at N =
150 then degrades to 61.88% at N = 250. Both the
protocols have nearly comparable PDF (DSR 95.79%
and CBRP 95.62%) at N = 50 but as the number of nodes
per sq. km increase, CBRP performs better than DSR and
with N = 200 or more CBRP clearly outperforms DSR at
a factor of ~ 2.
DSR has a lower PDF than CBRP in high mobility (0s
pause time) scenarios and more or less same PDF in
stationary (100s pause time) scenarios. In high mobility
scenarios, the PDF due to 30% traffic sources is better
(with 200 or more nodes PDF is two times more) than
that due to 70% traffic sources. The performance
degradation in PDF is due to packet drops by the routing
algorithm after being failed to transfer data in the active
routes. The packet drops are due to network partitioning,
link break, collision and congestion in the ad hoc
network.
B. Average end to end delay
Figure 2 shows average end to end delay of both the
protocols, DSR and CBRP, as a function of both nodes
density- 50, 100, 150, 200 and 250 nodes per sq. km and
different number of traffic sources- 30% and 70% CBR
sources. Each chart shows throughput of both the
protocols in a high mobility (p = 0s) and a stationary (p =
100s) scenarios.
In high mobility (p = 0s) scenarios with 30% CBR
traffic sources, average end to end delay for DSR
decreases 10 times from 0.2221s at N = 50 to 0.0207s at
N = 100 after that it increases gradually till N = 200
where it increases 5 times from 0.0768s to 0.3924s at N =
250, whereas for CBRP average delay decreases up to N
= 200 after that it increase slightly. For low density (up to
100 nodes per sq. km) DSR performs better than CBRP
whereas CBRP performs better for large density (more
than 100 nodes per sq. km). For stationary (p = 100s)
scenarios, both the protocols show comparable average
end to end delay except at N = 50 where DSR (0.0059s)
performs better than CBRP (0.1955s).
In high mobility (p = 0s) scenarios with 70% traffic
sources, the average end to end delay for DSR is below
0.1s with 50 and 100 nodes per sq. km. As the number of
nodes increases, delay grows quickly and at 250 nodes
per sq. km delay is about 0.9s, whereas in case of CBRP
the average delay decreases from 0.3211s at N = 50 to
0.0202s at N = 150 and with increasing number of nodes
the delay grows slowly (with 250 nodes this delay is 4
times less than that of DSR). In stationary (p = 100s)
scenarios both protocols show more or less same average
delay as the number of nodes increases.
With number of nodes less than 150, DSR has lower or
same average end to end delay in high mobility (p = 0s)
and static (p = 100s) scenarios with both traffic sources.
With increasing number of nodes beyond 150, DSR has
higher average delay than CBRP with both traffic sources
and in high mobility scenarios. Since an increasing
number of nodes means an increase in path length, longer
routes takes longer time to discover and to travel along
them. They also have a higher chance of breaking than
30% sources
0
20
40
60
80
100
120
50 100 150 200 250
Number of nodes per sq. km
Packetdeliveryfraction
(%)
DSR, p=0s
CBRP, p=0s
DSR, p=100s
CBRP, p=100s
70% sources
0
20
40
60
80
100
120
50 100 150 200 250
Packetdeliveryfraction
(%)
DSR, p=0s
CBRP, p=0s
DSR, p=100s
CBRP, p=100s

RESEARCH PAPER
110
shorter ones; as a result more route repairs are needed,
which in turn result in a higher delay of the data packets,
whereas for CBRP this behavior is due to their cluster
based structure. Low density may form large number of
small loosely connected clusters and may cause network
to be frequently disconnected by movement of any node,
which introduce delay in route setup, results in higher
average end to end delay of data packets.
(a) 30% sources
(b) 70% sources
Figure2 Average end to end delay as a function of 50, 100, 150, 200,
250 nodes per sq. km and 30%, 70% traffic sources.
C. Normalized routing load
Figure 3 shows the normalized routing load as a
function of both nodes density and different number of
traffic sources. Each graph shows the PDF of both the
protocols, DSR and CBRP, in a high mobility (pause time
= 0s) and stationary (pause time = 100s) scenarios.
In high mobility (p = 0s) with 30% traffic sources,
DSR exhibits NRL below 1 with 50 nodes per sq. km but
as the number of nodes increases, NRL grows and with
250 nodes NRL is about 12, whereas NRL is well below
4 ( lies between 2.8 and 3.3) for CBRP. In static (p =
500s) scenarios, NRL for DSR grows with number of
nodes but is well below 1 except at N = 250 where it is
just above 1 (1.06). NRL for CBRP is more or less
similar to NRL in high mobility (p = 0s) scenarios. In
stationary scenarios DSR performs better than CBRP.
NRL for DSR is just above 1 (1.08) with 50 nodes per sq.
km but as the number of nodes increases, NRL grows and
with 250 nodes NRL is of the order of 30, whereas NRL
is well below 3 ( lies between 1.25 and 2.97) for CBRP.
In static (p = 100s) scenarios, NRL for DSR grows with
number of nodes but is well below 1 except at N = 250
where it is more than 1 (1.84), whereas NRL for CBRP is
always more than 1 (lies between 1.21 and 1.75) as the
number of node increases.
(a) 30% sources
(b) 70% sources
Figure3 Normalized routing load as a function of 50, 100, 150, 200, 250
nodes per sq. km and 30%, 70% traffic sources.
NRL for DSR is just above 1 (1.08) with 50 nodes per sq.
km but as the number of nodes increases, NRL grows and
with 250 nodes NRL is of the order of 30, whereas NRL
is well below 3 ( lies between 1.25 and 2.97) for CBRP.
In static (p = 100s) scenarios, NRL for DSR grows with
number of nodes but is well below 1 except at N = 250
where it is more than 1 (1.84), whereas NRL for CBRP is
always more than 1 (lies between 1.21 and 1.75) as the
number of node increases.
CBRP has lower NRL than DSR in high mobility (0s
pause time) and a higher NRL in case of stationary (100s
pause time) scenarios. This is because CBRP only
broadcasts route requests to cluster heads. Gateway nodes
receive the route requests as well but they forward them
to the next cluster-heads. This largely reduces the route
discovery packets which in turn reduces NRL.
D. Throughput
Figure 4 shows throughput of both the protocols, DSR
and CBRP, as a function of both nodes density- 50, 100,
150, 200 and 250 nodes per sq. km and different number
of traffic sources- 30% and 70% CBR sources. Each chart
30% sources
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
50 100 150 200 250
Averageendtoend
delay(sec)
DSR, p=0s
CBRP, p=0s
DSR, p=100s
CBRP, p=100s
70% sources
0
0.2
0.4
0.6
0.8
1
50 100 150 200 250
Averageendtoend
delay(sec)
DSR, p=0s
CBRP, p=0s
DSR, p=100s
CBRP, p=100s
30% sources
0
2
4
6
8
10
12
14
50 100 150 200 250
Normalizedroutingload
DSR, p=0s
CBRP, p=0s
DSR, p=100s
CBRP, p=100s
70% sources
0
5
10
15
20
25
30
35
50 100 150 200 250
Normalizedroutingload
DSR, p=0s
CBRP, p=0s
DSR, p=100s
CBRP, p=100s

RESEARCH PAPER
111
shows throughput of both the protocols in a high mobility
(p = 0s) and a stationary (p = 100s) scenarios.
In high mobility (p = 0s) and zero mobility (p = 100s)
scenarios with 30% CBR sources the throughput of both
the protocols improve with increasing nodes density. In
high mobility (p = 0s) both the protocols depicts nearly
comparable throughput (DSR 5.42kB/s, CBRP 5.54kB/s
at N = 50 and DSR 9.44kB/s, CBRP 9.92kB/s at N = 100)
up to 100 nodes per sq. km. After that CBRP outperforms
DSR and at N = 250 this difference is quit significant
(DSR 19.06kB/s and CBRP 27.55kB/s). In static (p =
100s) scenario both the protocols show no significant
difference in throughput and scales well with increasing
nodes density. Chart (a) depicts that DSR at p=100s and
CBRP at p=0s, p=100s shows similar behavior with
increasing number of nodes per sq. km.
(a) 30% sources
(b) 70% sources
Figure 4 Throughput in kB/s as a function of 50, 100, 150, 200, 250
nodes per sq. km and 30%, 70% traffic sources.
In high mobility (p = 0s) and with 70% CBR traffic
sources the throughput of both the protocols increase up
to N = 150 for DSR and N = 200 for CBRP, after that the
throughput degrades. Both have nearly same throughput
up to N = 100 and with increasing nodes density CBRP
performs better than DSR and with N = 200 or more
CBRP clearly outperforms DSR at a factor of ~ 2. In
stationary (p = 100s), throughput of CBRP improves with
increasing number of nodes per sq. km, whereas for DSR
throughput improves up to N = 200 after it degrades.
CBRP has a better throughput than DSR in high
mobility and stationary scenarios with both traffic
sources. This better throughput comes from its cluster
based structure which largely reduces network traffic.
V. CONCLUSION
This paper compared the performance of DSR and
CBRP, two on-demand source routing protocols for
mobile ad hoc networks. DSR and CBRP both use on-
demand route discovery, but with different flooding
behavior. In particular, DSR use network-wide flooding
for route discovery and does not depend on any periodic
hello message or timer-based activities. CBRP, on the
other hand, only broadcasts route requests to cluster
heads, largely reducing the network traffic. Hello
messages are the integral part of CBRP and the size of a
hello message may be large as it contains the neighbor
table and cluster adjacency table of the sender. As a
result, while CBRP uses hello messages to establish
clusters and in turn reduce the flood in route discovery,
the hello message itself is another kind of overhead. The
general observation from the simulation is that in static
(pause time 100s) scenarios, due to similar performance
any, either DSR or CBRP, can be used in large scale
mobile ad hoc networks but in high mobility (pause time
0s) scenarios, CBRP outperforms DSR. CBRP scales
well with increasing number of nodes per sq. km.
Further when compared results of this work with
earlier reported similar works [16, 19-22, 25,26] it clearly
show that the previous results for 50 and 100 nodes
closely matches with the result of this work for number of
nodes up to 100. However earlier works did not consider
number of nodes above 100. Hence results of this work
are inline with previously reported results and can be
considered as statistically confident for those scenarios.
We are currently analyzing more precisely the effect of
mobility and network load on the performance of routing
protocols (DSR and CBRP) in large scale mobile ad hoc
networks. CBRP broadcasts periodic hello messages to
establish and maintain clusters, clusters reduce network
traffic but at a cost of extra overhead in the form of hello
messages. We believe that there are some ways to reduce
this overhead. We plan to investigate and study on how to
improve the performance of CBRP by reducing clustering
overhead.
REFERENCES
[1] E. M. Royer and C. K. Toh, “A review of current routing
protocols for ad hoc mobile wireless networks”, IEEE
Personal Communications magazine, April 1999, pp. 46–
55.
[2] C. E. Perkins and P. Bhagwat, “Highly Dynamic
Destination-Sequenced Distance-Vector Routing (DSDV)
for Mobile Computers”, in Proceedings of ACM
SIGCOMM 1994, August 1994, pp. 234-244.
[3] S. Murhty and J. J. Garcia-Luna-Aceves, “An Efficient
Routing Protocol for Wireless Networks”, ACM Mobile
Networks and Applications Journal, Special Issue on
Routing in Mobile Communication Networks, Vol. 1, no.
2, October 1996, pp. 183-197.
[4] C. C. Chiang, H. K. Wu, W. Liu and M. Gerla, “Routing
in Clustered Multi-Hop Mobile Wireless Networks with
30% sources
0
5
10
15
20
25
30
50 100 150 200 250
Throughput(kB/s)
DSR, p=0s
CBRP, p=0s
DSR, p=100s
CBRP, p=100s
70% sources
0
10
20
30
40
50
50 100 150 200 250
Throughput(kB/s)
DSR, p=0s
CBRP, p=0s
DSR, p=100s
CBRP, p=100s

RESEARCH PAPER
112
Fading Channel”, in Proceedings of IEEE SICON 1997,
April 1997, pp. 197-211.
[5] J. J. Garcia-Luna-Aceves and M. Spohn, “Source-Tree
Routing in Wireless Networks”, in Proceedings of IEEE
ICNP 1999, October 1999, pp. 273-282.
[6] D. B. Johnson and D. A. Malta, “Dynamic Source
Routing in Ad Hoc Wireless Networks”, Mobile
Computing, Kluwer Academic Publishers, vol. 353, 1996,
pp. 153-181.
[7] C. E. Perkins and E. M. Royer, “Ad Hoc On-Demand
Distance Vector Routing”, Proceedings of IEEE
Workshop on Mobile Computing Systems and
Applications 1999, February 1999, pp. 90-100.
[8] V. D. Park and M. S. Corson, “A Highly Adaptive
Distributed Routing Algorithm for Mobile Wireless
Networks”, in Proceedings of IEEE INFOCOM 1997,
April 1997, pp. 1405-1413.
[9] Y. Ko and N. H. Vaidya, “Location-Aided Routing
(LAR) in Mobile Ad Hoc Networks”, in Proceedings of
ACM MOBICOM 1998, October 1998, pp. 66-75.
[10] C. K. Toh, “Associativity-Based Routing for Ad Hoc
Mobile Networks”, Wireless Personal Communications,
vol. 4, no. 2, March 1997, pp. 1-36.
[11] P. Sinha, R. Shivkumar and V. Bharghavan, “CEDAR: A
Core Extraction Distributed Ad Hoc Routing Algorithm”,
IEEE Journal on Selected Areas in Communications, vol.
17, no. 8, August 1999, pp. 1454-1466.
[12] Z. J. Haas, “The Routing Algorithm for the
Reconfigurable Wireless Networks”, in Proceedings of
ICUPC 1997, vol. 2, October 1997, pp. 562-566.
[13] R. S. Sisodia, B. S. Manoj and C. Siva Ram Murthy, “A
Preferred Link-Based Routing Protocol for Ad Hoc
Wireless Networks”, Journal of Communications and
Networks, vol. 4, no. 1, march 2002, pp. 14-21.
[14] Andrew S. Tanenbaum, Computer Networks. Fourth
Edition, Prentice Hall, ch. 5, pp. 357-360.
[15] Andrew S. Tanenbaum, Computer Networks. Fourth
Edition, Prentice Hall, ch. 5, pp. 360-366.
[16] J. Broch, D. A. Maltz, D. Johnson, Y. C. Hu and J.
Jetcheva, “A Performance Comparison of Multi-Hop
Wireless Ad Hoc Network Routing Protocols”, in
Proceedings of ACM/IEEE MOBICOM 1998, October
1998, pp. 85-97.
[17] S. R. Das, C. E. Perkins and E. M. Royer, “Performance
Comparison of Two On-Demand Routing Protocols for
Ad Hoc Networks”, in Proceedings of the IEEE
INFOCOM 2000, March 2000, pp. 3-12.
[18] Samir R. Das, Charles E. Perkins, Elizabeth M. Royer and
Mahesh K. Marina. ”Performance Comparison of Two
On-demand Routing Protocols for Ad hoc Networks”,
IEEE Personal Communications Magazine special issue
on Ad hoc Networking, February 2001, pp. 16-28.
[19] A. Boukerche, “Performance comparison and analysis of
ad hoc routing algorithms”, in Proceedings of IEEE
IPCCC 2001, April 2001, pp. 171-178.
[20] A. Boukerche, “A performance comparison of routing
protocols for ad hoc networks”, in Proceedings of 15th
IPDPS 01, April 2001.
[21] A. Boukerche, “Simulation based performance
comparison of routing protocols for mobile ad hoc
networks”, Simulation, Vol. 78. Issue 7, July 2002, pp.
401-407.
[22] A. Boukerche, “Performance evaluation of routing
protocols for ad hoc wireless networks”, Mobile
Networks and Applications 9, 333–342, 2004.
[23] P. Johansson, T. Larsson, N. Hedman, B. Mielczarek and
M. Degermark, “Scenario-based Performance Analysis of
Routing Protocols for Mobile Ad-hoc Networks”, in
Proceedings of the ACM/IEEE MOBICOM 1999, August
1999, pp.195-206.
[24] S. R. Das, R. Cestaneda, J. Yan and R. Sengupta,
“Comparative performance evaluation of routing
protocols for mobile ad hoc networks”, in proceedings of
the IC3N 1998, October 1998, pp. 153-161
[25] Narendra Singh Yadav and R.P.Yadav, “The Effects of
Speed on the Performance of Routing Protocols in Mobile
Ad Hoc Networks”, International Journal of Electronics,
Circuits and Systems, vol.1, no. 2, 2007, pp. 79-84.
[26] Narendra Singh Yadav and R.P.Yadav, “Performance
comparison and analysis of table-driven and on-demand
routing protocols for Mobile Ad Hoc Networks”,
International Journal of Information Technology, vol. 4,
no. 2, 2007 , pp. 101-109.
[27] D.Johnson, D.Maltz, Y.Ho, “The dynamic source routing
protocol for mobile ad hoc networks”, IETF Manet
working group, Internet RFC 2026.
[28] Mingliang Jiang, Jinyang Li and Y.C.Tay, “Cluster Based
Routing Protocol”, August 1999 IETF Draft.
http://www.ietf.org/internet-drafts/draft-ietf-manetcbrp-
spec-01.txt
[29] K. Fall and K. Vardhan, The Network Simulator (ns-2).
Available: http://www.isi.edu/nsnam/ns
[30] T. S. Rappaport, Wireless Communications, Principles &
Practices. Prentice Hall, 1996, ch. 3, pp. 70-74.

Source routing in Mobile Ad hoc NETworks (MANETs)

Recommended

Recommended

More Related Content

What's hot

What's hot (18)

Similar to Source routing in Mobile Ad hoc NETworks (MANETs)

Similar to Source routing in Mobile Ad hoc NETworks (MANETs) (20)

More from Narendra Singh Yadav

More from Narendra Singh Yadav (7)

Recently uploaded

Recently uploaded (20)

Source routing in Mobile Ad hoc NETworks (MANETs)