62 (IJCNS) International Journal of Computer and Network Security,

Vol. 2, No. 5, May 2010

Performance of DSR Protocol over Sensor

Networks
Khushboo Tripathi1, Tulika Agarwal2 and S. D. Dixit3
1
Department of Electronics and Communications
University of Allahabad, Allahabad-211002, India
khushboo83@live.com
2
Department of Electronics and Communications
University of Allahabad, Allahabad-211002, India
tulikabbs09@gmail.com

Abstract: Routing protocols play crucial role in determining

performance parameters such as throughput parameter, packet 2. Related Work
delivery fraction, end to end (end 2 end) delay, packet loss etc. of Dynamic Source Routing (DSR) Protocol: DSR is
any ad hoc communication network. In this paper the
characterized by source routing i.e. sender knows the
performance of DSR protocol in sensor network of randomly
complete hop-by-hop route to the destination. It emphasizes
distributed static nodes with mobile source node is investigated
with source velocities 10, 20, 40 and 60 m/sec. and node the aggressive use of route caches that store full paths to the
densities 40-100 nodes/km2 . On varying velocity it is observed destination as given in figure1 (a) & (b) [2,7 ]. This protocol
that throughput of received packets is higher which reduces the has the advantages of loop-free routing and avoidance of the
overheads in the network. The delivery ratio is also maximum need for up-to-date routing information in the intermediate
except at node density 60/km2.When the number of dropped nodes.
packet is considered the average end to end delay is decreases
except at 80 nodes/km2 and so throughput is high. Hence DSR
performs better in this scenario and reduces minimum
overhead.

Keywords: Sensor Networks, DSR, Node Density.

1. Introduction
The recent development in small embedded sensing devices
and the wireless sensor network technology has provided
Figure1 (a). RREQ Broadcast
opportunities for deploying sensor networks to a range of
applications such as environmental monitoring, disaster
management, tactical applications etc [1].Main requirement
for such application is that motes carrying onboard sensors
should be physically small, low power consuming and
include wireless radio.
For data collection a straight forward solution is that each
mote transmits its data to a centralized base station.
However in such cases the energy requirement of each node
would be large which reduces mote life and also there would
Figure1 (b). RREP Propagation
be interference problem. Alternative approach for harvesting
data from sensor fields uses mobile data collector such as
It is composed of two main mechanisms: Route Discovery
robots which move in the sensor field to collect data and
and Route Maintenance. For route discovery the source node
transmit the same to base station in real /non-real time [6].
floods the route request (RREQ) packets in the network. The
In the present paper, the performance of dynamic source
nodes receiving RREQ rebroadcast it and the process repeats
routing (DSR) protocol has been analyzed keeping in mind a
until the destination node or an intermediate node having a
sensor network scenario wherein all the nodes are static and
route to the destination is found. Such a node replies back to
source node is moving ( one of which is a data harvester
the source with a RREP packet. When a link-break in an
from static nodes and other one acting as a sink).
established route occurs, the route maintenance phase is
Section 2 describes the related work and Section 3 is about
initiated wherein upstream node at the link-break site
the network scenario and all about the definition of
creates a route error (RERR) message and sends it to the
simulation parameters and details of simulation experiment.
source node. On receiving RERR the source node utilizes
Section 4 gives the results and discussion of the simulation.
alternate routes from the route cache, if they are available, to
Section 5 concludes the work.
prevent another route discovery. The drawback with DSR is
that it needs to place entire route in both the route replies
and the data packets and thus requires greater control
overhead.
 (IJCNS) International Journal of Computer and Network Security, 63
Vol. 2, No. 5, May 2010

3. Network Scenario and Simulation Details

The simulation scenario consists of tcl script that runs over
TCP connections for various number of nodes (40, 60, 80
and of 100 nodes) in an area of size of 1000 1000m2. Table1: Parameter values for simulation
Figure 2 shows the random node topology of the network.
The simulation time is set to 150 seconds. TCP connection Maximum simulation 150 seconds
time
has been used to set communication between the source
Area size (Flat area) 1000×1000 m2
node and the sink node.
Routing protocol DSR
(proactive)
Propagation Model Two Ray Ground
Propagation
MAC layers protocol IEEE802.11
Node placement Static Random
Distribution
Number of nodes 40,60,80,100
Velocities of source 10,20,40,60 m/sec
and sink nodes

4. Results and Discussion

The throughput vs. node density graph is given in figure3.
According to our topology considerations, in sparse
network, for the velocities 10, 20, 40 and 60 m/sec different
throughput graphs are obtained. For all the velocities at the
Figure 2. NAM Windows showing network topology for
node density 60 per square km minimum throughput is
100 nodes
obtained while higher velocities give maximum throughput
with node density parameter. At velocity 10m/sec
The simulation is carried out in the network simulator NS-2
throughput is minimum whereas at velocity 60 m/sec
(version-2.31) over LINUX (FEDORA 8) environment [3]
throughput is higher in comparison to other velocities. This
[4].The experiment has been performed for a set of 5
is because of the nature of DSR on-demand routing protocol.
random network topologies having number of nodes
Once the route is formed the possibility of (link-breakage) is
40,60,80 and 100 respectively. In each topology all nodes
lesser as compared to the former case where the probability
are fixed except the source node. For a given topology two
of route formation is lower. Hence it is clear from figure
nodes are selected to be source-sink nodes and the
after node density 60 per square km throughput is generally
observations are made for five different velocities 10, 20, 40
increasing in nature with different node densities due to less
and 60 m/sec. The experiment was repeated by selecting
link breakage in route formation. It is obtained that When
another pairs of source-sink nodes chosen randomly. Five
the throughput of received packets is maximum the number
such observations were taken for each of the following three
of overheads reduces in the network.
performance parameters and their averages are obtained to
plot graphs: 6000
nv10
1. Bytes Delivery Fraction: The ratio of the number of nv20
nv40
5500
data sends (in bytes) successfully delivered to the nv60

destinations to those are generated by TCP agents/sources. 5000

Thus, Bytes Delivery Fraction = (Received data /Total sent
Throughput

data) 4500

2. Average Number of Dropped Packets: This parameter

is worth mentioning while taking the effectiveness of this 4000

routing protocol. These are the average number of dropped

3500
packets in simulation study of our scenario.
3. Average Simulation End to End Delay: The delivery 3000
40 50 60 70 80 90 100
delay for the data is the interval between when it is Node Density(per square km)

generated at a sensor node and when it is collected by a sink

[2]. It does not depend on the time for collecting data from
the network. These are the possible delays caused by
buffering during route discovery latency, queuing at the
interface queue, retransmission delays at the MAC and
propagation and transfer times.
Simulation Parameters: The parameter values for
simulation are given in table 1.
Figure3.Throughput (of received packets) vs. node density
The packet delivery fraction vs. node density behavior is
given in figure 4. We have seen the effect of delivery ratio
with node density on varying the velocities 10, 20, 40 and
60 m/sec. The delivery fraction graph is almost similar to
the throughput vs. node density figure. On varying node
density the delivery fraction is increased for all the velocities
except at node density 60 per square km.
 64 (IJCNS) International Journal of Computer and Network Security,
Vol. 2, No. 5, May 2010

[3] The Network Simulator NS-2 homepage,

0.8
http://www.isi.edu/nsnam/ns
[4] The Network Simulator NS-2 tutorial homepage
http://www.isi.edu/nsnam/ns/tutorial/index.html
[5] E. M. Royer, and C. K. Toh. “A review of current
0.75
routing protocols for ad hoc mobile wireless
Delivery Ratio

networks”,IEEE Personal Communications, (April),

1999.
[6] J. Rao, and S.Biswas, “Data Harvesting In Sensor
0.7 Networks Using Mobile Sinks”, IEEE Wireless
nv10 Communication, pp.1536-1284, 2008.
nv20
nv40
[7] M Tamilarasi, V.R.S.Sunder, “Scalability Improved
nv60 DSR Protocol for MANETs”, ICCIMA-pp.283-287,
0.65 2007.
40 50 60 70 80 90 100
Node Density(per square km)

Figure4. Delivery Ratio vs. node density Authors Profile

The average simulation end to end delay is given in figure5.
Khushboo Tripathi is a D.Phil. student
According to the topology we found a vibrating nature of delay
in the Department of Electronics and
in graph. But generally it decreases with the node density. An Communication of J.K.Institute of
exceptions is attained at node density = 60 nodes/km2.When the Applied Physics and Technology
delay is high i.e. high node density in the network and the route ,University of Allahabad, Allahabad She
finding is difficult in that scenario whereas in case of less delay obtained her master’s degree in
the possibility of finding routes is more. From above graph it is mathematics in 2005 from University of
evident that for the higher velocity the throughput is maximum Allahabad. Also she received the MCA
and corresponding delay is less as in figure5 and vice versa. degree from UPRTOU, Allahabad. Her research is in the
area of Wireless Communications and networking,
0.41 particularly in MANETs, WSN and coding, cryptography.
nv10
0.4 nv20
nv40 Tulika Agarwal is a M.Tech. student in
Avg.Simulation End to End Delay(sec.)

0.39 nv60 the Department of Electronics and

0.38
Communication of J.K.Institute of
Applied Physics and Technology,
0.37 University of Allahabad, Allahabad. She
obtained her B.Tech. in Electronics and
0.36
Communication Engineering from
0.35 BBSCET, Allahabad in 2007. She is
carrying out her M.Tech. project on “A Comparative Study
0.34
of Routing Protocols in Ad Hoc Networks”. She has also
0.33 interest in “Digital Electronics” and “Wireless
Communication”.
0.32
40 50 60 70 80 90 100
Node Density(per square km)
S.D.Dixit is a professor in the
Figure5. Avg.simulation end2end delay vs. node density Department of Electronics and
Communication Engineering, University
of Allahabad at Allahabad. He has
teaching experience of 33 years in the
5. Conclusion field of Electronics and Communication
In the given scenario with increase in velocity throughput of Engineering at J.K.Institute of Applied
received packets is higher which reduces the overheads in Physics and Technology, University of
the network. Packet delivery ratio is maximum as Allahabad, Allahabad. He obtained M.Tech. and D.Phil.
throughput of received packet is high except at node density degrees in 1975 and 1981 respectively. He has published a
60/km2.When the number of dropped packet is considered number of research papers in National and International
Journals. He is the fellow of IETE, India. His current
the average End to End delay is decreases except at 80
interests are in the field of Wireless Communications,
nodes/km2 and so throughput is high. Therefore DSR Coding and Cryptography.
performs better in this scenario and reduces minimum
overhead.

References
[1] Anna Hac, “Wireless Sensor Network Designs”, John
Wiley and Sons, 2003.
[2] M.K. Marina, S. R. Das, “Performance of Route Caching
Strategies in Dynamic Source Routing,” ICDCS, pp.
425-432,2001.