Enhanced DSR: An Efficient Routing Protocol for MANET

Uma Rathore Bhatt* ,Neelesh Nema, Raksha Upadhyay

Department of Electronics &Telecommunication Engineering, Institute of Engineering & Technology, Devi Ahilya University,

Indore -452017, India

* Corresponding author- umarathore@rediffmail.com

Abstract- An ad-hoc network is a group of wireless mobile

nodes which are self-creating, self-organizing, and self­

administering. In these networks, selection of routing protocol is

an important concern, which may enhance the performance of

ad-hoc networks in terms of residual battery, average end to end

delay, average jitter and throughput. To fulfill this aim, an

efficient algorithm viz. enhanced DSR (DSRl) is proposed in the

present work. This algorithm reduces the flooding of RREQ

(route request) packets in the network which results in the

reduced congestion and energy consumption. In this when a node

receives a RREQ packet it checks, its own residual battery,

received signal strength and speed. If the defined thresholds for

the node for these mentioned parameters satisfy then the RREQ

packet will be forwarded in the network. If it doesn't meet the

defined threshold, the RREQ packet will be discarded. Hence the

proposed algorithm prohibits the unnecessary flooding of RREQ

packets. This results in the improved system performance.

Keywords-Ad hoc networks, DSRl, flooding, RREQ,

routing protocols,

1. INTRODUCTION

Ad-hoc networks are wireless networks that have no fixed infrastructure. Such kind of networks is created dynamically

and they provide special challenges beyond those in standard

data network. The structure of wireless ad-hoc network was

initially designed to be a flat type, which means that the ad­hoc network is assumed to be homogeneous. Based on the flat

structure, the routing work can be simply done by flooding. However, when the number of nodes in the network grows to hundred or even thousands, the mass control message will

even block the data transmission. It has been proved that even under the optimal circumstances, the throughput for each node

declines rapidly towards zero while the number of nodes is increased. Because ad-hoc networks do not rely on existing

infrastructure and are self organizing, they can be rapidly deployed to provide robust communication in variety of

hostile environments [1]. A wireless ad-hoc network consists

of various mobile self -organized and battery operated wireless devices having different battery capacity and power dissipation. Hence Battery lifetime becomes a key factor that affects the performances of mobile ad-hoc network.

Various route discovery mechanisms using DSR for

MANET have been given in literature. In [2] the author proposed a mechanism for highway environment which is based on the feature of vehicle movement. The work done in

[3] proposes a route selection mechanism and mobility

prediction for making DSR an energy efficient routing

978-1-4799-2900-9/14/$31.00 ©2014 IEEE

protocol, in this author investigated the impact of mobility on the performance of an on-demand power aware routing

protocol. In [4] author proposed an algorithm which selects a

route having the greatest available bandwidth. Flooding management in ad-hoc network by controlling the mobility

and transmission power is shown in [5]. Various ad-hoc routing protocol and their comparison is shown in [6]. In [7]

author proposed the mechanism for reducing the route request

flooding attack. The design and performance evaluation of a secure on-demand ad-hoc network routing protocol viz. Ariadne has been shown in [8]. It prevents many types of

denial-of-service attacks. In [9] author proposed a technique that can reduce the routing overhead by utilizing prior routing

histories.

In present work we have proposed an efficient and

simple algorithm which reduces the number of RREQ packets flooded in order to enhance the performance of an Ad-hoc

network. This algorithm considers the alteration in route

discovery mechanism at intermediate nodes. The alteration process includes the checking of parameters like mobility,

residual battery and received signal strength of node itself by the defined condition. In this process those nodes are

discarded which don't possess the defined conditions. Therefore those nodes will not further forward route request packets. Hence congestion in network will reduce.

II. NEED FOR REDUCING THE ROUTE REQUESTS

IN AD-HOC NETWORK

The dynamic source routing protocol (DSR) is a reactive MANET routing protocol. When communicating a destination

node which is not present in the route cache of the source

node, it will buffer the data packets and broadcast a route request (RREQ) into the network. The other intermediate DSR nodes broadcast the RREQ packet to the intended destination

node. The destination node then will send a Route Reply (RREP) on the reverse route back to the source node.

If no RREP is received after a fixed number of

attempts and within the net _traversal _time, the data packets from the buffer will be dropped. If more data packets are received, a new route discovery process will be initiated.

Increasing number of route requests will saturate the network

which makes it unable for further transmission of data packets

[6].

Reactive routing protocols generally require less

routing overhead than proactive routing protocols because they work on demand. Proactive routing protocols are not

215

affected by large number of flooding of route request because

if the destinations are not found, data packets are dropped at

the source or any intermediate nodes. But it can seriously degrade the performance of reactive routing protocols and

affect a node in the following ways:

During the process of route discovery the data packets

are buffered in the memory. And, if the large numbers of data packets originating from application

layer are actually unreachable, genuine data packets

in the buffer memory may be replaced by these unreachable data packets.

The increased number of RREQ packets in the network results in more collisions and consequently,

congestion as well as delay increase for the data

packets in the network. Power and bandwidth management is important in

MANET and transmission of large number of RREQ

can increase the power and bandwidth consumption.

III. BASIC DSR OPERATION

Existing DSR Algorithm: In existing DSR algorithm when

the packet arrives, first of all, node checks following conditions/possibilities:

• If it is RREQ packet then the route discovery

mechanism initiates and the step one of the algorithm

(shown in fig.3) will be executed. • If it is RREP packet then route reply mechanism

works and the step two of the algorithm (shown in

figA) will be executed. • If it is data packet then data is forwarded from source

to destination. This is shown in figure 2.

Extract the packets

True True True

Route discovery mechanism Route reply mechanism Data transm i ssion

Fig.2 Basic operation- Flow chart of Dynamic Source routing protocol [6]

Step-I: Route discovery mechanism Route discovery mechanism is as follows:

• When RREQ flooded, it reaches to the nodes which

are in the transmission range of source node. When a

node receive a RREQ packet, following conditions/possibilities will be checked:

(i) If it is the destination node, it sends a route reply to the sender.

(ii) If this node receiving the route request has recently seen another route request message from this initiator

bearing the same request identification and target address it discard the request.

OR

If this node's own address is already listed in the

route record in the route request, this node discards

the request.

(iii) If the above condition does not hold then the node add it's on address in the request packet and forward

it.

False

True

Packet with

sameid

False

Fig.3 Basic Route discovery mechanism in DSR [6]

Step -II: Route Reply mechanism: Route Reply mechanism is as follows:

• If routes are bidirectional, then reverse path is set up

and place that path in the header of the packet and

forward it to the source. • Otherwise, if an entry to the source node exists in the

route cache of destination node, forward the packet to

the source by using that route. • If above condition does not happen, then destination

initiates the route discovery for source node, but to

avoid possible infinite recursion of route discoveries, it must piggyback this route reply on the packet

containing its own route request for source.

216 2014 International Conference on Issues and Challenges in Intelligent Computing Techniques (ICICT)

False An enbyto the

source node False Routes are

bidirectional exists in the

route cache of

the destination

node

Initiate route

discovetyfrom

\----.1 destination to

source node

True

Place the header and forward

the packet to the source

F oJWard the route reply

packet to the source by using

route

Fig.4 Basic Route Reply mechanism [6]

IV. PROPOSED DSRI ALGORITHM

The flow chart of proposed algorithm viz. DSRI is shown in figure 5. It shows the mechanism of forwarding the route request packets by the intermediate nodes.

Rout. reply mechanism

Fig.5 Route discovery for proposed DSRI

False

For proposed DSRI the modifications made in route discovery process only, which is as follows:

• When RREQ flooded, it reaches to the nodes which are in the transmission range of source node. When a node receive a RREQ packet, following

conditions/possibilities will be checked: (i) If it is the destination node, it sends a route

reply to the sender.

(ii) Elseif this node receiving the route request has recently seen another route request message from

this lllltiator bearing the same request identification and target address, or if this node's own address is already listed in the route record in the route request, this node discard the request. Else

If {value of the speed of the node is greater than 12 m/sec OR value of the residual battery is less than the threshold value of

residual battery i.e. = 4600 mAhr OR value of the received signal strength is less than

the threshold received signal strength i.e. = -

85 dBm} Discard the request.

Else the node adds its own address in the request packet and forwards it.

V. RESULTS AND DISCUSSION

To evaluate and compare the performance of DSRI protocol with existing DSR protocol, a simulation code has been developed in Qualnet 5.0.2 simulator. Various

performance metric like average jitter, average end to end delay, throughput, residual battery energy consumed in

transmit/ receive mode and packets received have been calculated.

An extensive simulation has been carried out to define the threshold value of residual battery, speed and RSS for DSRI (shown in table 1) using networks consisting of 40, 80, 120,

160, 200 and 240 nodes.

T ABLE. I THRESHOLD VALUE OF VARIOUS PARAMETERS FOR DSRI

Parameter Routing protocol-DSRI

Threshold value of residual battery 4600 mAhr

Threshold value of speed of node 2.98 m/sec (m/sec)

Threshold value of RSS -85 dBm

Following parameters (shown in table2) are taken to run the test scenario for DSR and DSRI protocol. The source and destination are connected through CBR (Constant Bit Rate) link.

TABLE2: SIMULATION PARAMETERS USED FOR DSR/DSRI

Routing Protocol DSRlDSRl

Packet size 512 bytes

Battery Model Linear

Energy Model Generic

Simulation Time 1000 sec

Number of nodes 40.80.120.160,200.240

20J4 Internationai Conference on Issues and Challenges in Intelligent Computing Techniques (ICICT) 217

.0 80 120 100 No. of nodes

200 2.0

Fig.l Average residual battery (mAhr) vs. number of nodes

Figure I shows the average residual battery vs. number of nodes for DSR and DSRI. The figure clearly demonstrates that in DSRI protocol, the residual battery of the network is more in comparison with DSR protocol. It is because, in DSRI less number of route request are forwarded by intermediate nodes with respect to DSR. Simultaneously, DSRI Protocol does not include those intermediate nodes, which are having less residual battery. That's why, average residual battery of the whole network increases.

0.45 .......... , ........... � ........... : ........... ; ........ . : : : '-;---.,,--' , , ,

0.4 ......... . : ........... , ........... : ........... , ........... : ......... . ;. , , , , I , , I I , I , , , , , I , I I I "

U' 0.35 .. (/)

� 0.3 >-

.."! .. 0.25 0 '0

<::

W E '0

<::

W

0.1 , "

I I I I I , _ _______ J ___________ L ___________ , ___________ J __________ .'- __________ J.

, , , , I , , I I , I , , , , , I , I I I I I ,

40 80 120 160 No. of nodes

200 240

Fig.2 Average End-to-End Delay (in sec) vs. number of nodes

Figure 2 shows the average end to end delay vs. number of nodes. It is clear from the fig. that delay decreases for DSRI

as number of nodes increases in the network. This is due to fact that DSRI forwards less number of route request as compared to DSR. This in turn reduced bandwidth consumption and interference. Hence for transmitting the data packets more bandwidth will be available. However DSRI is not suitable for a network with less number of nodes in terms of delay. DSRI offers higher delay as compared to DSR because of its own route discovery mechanism. Hence for the large size networks DSRI protocol will be suitable.

Figure 3 depicts the throughput of the network vs. number of nodes. It is clear from the fig. that network throughput is greater for DSRI as compared to DSR. This is happening due to reduction in congestion, bandwidth consumption and delay. Hence for DSRI, number of successful bits received per second increases and consequently, throughput increases.

u ., (/)

4000 i-:-----::::::F===t=:====:r:::===+====:::::-:i

3500

I I , I , I

3000 _________ J ___________ L ___________ , ___________ l ___________ '-________ .J. I I I I , I I I I I , I I I , I , I I , I , I I , I , I I , I ,

:;; 2500 ------- ---:- - - - - - - - - - - r - ----------:- - - - - - - - - - - + - ----------;-- -- -- --- -- �-Cl.

.� i 2000

� 1500 e

� f--1000

, , . , . , , , • I • , I I • I • , I I • I , I I I , I , I I I , I , I ----- ----,-----------r-----------,-----------T-----------,----------,-

I I I I , I I I I I , I I I , , , , , , , , , , , , , , , , , , , , , , .... ····· ,,··········· r ···········,··········· T ··········· ,. .......... ,.

.. ······· ,··········· r -----------,----------- y -----------,...---------- , -, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , 500 .-.... --.�-.... -.... -� ........... :---.-...... :.---......

............... DSR1

U' .. (/)

-- DSR 40 80 120 160 200 240

No. of nodes

Fig.3 Throughput (in bits/sec) vs. number of nodes

0.35r-:----:----r----:-----;:===c::;,

0.3 --. ---_ .. _,-------._ .. �-. ------... :-. __ . -. ---- ; -_ ... ----t,------,-.l-J

0.25

, . , , , . , , , . , , , • , I , . , , , • I , , ,

__________ J ___________ L ___________ , ___________ l ___________ '-________ _ J_ , • I , , , , , , , I , , " " , I I " , " " , " " , " "

2. 0.2 � '=j � 0.15 ..

0.1 --.--. -- .. '-.--.---.-- � ---.--.---. -.--.---.- ;----. - -. -.� -.--.---.�-

0.05

40

I I " , ' " , ' " , , ,

, , , ,

- � 0 0 __ 0 0 _ 0 0 __ '0 0 _ 0 0 _ 0 0 __ o . _ 0 0 __ 0 0 _ 0 0 _� 0 __

80 120 160 No. of nodes

200 240

Fig.4 Average Jitter (in sec) vs. Number of Node

218 2014 International Conference on Issues and Challenges in Intelligent Computing Techniques (ICICT)

In figure 4, average jitter for DSR and DSRI has been compared. Since DSRI reduces congestion by reducing route

request packets. This consumes lesser bandwidth as compared to DSR which in tum decreased variation in packets arrival. Hence average jitter is found less for DSRI .

VI. CONCLUSION

Need of controlling the flooding of RREQ is very important task in MANET because it consumes more power and bandwidth of the network. It also causes high congestion and delay in the network. The DSRI provides a better solution to overcome it. The simulations are performed using Qualnet

simulation tool to evaluate the performance of proposed and

existing protocol viz. DSRI and DSR. In DSRI route discovery mechanism reduces congestion and energy

consumption by not taking into account those nodes whose battery is lesser, mobility and distance is greater than the defined threshold level. If following assumptions are not

considered the chances of network failure increases. Because

high speed nodes will come into a network for short time period and the node having lesser battery charge will not remain in a network for long duration. DSRI algorithm gives a new technique for route discovery with efficient utilization of

the bandwidth. It results in increased throughput and average

residual battery of the node and decreased average end to end delay and jitter. In all, DSR 1 outperforms the existing DSR.

This protocol can be useful at the place where the node density is very high and delay is to be minimized.

[J]

[2]

[3 ]

[4]

[5]

[6]

[7]

[8]

[9]

REFERENCES

Cheng ,R.H , Wu,T.K , Yu,w.c. , "A highly topology adaptable ad hoc routing protocol with complementary preemptive link breaking avoidance and path shortening mechanisms", Wireless Network, DOl 10. I 007/s1 1276-009-0203-0, pp. I 218-I 3 I 1,20 I O.

Gang , Z., Han, T. , Wenwei,S., Chunfeng, L., Yantai, S., "A Hybrid DTN-DSR Routing protocol Based on Clustering", 8th International Conference on wireless communication Networking and mobile Computing (Wi COM), doi:IO.J 109IWiCOM.2012.647829J, 2012.

Luo , Y., Wang, J. , Chen, S., " An energy efficient DSR routing protocol based on mobility", World of wireless, International Symposium on Mobile and Multimedia Networks, doi:10.1109IWOWMOM.2006.25 publication, 2006. Kadri, B., Moussaoui, D., Feham, M., "A cross-layer design for QoS impletion in MANETs applied to DSR", 3rd International Conference on Infonnation and Communication Technologies doi:IO.J 109IICTTA.2008.4530206. Smys, s., Bala, GJ., Jennifer, S., "Mobility management in wireless networks using power aware routing", International Conference on Intelligent and Advanced System (ICIAS), doi: 10.11 09/ICIAS.20 J 0.57 J 6234 . Jacob, J., Seethalakshrni, V., Perfonnance Evaluation of Various Routing Protocols in MANET, An International Journal of Engineering Sciences, 5, pp.208-220, 20 J 1. Zhi Ang EU, Winston Khoon Guan seah, "Mitigating Route request Flooding Attacks in Mobile Ad-hoc Networks", International Conference on Infonnation Networking doi: 10.100711 19195 683 3, pp. 327-336,2006. Hu, Y.C., Perrig, A., Johnson, D. B., Ariadne, "A Secure On­Demand Routing Protocol for Ad-hoc Networks", Springer Science, Wireless Networks, voJ.] I , pp. 21-3 8, 2005. Castaneda, R. , Das, S. R., Marina, M. K., "Query Localization Technique for On-demand Routing Protocols in Ad-hoc Networks", Kluwer Academic Publishers, Wireless Networks vol. 8, pp. 137-151,2002.

20J4 Internationai Conference on Issues and Challenges in Intelligent Computing Techniques (ICICT) 219