IEEE International Conference on Advances in Engineering & Technology Research (ICAETR - 2014),

August 01-02, 2014, Dr. Virendra Swarup Group of Institutions, Unnao, India

Improved Energy Efficient Secure Multipath

AODV Routing Protocol for MANET

Hrishabha Raj Jain Sanjay Kumar Sharma

Department of computer science and engineering Department of computer science and engineering
Oriental institute of science and technology Oriental institute of science and technology
Bhopal,(M.P.) INDIA Bhopal,(M.P.) INDIA
hrishabha [email protected]
_ [email protected]

Abstract- In mobile ad hoc network (MANET), energy

consumption is one among the foremost vital restrictions that I. INTRODUCTION
deteriorate the performance of the whole network. Multi-path
routing is better than the single path routing in ad hoc networks, A mobile ad hoc network (MANET) is a collection of wireless
because multi path routing allows the establishment of multiple
path between a single source and single destination node. This
devices moving in seemingly random directions and
paper presents a new approach of energy efficient secure communicating with one another without the aid of an
multipath AODV (EESM-AODV) routing protocol for MANET established infrastructure. Communicating nodes in a Mobile
based on AODV protocol is modified and converted to work on
Ad hoc Network usually seek the help of other intermediate
multiple path .Differing types of routing protocols planned over
the years with minimal management overhead and network
nodes to establish communication channels. Thus, the
resources. AODV is well-liked routing protocol among others. It is communication may be via multiple intermediate nodes from
a loop free, no centralized authority, single path, On-demand source to destination. Topology changes occur unpredictably.
routing protocol and its performance is superior than different
Topology control is needed to determine appropriate topology
routing protocols in MANET. However, single path plan is that
the most vital drawbacks of AODV. Also, it is a heap of routing in adhoc network which saves energy, reduce interference
overhead at the time of route repair and route discovery method. between nodes and extends the lifetime of the network. Based
The mobile nodes in MANET also have restricted resources like
on connectivity, efficiency of energy, robustness to mobility
battery power, restricted information measure that solely single
path protocols cannot handle efficiently. so routing is also a really
and throughput the quality of topology is determined. High
vital issue inside the look of a MANET. Multipath routing permits level routing protocols are implemented over a suitable
the mUltiple methods between one source and sin�le destination topology .All nodes connected in a network must act as routers
node and during this paper conjointly offer security by filtering
to have accurate delivery of data packets. Routes contain links
route discovery process in AODV routing protocol. The proposed
approach contains an adaptive technique in which the our
which is the connection between two nodes. The route quality
proposed scheme is more energy efficient. The filtering is influenced by change in link quality. A varying link route
forwarding scheme slows down the spread of excessive RREQs does not produce good results. The network layer has received
originated by a node per unit time and with success prevents
a notice when working on Mobile Adhoc Network. Therefore
Denial of service attacks. This paper planned multipath extensions
of AODV routing protocol and a security improvement against plenty of routing protocols in such network with different
We compared results of AODV routing protocols in attacked objectives and with different specific needs have been
condition and after applied our algorithm and enhance the results
proposed. [1] . As a matter of fact, the two vital operations at
exploitation multipath scheme for AODV. We used ns-2 to
simulate. Simulation results show that the energy consumption of
the network layer are data forwarding and routing. Data
network is less after applying our proposed algorithm and our forwarding controls how packets are taken from one link and
Network gets more stable and that proposed technique has better put on another. Routing finds out the path which the packet
packet delivery ratio and improved throughput.
must follow to reach the destination from the source. Routing
Keywords: Routing Protocols, Mobile Ad-Hoc Networks Multipath protocols can be divided into single path and muItipath based
Extensions, AODY, Energy Efficient, DOS attack. on the nwnber of routes discovered [3] . Single path protocols
learn routes and select a single best route to reach each
destination. These protocols are unable to balance traffic load
[4] . Single route simplify the routing table and packet

978-1-4799-6393-5/14/$31.00 ©2014 IEEE

 IEEE International Conference on Advances in Engineering & Technology Research (ICAETR - 2014),
August 01-02, 2014, Dr. Virendra Swarup Group of Institutions, Unnao, India

forwarding but it has many disadvantages. Even though section 5 and result analysis in Section 6.and last Section 7
additional network resources may be available, using a single concludes the paper.
path, it is difficult to respond to a large burst in traffic. If the
path fails a new route discovery must be initiated resulting in II. ENERGY AWARE ROUTING IN MOBILE ADHOC
NETWORK
significant delay and packet loss. Multi path protocols learn
routes and select more than one path to reach the destination. Wireless mobile devices are useful if they can be used
They are better for load balancing. Multipath routing improves anywhere. But we have limited battery power of using it,
communication efficiency and promotes Quality of Service by Therefore, in wireless communication; one of the most
utilizing different paths simultaneously [5] . Also they are more challenging problems is power management. Several energy
reliable, robust and consequently reduces control overhead, aware routing protocols have been developed. Most of these
enhances data transmission rate, the network bandwidth is routing protocols aim to minimize the energy consumed per
increased and the energy is saved [6]. Contributions in our packet needed to deliver this packet to its destination. Some of
solution are as follows. the more sophisticated routing algorithms associate a cost with
routing through a node with low power reserve. Other routing
1. A multipath proactive source routing protocol is used as protocols aim to maximize the network lifetime. All previous
each node has complete knowledge of how to route data to protocols are using single path to distribute data traffic through
all nodes in the network at any time. Based on the number the network. The routing protocols, described previously are
of paths available to reach destination, the packets can be
based on the single path routing between a source and a
divided and sent simultaneously to destination.
destination. However, in a reasonably well-connected network,
2. When the data packets are forwarded towards destination
the intermediate nodes can adjust the route information there may exist several paths between a source-destination pair.
carried by them. Furthermore, as these packets are The concept of multipath routing is to give the source node a
forwarded along the new route, such updated information choice at any given time of multiple paths to a particular
is propagated upstream quickly without any additional destination by taking advantage of the connectivity redundancy
overhead. As a result, all upstream nodes learn about the of the underlying network. The multiple paths may be used
new route at a rate much faster than via periodic route
alternately, namely, traffic taking one path at a time, or they
exchanges. Opportunistic data forwarding is taken to
another level by allowing nodes that are not listed as may be used multiple paths simultaneously .Multi-path routing
intermediate forwarders to retransmit data if they believe consists of three components: route discovery, route
certain packets are missing. maintenance, and traffic distribution among multiple paths.
Malicious nodes may become part of actively used routes and
disrupt network operation. In such an environment, malicious A. Route Discovery: It finds multiple routes between a source
intermediate nodes can be a threat to the security of and destination nodes. Multipath routing protocols may be
node disjoint (no common nodes), link disjoint (no common
conversation between mobile nodes. In this paper, we focus on
links), or non-disjoint routes. Non-disjoint routes may have
a special type of denial of service attack due to RREQ flooding lower aggregate resources than disjoint routes for the reason
attack. In this type of attack, those malicious nodes behave like that non-disjoint routes share links and nodes. Disjoint
the normal nodes in all aspects except that they initiate routes provide higher fault-tolerance.
frequent RREQ control packet floods. This type of attack is B. Route Maintenance: It finds and repairs the broken paths.

hard to detect since any normal node with frequently broken C. Traffic Allocation: The traffic allocation strategy is used to
routes could legitimately initiate frequent route discoveries. deal with how the data is distributed amongst the paths.
One or more malicious nodes flooding the MANET with
III. DENIAL OF SERVICE ATTACK DUE TO ROUTE REQUEST
RREQ control packets related to bogus route discoveries can
(RREQ) FLOODING
cause a sharp drop in network throughout. The rest of the paper
The Route Request (RREQ) Flooding Attack is a kind of
is organized as follows. In section 2, we describe Energy
denial-of-service attack, which aims to flood the network with
Aware Routing In Mobile Ad Hoc Networks. In section 3 we
a large number of RREQs to the destinations in the network. In
describe the Denial of services attack caused by RREQ
this attack, the malicious node will generate a huge number of
flooding. In section 4 we describe some related work. An
RREQs, may be hundreds or thousands of RREQs, into the
energy efficient secure filtering mechanism is proposed in
network until the network is saturated with RREQs and unable
to transmit data packets. Many different reactive (on-demand)

978-1-4799-6393-5/14/$31.00 ©2014 IEEE

 IEEE International Conference on Advances in Engineering & Technology Research (ICAETR - 2014),
August 01-02, 2014, Dr. Virendra Swarup Group of Institutions, Unnao, India

dynamic routing protocols proposed for MANETs can suffer network will be improved by suitably reducing the requirement
from this type of attack. On-demand routing protocol, uses a of power for connections. This protocol is the enhanced
route discovery process to obtain a route when a node want to protocol of Ad-hoc On Demand Multipath Distance Vector
send a data packet to a destination for which it does not protocol (AOMDV). This routing protocol is used to avoid the
information about the route. The route discovery works by loop formation in network so that it reduces congestion in the
broadcasting the network with route request (RREQ) control channel.
packets. A node that receives a RREQ rebroadcasts it, unless it
has information about another neighbor, intermediate node or it In [13] , Meng Li proposed a cross-layer optimized energy­
has a route information to the destination indicated in the aware multipath routing protocol (EMRP) for mobile ad hoc
RREQ. If the received RREQ is a duplicate or it has a same networks (MANET). By sharing the information among the
sequence number, it will be dropped. If a node has the route physical layer, the MAC sub-layer and the network layer,
information because it is the destination then it replies to the EMRP efficiently utilizes the network resources such as the
RREQ with a route reply (RREP) packet that is routed back to node energy and the link bandwidth. In this paper, an energy­
the original sender of the RREQ. In an ad hoc wireless network aware multipath routing protocol for mobile ad hoc networks is
energy and traffic load are the two major elements for research, proposed. As a cross-layer design, EMRP utilizes the
the RREQ packets used for route discoveries may consume information from the physical and the MAC layers to select
more bandwidth than the data packets. Malicious nodes could better routes.
exploit this weakness of routing protocols. Attackers can
In [14], M. BHEEMALINGAIAH proposed a routing for
initiate much more REEQ control packets than the normal
MANET. In this paper the author said the Mobile Ad Hoc
nodes to consume network resource. Since control packets are
Networks are essentially suitable when infrastructure is not
given higher priority over data packets in transmitting, In this
present or difficult or costly to setup or when network setup is
situation, valid communication cannot be kept and normal
to be done quickly within a short period.
network nodes cannot be served, then it leads to a type of
denial-of-service attack. In some on-demand protocols, for In [15], S. Harouset.el. proposed a the performance analysis
example AODV, a malicious node can override the restriction of an Energy Aware Multi-path Dynamic Source Routing
put by RREQ_RATELIMIT (limit of initiating / forwarding protocol (EA-MPDSR). This protocol is based on the existing
RREQs) by increasing it or disabling it. A node can do so on-demand Dynamic Source Routing protocol (DSR). It is
because of its self-control over its parameters. The default energy aware and uses a multi -path technique. EA-MPDSR
value for the RREQ_RATELIMIT is 10 as proposed by RFC reduces the energy consumed per received data bytes as well as
3561. A compromised node may choose to set the value of prolongs the network lifetime which leads to improvement in
parameter RREQ_RATELIMIT to a very high number. This the performance of the network.
allows it to flood the network with fake RREQs and leads to a
type of Denial of service attack. In this type of Denial of In [16] , Hesham A. Ali proposed an approach of load
service attack a non-malicious node cannot fairly serve other balancing and multipath for MANET. In this paper they
nodes due to the network load imposed by the fake RREQs. employed a simple test scenario to be sure of proposed model
This will not only lead to the exhaustion of the network efficiency and to validate the proposed Load Balancing Parallel
resources like memory (routing table entries), but also lead to Routing Protocol. Load Balancing Parallel Routing Protocol
the wastage of bandwidth and the wastage of nodes' processing get load balancing, decreasing the end-to-end delay and
time. increasing the packet delivery ratio and throughput of the
network, therefore the performance of multi-path routing
IV. RELATED WORK protocols can be improved consequently.
Many research has been conducted on the energy efficiency
and multipath routing in MANET but still current result are not V. PROPOSED WORK

appropriated for MANET. AODV protocol is modified and converted to work on multiple
paths to send data .First we modified route discovery process
In [12] , S. Santhiet. al. proposed a power aware multipath of AODV when a destination receives a RREQ packed it
routing protocol. In this author told that Energy consumption in replies for all the RREQ packets received even for same
heterogeneous network is a major problem. The lifetime of

978-1-4799-6393-5/14/$31.00 ©2014 IEEE

 IEEE International Conference on Advances in Engineering & Technology Research (ICAETR - 2014),
August 01-02, 2014, Dr. Virendra Swarup Group of Institutions, Unnao, India

sequence number using back path. At the sender end sender ongmator during a preset time period. Whenever a RREQ
takes all the paths and short them on the bases of hope count, at packet is received, a check is performed. If the rate of this
the time of data forwarding sender uses best three paths among RREQ originator is below the RATE_LIMIT, the RREQ
all paths . packet is processed as normal.

This process also eliminates problem of route brakeage in The BLACKLIST_LIMIT parameter is used to specify a
AODV .When one route breaks source can send the data using value that aids in determining whether a node is acting
different path available. malicious or not. If the number of RREQs originated by a node
per unit time exceeds the value of BLACKLIST_LIMIT, one
can safely assume that the corresponding node is trying to
flood the network with possibly fakeRREQs. On identifying a
0 0 0 0 0 0 0 0
0 0 sender node as malicious, it will be blacklisted. This will
0 0 0 prevent further flooding of the fake RREQs in the network.
The blacklisted node is ignored for a period of time given by
S BLACKLIST TIMEOUT after which it is unblocked. The
0 0

0
proposed scheme has the ability to block a node till
0 0 0
BLACKLIST_TIMEOUT period on an incremental basis. By
(a) Under light data traffic load (b)Under heavy data traffic load
blacklisting a malicious node, all neighbors of the malicious
Fig.1 AODV routing under (a) Light data traffic load (b) Heavy data traffic node restrict the RREQ flooding. Also the malicious node is
load isolated due to this distributed defense and so cannot hog its
neighbor' s resources. The filter process is shown in figure 3.

The proposed method is more energy efficient because it

divides the load of packet forwarding to all the nodes available
in multiple paths .As i already describes P, Q and R
intermediate nodes will not busy most of the time and there [ recei ve a RREQ
energy will save because there load is shared by the other path
as A, B, C and D or X, Y and Z. We propose a simple,
distributed, and adaptive technique to automatically control the
spread of RREQ packets and reduce the effects of broadcast
attacks using RREQ. We assume that there exists a security
mechanism, such as public key cryptography and digital
signatures or MAC (Message Authentication Code) that y
enables a node to authenticate routing messages from any node
in the network. Therefore, a malicious node cannot spoof the
Process as OOfmll
originator and destination IP addresses in a RREQ packet
although the destination IP address may not be reachable in the
network. The proposed technique uses a filter to detect
misbehaving nodes and reduces their impact on network
performance. The aim of the filter is to limit the rate of RREQ
packets. Each node maintains two threshold values. The
threshold values are the criterion for each node' s decision of
how to react to a RREQ 1message.

The RATE_LIMIT parameter denotes the number ofRREQs

FigJ Processing of the received RREQ
that can be accepted and processed as normal per unit time by a
node. Each node monitors the route requests it receives and
The neighboring nodes of the malicious node are therefore free
maintains a count of RREQs received for each RREQ
to entertain the RREQs from other genuine nodes. In this way

978-1-4799-6393-5/14/$31.00 ©2014 IEEE

 IEEE International Conference on Advances in Engineering & Technology Research (ICAETR - 2014),
August 01-02, 2014, Dr. Virendra Swarup Group of Institutions, Unnao, India

genuine nodes are saved from experiencing the Denial of If table contain more than three entries than retrieve
services attack. If the rate of RREQs originated by a node is fIrst three otherwise take all .
between the RATE_LIMIT and the BLACKLIST_LIMIT, the
Step 5: (At the time of sending data)
RREQ packet is added to a "delay queue" waiting to be
processed. Every time a DELAY_TIMEOUT expires, if there if data available to send is big than divide it in parts
is anything in the delay queue (RREQ packet waiting to be and send through available paths
processed), then the fIrst packet is removed to be processed. To
else send data using fIrst path(shortest) in table .
do so, malicious node that has a high attack rate will thus be
severely delayed. Meanwhile, the proposed rate control Step 6: If a path breaks send the data using other available
mechanism will have no impact on other nodes and also have paths .
minimal impact on the normal nodes that send abnormally high
RREQs. VI. SIMULATION AND EVALUATION

The fIltering forwarding scheme slows down the spread of According to Shannon, simulation is the process of creating a
excessive RREQs originated by a node per unit time and model of a real system and conducting experiments with this
successfully prevents DoS attacks. The proposed scheme system model for the purpose of learning the behavior of the
incurs no extra overhead, as it makes minimum modifIcations system or evaluating various strategies for the operation of the
to the existing data structures and functions related to system. With the vibrant nature of computer networks, we thus
actually deal with a energetic model of a real dynamic system.
blacklisting a node in the existing version of pure AODV. Also
the proposed scheme is more efficient in terms of resource
A. Simulation tool
reservations and its computational complexity. In addition to
limiting the clogging up of resources in the network, the NS-2 is an open-source event-driven simulator designed
proposed scheme also isolates the malicious node. especially for research in computer communication networks.
Since its inception in 1989, NS-2 has always gained
Algoritbm: tremendous interest from industry, Academia and various
government and private organization. Having been
DSN - Destination Sequence Number, NID - Node ID,
investigation and enhancement for years, To investigate
Hop_count. network model performance, researchers can simply use an
easy-to-use scripting language to confIgure a network, and
Step 1: (Initialization Process)
observe results generated by NS-2 Undoubtedly NS-2 has
Retrieve the current time become the most widely used open source network simulator.
Here we have selected NS-2.35 for our simulation.
Add the current time with waiting time
Step2: ModifIed route discovery process of AODV In order to evaluate the performance of our proposed Scheme
and compare with Ad hoc On-Demand Distance Vector
when a destination receives a RREQ packed it replies for all
(AODV) Routing [2] and AODV under attack .we simulate the
the RREQ packets received even for same sequence number proposed mechanisms using NS-2.35 Simulator. The
simulation environment, performance metrics and results are
using back path.
discussed in the subsequent sections.
Step 3: (Storing Process)
In our simulations bandwidth set to 1 Mbps and the
Store all the Route Replies Destination Sequence
transmission range is set to 250 m. The evaluations are
Number and Node ID in Route Routing Table conducted with a total of 20 nodes that are randomly
distributed in an area of 1100m x 1100m. We use Random
Do again the above process until the time exceeds
Waypoint to model node mobility. The size of each Constant
Step 4: (Identify top 1 to 3 shortest path on the bases of hop Bit Rate (CBR) packet is 1024 bytes and packets are generated
at the fIxed interval rate of 5 packets per second.
count) .
Short the table on the bases of Hop_count.

978-1-4799-6393-5/14/$31.00 ©2014 IEEE

 IEEE International Conference on Advances in Engineering & Technology Research (ICAETR - 2014),
August 01-02, 2014, Dr. Virendra Swarup Group of Institutions, Unnao, India

B. Simulation parameter D. Result Analysis

Channel type Wireless channel This result shows the working of AODV under attacked
condition, here show three results. Graph 6.1 A shows the
Radio propagation model Propagation Two ray packet delivery ratio. Graph 6.1B is showing throughput of
ground network and graph 6.1C showing energy graph of node present
in my simulation environment. Here it is clearly shown that
Network interface type Phy/wirelessphy performance is degraded in under attack condition.

Mac type Mac 802 11 -

xgraph
I Close II Hdcpyll Aboull X Graph
Interface queue form Queue/droptail/priqueue Packet_Delivery_Ratio

on

Link layer type LL lit,

RS

Antenna model AntennalOmni antena Rn ��� �"'" ......,

II'
7S
Max.packets in infq. 50
7n

Routing protocol AODV eo;

X dimension topology 1100 en II

"" II
Y dimension topology 1100
sn

Model of simulation Energy model 4S II

II
0.0000 20.0000 40.0000 60.0000
Initial energy in joules 100

6.1A
Time of simulation 75.0sec.
xg.-aph 1£1
I Close II HdcpyllAbmtl X Graph
Number_oCPackets x 103
Table 1: simulation Parameters
o
30000 ..·--...... ·.. ·.. · /'. . ./"'�
C. Performance Metrics 2.8000_
2.6000-·.. ...... .. ,V
. .. .. ,/rl �II'�I\ �/+v.. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. , .. .. .. .. .. .. .. .. .. ..\.. .."".. .. .. .. .. .. ,.. .. .. .. .. .. .... .... .... .. .. .. .. .. .. .. ..
. .. .. .. .. ..
.. ..

Three metrics were used to evaluate AODV protocol and

proposed EESM-AODV protocol.
: I
2.0000· --· K ..··....···....···....···....···....··· ..+ ..···....···....··....... ....... ....... + ........................................................................ + ....................................................

I)Packet Delivery percentage: The ratio of the data packets

1.4000._
succeed to deliver at the destinations to those generated by the .. .. ...... ...... ...... ...... ........ .. .... .. .... + .. .. .... .. .... .. .. .. .... .. .... .. .... .. .... +...................... ...... ................................................ , .. .. ...... ...... .. .. .. .. ...... .... ........ ......
1.2000)�1
CBR sources.
08
. 000 -· 1 .. .. .... .. .. .... .. .. .... .. .. .... .. .. .. .. .. .. .... .. .. .. ...... .. .. .. + ...... .. .. ...... .. .. ...... .. ...... .. .. ...... .. .. ...... .. .. ...... .. .. .. +.................................... ........................................ , ...... .. ...... .. .. .... .. .. .. .. .. .. .... .. .. .... .. .. .. .. .. .... .. .. ....
..

2) Total Throughout: The total number of application layer 06 1H .... .. .. .... .. .. .... .. .. .... .. .. .... .. .. .... .. .. .. .... .. .. .. .... .. .. . .... + ................. ...... .... ...... .... .......... + ........ ............................................................., .. .... .. .. .... .. .. .... .. .. ...... .. .. .... .. .. .... .. .. .... .. .. .... .. .. .. ..
. 000 -·
O.4ooo -l · H .... ........ ........ ........ ....... ....... ....... ..... .. ..... .. + ......................................... + ........................................................................ + ........................................................
data bits successfully transmitted in the network per second. 0.2000· -] 1 + ..··....···....··· ....··· ....···....···.. + ..... ................................ .................... + ............................................... !.................................................................
.. ..

0.0000 I� �
Time in Seconds
3)Energy levels : This shows energy at the time of simulation 0.000 0 20.0000 40.000 0 60.0000
starting and ending of simulation .
6.1B

978-1-4799-6393-5/14/$31.00 ©2014 IEEE

 IEEE International Conference on Advances in Engineering & Technology Research (ICAETR - 2014),
August 01-02, 2014, Dr. Virendra Swarup Group of Institutions, Unnao, India

In the previous chapters we have seen packet delivery ration packet delivery ratio. Graph 6.2B is showing throughput of
and throughput of AODV in under attack condition. In this network and graph 6.2 C showing energy graph of node present
section we are presenting the energy consumption by both in my simulation environment. It is visible that our proposed
condition in under attack condition and after applied algorithm
algorithm has better performance then AODV under attack.
condition and comparing energy consumption.
Energy utilization after applied Algorithm
Energy Consumption of nodes in AODV When attacking
condition AODV is run the energy consumption is more in
After practical this algorithm in AODV we minimize our
attacking condition case. There is no method for detection of
Energy consumption of each node. The energy of each node
unknown node. And the number of RREQ Packets send is
after this algorithm is given below with the help of energy
more so the energy consumption is more in attacking condition
graph.
of AODV .We plot a energy graph for attacking condition
AODV after simulation. Each node have 100 joules of initial
Energy of Node 10 56.125695 Joule. =

energy and after simulation the energy graph of each node is

Energy of Node 14 75.473575 Joule. =

given below:
Energy of Node 16 51.593463 Joule. =

Energy of Node 18 65.381331 Joule. =

Energy of Node 10 54.793256 Joule. =

Energy of Node 19 66.84058 Joule =

Energy of Node 14 68.086466 Joule. =

Average Energy of Nodes Total Energy of all No. of Nodes
=

Energy of Node 16 52.839517 Joule. =

Average Energy of Nodes =

Energy of Node 18 63.868596 Joule. =

56.125695+75.473575+51.593463+ 65.381331+66.84058
Energy of Node 19 64.603259 Joule. =
Average Energy of Nodes 63.0829286 Joules =

Average Energy Consumption Total Energy of all =

No. of Nodes
Energy graph is shown below
Average Energy Consumption =

54.793256+68.086466+52.839517+63.868596+64.603259
ENERGY GRAPH AFTER APPLIED ALGORITHM
Average Energy of Nodes 60.8382188 Joules. =

xgraph
ENRGY GRAPH

xgraph

l00.0000···-�/,·· ····-· ·I--··-··-·

· ·I··-·- ·- �·-··-··-·
· i ·-.•.••-.� · ········· ·+--····- 1

95.0000--...

00.0000

85.0000···,,· ················ +·················· 1

00.0000

75.0000- ._--"- - - -_.

70.0000··_ ..•..•...•...•..•.•...•...•....•...•... t ··············· · 1 · ········· .+ ............"k............._.+..'''':........ + . ..........

65.0000 __,'.N.

60.0000- ---- .-
O.CKXlO 10.0000 2O.QCKX) 3O.(XX)Q 40.0000 50.0000 60.0000 70.0000

6.2C

After applied this algorithm our network energy consumption

6.1C is less as compared to prior one. Because the average energy
consumption of each node after applied this algorithm is less.
This result shows the working of AODV after applied our And our network is more energy stable.
mechanism. Here shows three results. Graph 6.2A shows the

978-1-4799-6393-5/14/$31.00 ©2014 IEEE

 IEEE International Conference on Advances in Engineering & Technology Research (ICAETR - 2014),
August 01-02, 2014, Dr. Virendra Swarup Group of Institutions, Unnao, India

View Go Bookmarks Help xg.-aph

��� X Graph
xgraph Packet_Deliverv _Ratio

n IClosellHdcpyliAboutl PDR Result

B Packet_Delivery_Ratio 100.0000

pdrgraph
00.01000 r······································ ,....................................... t······································· , ..............................................

65.0000'"

60.0000 ..· ..· II ....·..···· , ........................; ........... ; ......................., ........... t··....·....·..···..··..; ·....·..··· ; .......................
55.0000·····

50.0000 ..· ..· It ....·..···· ; ........................; ..........

TlmeJn_Second�
0.0000 60.0000 70.0000
0.0000 20.0000 40.0000 60.0000

6.3A
6.2A

-. xgraph

�t3 ���
I,
�
IClosellHdcpyllAbolll
Number_of_Packets x nil
xgraph

X Graph
Number_of]ackehx 103
Throughput comparision
attackedth

recDveredth

II 4.0000·····

4.0000-- ;tw ��
3.5000"'"

3.5000-·- I V \ � 3.0000"'"

1 , .......... ....... . . .......... .........

2.5000

2.5000---
5
2.0000 · -liV··········· , ......................, ............ t·······················; ·················

2.0000-
1.5000···..

1.5000-
1.0000···..
,
1.0000j

05000- il
0.0000

0.0000 ;,= = 1= �
Time-in-Seconds
0.0000 20.0000 40.0000 60.0000
6.3B

6.2B

The below graph shows the comparative study of EESM­

AODV, attacked AODV and after applied algorithm AODV ,
red lines is showing AODV under attacked condition and green
lines is showing our proposed mechanism EESM-AODV. See
fig 6.3A and fig.6.3B

978-1-4799-6393-5/14/$31.00 ©2014 IEEE

 IEEE International Conference on Advances in Engineering & Technology Research (ICAETR - 2014),
August 01-02, 2014, Dr. Virendra Swarup Group of Institutions, Unnao, India

PARAMETER Attacked Recovered Improve [4] Online]http://www.inetdaemon.com/tutorials/internet/ip/routing/single_

vs_multi.shtml.
condition condition performance
[5] KoojanaKuladinithi, Carmelita Gorg and Samir Das, -Radio Disjoint
Multi-Path Routing in MANETII, in CEWIT 2005 Conference, 2005, pp
SEND 57078 59008 1930 1-2.
[6] Ali AbdallaEtorban, -The Design and Performance Evaluation of a
Proactive Multipath Routing Protocol for Mobile Ad Hoc Networksll , at
RECV 45793 53127 7334 Heriot - Watt University in the School of Mathematical and Computer
sciences, May 2012.
PDF 80.23 90.17 9.94 [7] W. Yu and J. Lee, "DSR-based energy-aware routing protocols in ad
hoc networks," Proceedings of the International Conference on Wireless
Networks (lCWN), Las Vegas, Nevada, June 2002.
Throughput 2.65 3.52 0.87 [8] J. Broch, D. Maltz, D. Johnson, Y -C. Hu, J. Jetcheva, "A performance
comparison of multi-hop wireless ad hoc network routing protocol", The
Actual 87.24% 95.29% 8.05% 4th Annual ACMlIEEE International Conference on Mobile Computing
and Networking (MobiCom'98), pp. 85-97, Dallas, TX, October 1998.
Performance
[9] A., c.-c. Chiang, Iwata, G. Pei, M. Gerla and T.-W. Chen, "Scalable
Routing Strategies for Ad Hoc Wireless Networks", JSAC99, IEEE
TABLE 2: ANALYSIS OF OVERALL RESULTS Journal on Selected Areas in Communications, Vol. 17, No. 8, pp. 1369-
1379, August 1999.
VII. CONCLUS]ONS [ 10] T. Clausen, P. Jacquet, A. Laouiti, P. Muhlethaler, A. Qayyum and L.
In this paper we have proposed energy efficient secure Viennot, "Optimized link state routing protocol for ad hoc networks",
IEEE INMIC 2001, pp. 62 - 68, Dec. 2001.
multipath AODV routing protocol for mobile ad hoc network
[ 1 1] D. B. Johnson and D. A. Maltz. "Dynamic source routing in ad hoc
and fount that the proposed scheme is more secured and has
wireless networks". In Mobile Computing, edited by Tomasz Imielinski
better results in compared the previous techniques. In this we and Hank Korth, chapter 5, pp. 153-181. Kluwer Academic Publishers,
simulate under attacking condition AODV and find the route 1996.
request send in each session. We also calculate the energy [ 12] S. Santhi, Dr. G. SudhaSadasivam "POWER AWARE QOS
consumption of each node in this we calculate the average MULTIPATH ROUTING PROTOCOL FOR DISASTER RECOVERY
NETWORKS" International Journal of Wireless & Mobile Networks
energy consumption of each node and generate the energy
(IJWMN) Vol. 3, No. 6, December 20 1 1.
Graph. After this we apply our proposed algorithm and then
[ 13] Meng Li, Lin Zhang, Victor O. K. Li, Xiuming Shan, Yong Ren "An
again calculate the energy consumption of each participating Energy-Aware MuItipath Routing Protocol for Mobile Ad Hoc
node and calculate the average energy of consumption of each Networks" ACM Sigcomm Asia 'OS, Apr. 10-12, 2005
node. Here we find the discard limit of each node on the basis [ 14] M. Bheemalingaiah. M. M. Naidu. D. SreenivasaRao, G.Varaprasad
of route request send in each session. After applying this "ENERGY AWARE NODE DISJOINT MULTIPATH ROUTING IN
MOBILE AD HOC NETWORK"Journal of Theoretical and Applied
algorithm the average energy consumption is less as compared
Information Technology 2005-2009.
to earlier and our network is more energy stable. Average
[ 15] S. Harous, M. A1dubai, Q. Nasir "Performance Analysis of An Energy
energy of participating node is 60.8382188 joules before Aware Multipath Routing Algorithm for Mobile Ad Hoc Networks"
applying the algorithm. The energy of all nodes after applying University of Sharjah.
the algorithm is 63.0829286 Joules when we apply our [ 16] Hesham A. Ali, Taher T. Hamza and ShadiaSarhan "Manet Load
proposed algorithm. So the energy consumption of the network Balancing Parallel Routing Protocol" IJCSI International Journal of
is reduced with our proposed algorithm, & our Network gets Computer Science Issues, Vol. 9, Issue 4, No 1, July 2012.

more stable. This multipath technique can also we use to [ 17] Sheela Rani Arasu, Immanuel JohnrajaJebadurai " Multipath Routing
for Opportunistic Data Transfer in Mobile Adhoc Network" IJERA Vol.
enhance DSR, DSDV, and other routing protocols .We can say 3, Issue 2, March -April 2013, pp.1567-1573.
that the use of this technique can also make DSR and other
protocols like this more energy efficient and achieve better
PDR and throughput.

REFERENCES
[ 1] Rajaraman, R. -Topology Control and Routing in Ad hoc Networks:A
Survey,1I SIGA News, vol. 33, pp. 60-73, June 2002.
[2] Sheela Rani Arasu, and Immanuel JohnrajaJebadurai, -Analysis of
Different Routing Techniques for Opportunistic Data Transfer,
International Journal of Computer Applications (0975 - 8887), volume
62 - No.5, January 2013.
[3] Ash Mohammad Abbas, and BijendraNath Jain, -Path diminution in
node - disjoint multipath routing for mobile ad hoc networks is
unavoidable with single route discoveryll ,International Journal Ad Hoc
and Ubiquitous Computing.

978-1-4799-6393-5/14/$31.00 ©2014 IEEE