IRJET- Energy Efficient Reactive Routing to Enhance QOS of Manets

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 12 | Dec 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.34 | ISO 9001:2008 Certified Journal | Page 850
Energy Efficient Reactive Routing to Enhance QOS of MANETs
Amulya M.A1, Hemanth Kumar A.R2
1M.Tech student, Dept. of ECE, Bangalore Institute of Technology, Bengaluru, India-560004
2Prof, Dept. of ECE, Bangalore Institute of Technology, Bengaluru, India-560004
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract - MANETS are the nodes, which are dynamic in
nature and keep moving from location to location. When the
nodes moves, the network topology changes and the link
breaks between the nodes and the new link are formed. To
deliver high ratio of data packets we propose energy efficient
multipath routing scheme, which is an adaptation of AODV
routing protocol to improve network lifetime. The proposed
system determines multiple paths from source to destination
by selecting nodes having high residual energy and also
controls congestion and delivers high throughput. The
proposed system is simulated in the NS2 simulator and the
network parameters like, throughput, delay, packet delivery
ratio, energy consumption and overhead is analysed
Key Words: MobileAd-hoc Networks(MANETs),Reactive
Routing Protocol, Ad-hoc On Demand Distance Vector
Routing (AODV), Energy Efficient Multipath Ad-hoc On
Demand Distance Vector Routing (EMAODV);
1. INTRODUCTION
A Wireless Sensor Network (WSN) consists of hundreds or
thousands of low cost nodes whichincludesheapsandheaps
of low value points which can be deployed fixed position or
randomly communicating nodes to monitor the
surroundings. Deploying big quantity small nodes are
included in the WSNs which are trend from past few years.
The nodes then experience environmental modificationsand
document these different nodes over the network design.
The sensor network contains several lots of nodes and
contains controlled resources. Topologyadjustmentsoccurs
regularly due to the nodes disasters, it specifically uses
broadcast communication. The huge environmental sensor
nodes are great for deploying consists of theself-powerunit,
Sensing unit, Processing unit, memory, self and remote
testing unit, sync and time unit, transmitter and receiver,
routing tables and also security units. The nodes in the
system are spread in the network.
Mobile ad hoc networks (MANETs) consists of mobile nodes
which exchange information dynamically among them over
wireless links. MANETs are an infrastructure-less,
decentralized network which consists of a series of mobile
wireless nodes that communicate with each other without
any centralized authority being used. By its basic
characteristics, such as wireless media, dynamic topology,
distributed cooperation, MANETs is vulnerable to various
kinds of security. The most important element of MANET is
Routing Protocols which are needed to handle dynamic
communication and also find route so as to deliver data
packets to the correct destination. [1]. The nodes serve as
routers to deliver data to their destinations from sources.
The routers have to work together efficiently in order to
achieve performance requirements of this kind of network.
Routing algorithms moreover play a key role in the design
for MANET's architecture, so they need presents tests and
evaluation of reliability for two types of flat routing
protocols: proactive and reactive routing protocols. [2].
Routing information of each node propagates through
network by certain routing protocols:ReactiveRouting,Pro-
active Routing and Hybrid Routing are the three types of
routing protocol.
Reactive routing protocol isanondemand routingprotocols,
when a source node wants to communicate with the
destination node then only the route will get searched
otherwise, there is no pre-defined routes between any
nodes. Updated routing information will not be kept in the
nodes when no communication is active. Ad-hoc on demand
distance vector (AODV) and Dynamic Source Routing (DSR)
are the reactive routing algorithms. Pro-active routing are
table driven, they keeps updatedroutinginformationofeach
nodes will be stored in the routing table. Destination
Sequenced Distance Vector (DSDV) Routing and Fisheye
State Routing (FSR) are pro-active routing protocols.Hybrid
Routing Protocols consists of combined features of reactive
and pro-active Routing protocols. [3]
In the wireless sensor network as the node moves, the
network topology will change this leads to the breaks
between the nodes. Due to dynamic characteristics of nodes
links are unstable and lossy in such environments. To
guarantee QoS parameters it is essential to estimate the link
quality among nodes to increase the performance of routing
protocol. Single path between source and destination was
the principle of routing protocol and it was limited with
minimum hop counts. In case of route fail, new route is
evaluated this led to high communication cost and more
resource consumption. Multipath routing provides key
solution under various conditions to achieve QoS
requirements. QoS based protocols have to find a trade-off
between energy consumption and the quality of service.The
network has to balance between energy consumption and
data quality. In particular, the network has to satisfy certain
QoS metrics, e.g., delay, energy, bandwidth, etc. when
delivering data to the BS.
This paper proposes anEnergyEfficientMultipathAd-hocon
demand distance Vector Routing. It specifies the level of
energy a node has in the network. The node owns the initial
energy value of 100 joules, which is referred as the energy
level the node has at the start of the simulation. This energy
level is called initial energy. The node reachescertain energy

level after transmitting and receiving the data, hence that
energy level is referred as tx-Power and rx-Power.Inthis we
have considered energy module which assign nodes initial
energy, by calculating the consumed energy and residual
energy of individual nodes is carried out. By calculating the
residual energy of the each node the path which is more
energy efficient can be found. So by this the network
resources can be utilized effectively. The calculation of
energy of nodes is very important as the nodes lifetime has
to be increased, for prolong network lifetime, many
optimization techniques have to be used to optimize and
monitoring of the nodes.
Simulations were carried out on NS2 tocomparetheexisting
Reactive protocol to the proposed EMAODV basedonpacket
loss, packet delivery ratio, end to end delay throughput and
energy. The results shows that proposed protocol works
better in five aspects.
The organization of this paper is as follows. Section 2
discusses about related work. Section 3 describes the
proposed EMAODV with energy module. Section 4 presents
simulation techniques and result justifications, finally,
section 5 presents concluding remarks.
2. RELATED WORK AND PROBLEM STATEMENT
2.1 Related work
MANET routing protocols are based on the premise that all
nodes comply without the routingprotocolbeingmaliciously
disrupted. MANET is affected by a large number of attack
forms of varyingseverity. Dynamic Source Routing(DSR)isa
well-known MANET reactive routing protocol, path finding
and maintenance are the two important phases of this
protocol. It can send the packets to the destination by
choosing alternative route at any time since, the nodes keep
the multiple routesinformation.Whenthesourcenodewants
to communicate with the destination it will send the route
request packet which is RREQ, it consists of source,
destination address and ID number. After receiving route
request packet the sink or intermediate nodes generates
replay message whichisroutereplaypacketinthesamepath.
DSR does not support routing messages protection. [5].
In AODV all the nodes in the network need to keep a routing
table that stores routing information of its neighbor nodes.
Node sequence number and a broadcast-id will be
maintained for each of the nodes. When a source node wants
to communicate with destination node, it first increments its
broadcast-id and broadcast a request packet to its neighbors
to set up the route.
In this protocols route discovery is done by blindly
forwarding route request packets from source to all its
neighbor nodes in the network.Then,theneighbornodeswill
receive and process the information. Nodes which receives
the request packet checks there routing tableforthepossible
route. If there is a route, the nodes send out RREP packet to
its neighbors. Nodes silently discards the RREQ packet that
do not arrives first. The other nodeatthedestination,itsends
out a route replay error packet (RREP). It blindly floods the
entire network with route request packets for route
discovery, congestion builds up in the network.
2.2 Problem Statement
Existing system schemes deals with the alternates routes
methods, which was notenergyefficientandtimeconsuming.
The existing systems update the node position using
localization algorithm and shortest path, which does not
relate to energy efficiency. The problem of this is to find
proper positions forthe sensorsatanypointintimesuchthat
some desired specifications are satisfied, while the total
energy consumption of the sensors is minimized. When a
source node wants to communicate with the destination it
should find a path which is efficient, the nodes which are
involved in the path should have maximum energy which is
sufficient forthe transmission and receptionofpackets,soby
finding the residual energy of a nodes efficient path can be
discovered, which is energyefficientpath.Sowhichyieldsthe
prolong increase of network lifetime.
3. IMPLEMENTATION
3.1 Existing System
The existing system proposes an Efficient Multipath AODV
(EMAODV) which decides whether a node in a network is
forwarding ormutetosenddatapacketsfromsourcethrough
destination in the context of route discovery. This could be
achieved by using the Time to Live (TTL) value, that decides
how many hops a routerequest can go through,aswellasthe
Predecessor address by adding an extra field.
EMAODV defines a route discovery path while using the
factor Time to Live (TTL). TTL is a component deciding the
hop distance until it is possible to transfer a RREQ packet.
The TTL value throughout the RREQ packet is initially set by
the origin node to a certain initial value INITIAL TTL and the
RREQ packets being distributed over a hop array equal to
TTL cost. If the source does not obtain a RREP packet from a
destination within in the discovery duration of the path, the
destination is not withininitialhopdistance.Thesourcenode
instead increases the TTL value to improve its lookup range
and transmitted in the new range the RREQ packets.

3.2 Proposed Architecture and Methodology
The below fig-1 shows the architecture of the proposed
system,
Fig-1: Architecture of proposed system
Sensors are deployed and it is communicated. When it is
communicating with nodes the routing protocols like AODV
and EAODV is analysed and performance of the routing
protocols by considering network parameters is verified for
better results.
This design scheme is iterative and requires attention of
various options at each stage. Several assumptions regarding
the hardware required and the operating surroundings of the
systeminfluencedesignselections.Theassumptionshavebeen
made after giant consultation with the cease user and are
greater or muchless affordable.Thegadgetmightbeappliedat
the Ubuntu working gadget, using C++ and TCL Script.
Fig-2: Data flow model for performance analysis
Fig-2 Describes the data flow of the model for the
performance analysis, in which the nodes are randomly
deployed and routing of packets using different routing
protocols are used toanalysetheefficienttransmissionandto
evaluatethe parameterslikethroughput,delay,overheadand
packet drop ratio. The network topology design gives the
information about how the nodes are deployed, what is its
configuration. It consistsofthenetworkparameterslike,MAC
protocol, Network area, No of nodes, Propagation Model,
Queue type used.
3.3 ENERGY MODULE
In this energy module, the nodes are assigned with initial
energy and the nodes are homogeneous. The energy of the
node is calculated as, at the start time of simulation all the
nodes has higher initial energy, as the nodes moves
randomly the nodes energy decreases. Each node consume
energy while transmitting and as well as receiving data
which is represented as t-x and r-x power. Forsavingenergy
the parameter like sleep power, idle power and transition
power are assigned for awake and sleep modes. These
modes will be automatically be shifting to awake when the
Mobile nodes deployment
Routing Protocols
Check for Throughput
AODV, EAODV
Yes
No
Check for Delay
Check for overhead, pdf
and energy

node has data transmission and goes to sleep after
transmission. The residual energy is defined as remaining
left out energy after data transmission. The residual energy
should be balanced amongnodestoextendnetworklifetime.
The residual energy is formulated as
RE = IE (initial energy)- CE (consumed energy)
Fig-3: Energy calculation Algorithm for Nodes
4. RESULT AND PERFORMANCE ANALYSIS
The Network Simulator-2 has been used to carry out the
simulation, parameters which are used to set up the
simulation environment is as shown in the Table-1.
Table-1. Parameters
Parameters Value
Operating system Ubuntu 11.10
Simulator NS2 (ns-2.35)
Channel type Wireless channel
Number of nodes 20,40,60,80
MAC protocol 802.11
Data packet size 512
Simulation area 1000*1000
Radio propagation model Propagation model/Two
Ray ground
Routing protocols AODV, EMAODV
These parameters setup for simulation have been kept same
for different number of nodes. i.e. 20, 40, 60, 80, and 100.
Few parameters are considered in this project whichyieldin
estimating the system performance and also which provides
proof for network lifetime improvement. Based on five
performance metrics, the network quality of services (QOS)
is checked using proposed energy efficient protocol.
4.1 Simulation Results
We compare the performance of proposed system with an
existing system. In Existing system alternate routesmethods
are used, which was not energyefficientandtimeconsuming.
The existing systems update the node position using
localization algorithm and shortest path, which does not
relate to energy efficiency. Where in proposed system we
provides alternative path with high residual energy. It
controls congestion in the network and performances like
throughput, delay, packet delivery ratio, overhead, and
energy of existing and proposed systems are compared, this
also checks performance for varying network size. The
simulation is carried out in network simulator 2, NAM
window provides the graphical representation of the
simulation. The result and performance analysis are
discussed below, the red line indicates the existing system
and green color indicates the proposed system.
1) Throughput
The number of packets sent and received per unit time. It is
expressed in kbps. The throughput values of the proposed
system and existing system is given in Table-2
Table-2. Throughput
Number
of Nodes
Existing system
(MAODV)
Proposed system
(EMAODV)
20 23.60 35.30
40 35.99 37.74
60 36.84 38.03
80 37.56 37.95

Chart-4: Throughput w.r.t existing and proposed system
The above graph shows the throughput of existing system
v/s proposed system. For the proposed system the X axis
indicates the number of nodes and the Y axis indicates
Throughput. As the node number varies, the number of bits
processing time per seconds increases in existing system.
Thus delivering less throughput inexistingsystem.Butin the
proposed system, due to high frequency cycles and data
processing rate, delivers high throughput.
2) End to End Delay
End to end delay can be calculated by sent time of packet by
source – received time of packet by destination. It is
expressed in milli seconds (ms).The belowtable-3andgraph
shows the end to end delay of existing system v/s proposed
system. For the proposed system the X axis indicates the
number of nodes and the Y axis indicates delay in terms of
seconds.
Table-3. End to End delay
Number of
Nodes
Existing system
(MAODV)
Proposed system
(EMAODV)
20 725.393 168.182
40 162.293 43.489
60 49.44 14.62
80 35.179 128.255
Chart-5: End to End Delay w.r.t existing system and
proposed system
As the number of nodes varies and due to nodes mobility,
packet arrival time increases in existing system which
reflects the delay of existing system to increase. Due to high
data processing rates, packet sends time and arrival timeare
less in proposed system.
3) Overhead
It is the number of routing packet processed in a network.
The overhead values and simulation results of existing and
proposed system are given below.
Table-4. Overhead
Number of
nodes
Existing system
(MAODV)
Proposed
system
(EMAODV)
20 6.086 4.122
40 8.019 4.736
60 6.869 4.896
80 8.305 6.312

Chart-6: Overhead w.r.t proposed system and existing
system
The fig-6 shows the overhead of existing system v/s
proposed system. For the proposed system the X axis
indicates the number of nodes and the Y axis indicates
overhead in terms of load. As the number of nodes varies
overhead of the proposed system increases due to
computational function. In the proposed system the
overhead is low due to increased processing rates.
4) Packet Delivery Ratio
It is the measure of ratio of numberofpacketstransmittedby
source and the number of packets acknowledge by
destination.
Table-5. Packet delivery ratio
Number of nodes Existing system
(MAODV)
Proposed system
(EMAODV)
20 0.6014 0.8
40 0.9289 0.9707
60 0.9595 0.9709
80 0.9562 0.9792
Chart-7: Packet Delivery Ratio w.r.t existing system and
proposed system
The graph shows the PDR of existing system v/s proposed
system. For the proposed system the X axis indicates the
number of nodes and the Y axis indicates PDR. In the existing
system due to large data rates packet processing takes more
time and as the node moves apart the link between two
nodes is not stable enough to deliver the packets. In
proposed system enhanced data rate increases PDR and
achieves more reliability in delivering the packets.
5) Energy
The sensor nodes are battery operated so, the nodes require
some energy for communication. Below table simulation
result shows energy of existing system v/s proposedsystem.
Table-6. Energy consumption
Number
of nodes
Proposed system
(MAODV)
Existing system
(EMAODV)
20 21.7 12.3
40 31.2 10.1
60 23.7 8.6
80 26.3 9.45

Chart-8: energy consumption w.r.t existing system and
proposed system
For the proposed system the X axis indicates the number of
nodes and the Y axis indicates energy. As the number of
nodes varies average energy consumption of existingsystem
increases. If the nodes are not energy efficient, network
lifetime can’t be increased. In proposed system energy
consumption gradually decreases due to tune up in the MAC
layer thus, extending the network lifetime.
5. CONCLUSION
The multipath AODV system schemes deals with the
alternates routes methods, which was not energy efficient
and time consuming. In this paper we focuses on energy
efficiency of AODV routing protocol by finding high residual
energy of each node. As per the simulation results after
considering energy module the variation in the network
parameters can be seen. So that the route which is having
higher energy will be considered to create path for route
discovery in the proposed EMAODV routing protocol rather
than just flooding the network with route request packet
every time, which leads to congestion in the network. the
problem of this is to find proper positions for the sensors at
any point in time such that some desired specifications are
satisfied, in proposed system total energy consumption
gradually decreases due to tune up in the MAC layer so this
will increases the network lifetime,overheadand congestion
in the network can be minimized.
In the future we can use light weighted key management
system for secure and efficient data management.
ACKNOWLEDGEMENT
We express our thanks to Department of Electronics and
Communication, Bangalore Institute of Technology (BIT).
REFERENCES
[1] S. Kumar, “Study of MANET: Characteristics, Challenges,
Application and Security Attacks.”, in International
Journal of Advanced Research in Computer Science and
Software Engineering (IJARCSSE),Vol 3(5),pp.266-274,
2013.
[2] K. Gorantala, “Routing Protocols in Mobile Ad-hoc
Networks” (master's thesis), Department of Computer
Science, UMEA University, 2006.
[3] Y. Bai, Y. Mai, amd N. Wang, “Performance Comparison
and Evaluation of the Proactive and Reactive Routing
Protocols for MANETs”, in Proc. IEEE Wireless
Telecommunication Symposium (WTS 2017), Chicago,
USA, pp. 1-5, April 2017.
[4] R. Gupta, “Firefly based Optimized Routing over
MANETs (master's thesis).” Department of Computer
Science and Engineering, Thapar University, 2015.
[5] T. A. Murshedi, X. Wang, and H. Cheng, “On-demand
Multipath Routing Protocols for Mobile Ad-Hoc
Networks: A Comparative Survey”, in the International
Journal of Future Computer and Communication, Vol.
5(No.3), June 2016.
[6] Y. Mai, Y. Bai, and N. Wang, “Performance Comparison
and Evaluation of the Routing Protocols for MANETs
using NS3”, in Journal of Electrical Engineering, Vol.5,
2017, ISSN 1582-4594. Doi:10.17265/2328-
2223/2017.04.001, pp. 187-195.
[7] M. A. Abdelshafy, and P.J. King,“DynamicSourceRouting
under Attacks”, in IEEE 7th International Workshop on
Reliable Networks Design and Modeling (RNDM 2015),
Munich, Germany, pp. 174-180, Oct. 2015.

IRJET- Energy Efficient Reactive Routing to Enhance QOS of Manets

More Related Content

What's hot (16)

Similar to IRJET- Energy Efficient Reactive Routing to Enhance QOS of Manets (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Energy Efficient Reactive Routing to Enhance QOS of Manets