1

1 INTRODUCTION

MANET is a self-organizing network of mobile nodes

connected by wireless links without any centralized control. It is a
homogeneous system, a mobile devices group together and share
information to reach the goal. These devices progressively move various
locations. So they structure self – configuring, self-arranging and self-
directing. Nodes communicate directly with in transmission range is
single hop. A indirectly communicate depends on neighbor nodes outside
of transmission range is multi hop communication. Normally the route
discovery in AODV (Adhoc on demand distance vector routing protocol)
performed on demand and each device have access the medium which
involves adhoc networks group together an entire system. All Devices
broadcast a message to every single other node in the system. Multi-hop
message transmission can be brought out through the FLOODING, it
provides drawback as increasing repetitive data, transfer speed uses and
more delay. Using this transmission message, there may be affected by the
path. Hence it eliminates the link failure and guarantees to deliver data
packets a Stable neighbor nodes are selected for the transmission path. A
Path includes high received signal strength, it improves throughput. A
modified AODV new routing algorithm will be designed in such a way
that it effectively resolves issues such as redundant information,
Bandwidth utilizations, delay, error in the channel, Remove unstable
nodes and Reducing the number of dropped packets based on fading signal
by improving the packet delivery in that channel path.
 2

1.1 MOTIVATION OF THE RESEARCH

The AODV Routing protocol that has any of the helpful device are
not reacting with different devices, which means a device is heavily
loaded which involves the energy level may be reduced. In this network a
node becomes unavailable due to radio link fluctuation affected by the
path away disappointment on the flooding of the route request packet, it
will increase the number of RREQ and RREP packet. It produces high
routing overhead and reduces the routing stability and also packet
delivery ratio. This ensures that there is an initial route discovery process
depends on stable node are chosen instead of alternate path are
discovered. It reduces the routing overhead and route stability. When
there is a situation the Route changes due to the rapid node mobility of
AODV continuously that affects the link qualities, which leads to a path
loss and also does not provide the utilization of the resources shared by
the all the nodes are energy level, bandwidth and storage capacity etc. It
reduces the network performance. In order to overcome such problems, a
dynamic clustering mechanism with handoff technique is implemented to
avoid path loss and establishes a stable link between source and
destination. The single optimal path instead of multiple feasible solution
concept is deal with the swarm intelligent behavior of bees Colony
Algorithm (BCA). These basic concepts of the BCA can be used to
recognize the already feasible solution stable neighbor node and also
given me best results in Mobile ad hoc network to increase the packet
delivery ratio and reduce the routing overhead.
 3

1.2 PROBLEM DEFINITION

This work investigates

1. Detecting unstable node itself identification of Stable Neighbor Node

delegate all the activity concepts. It increases, the route stability and
packet delivery ratio.
2. The initial route discovery process depends on stable node are chosen
instead of alternate path are discovered. It reduces the routing overhead,
delay and route stability.
3. Resourceful Handoff Mechanism over Cluster Nodes is a node
becomes unavailable, it involves intra (nodes within the same cluster) and
inter (between nodes in different cluster) hand off is done to maintain the
node stability and link stability based on a stable path is provided. Once
creating the link, it Defective signal recovered due to delay without
affecting the transmission. The path stability is maintained and further
route discovery is eliminated, it reduces the transmission delay and
routing overhead.
4.Optimal path Based Artificial Bee Colony with Adhoc on Demand
Distance Vector (OPABC-AODV). Using Fitness of stable node and
Minimization of Broadcasting Delay depends on best path is selected,
which will be successfully transmitted a number of packets from source
to destination. With FSN factor, we eliminate the affected link and attain
successful data transmission. Even though during the unwanted situation
a link may fail, then handoff technique is implemented depends on the
Cluster Head (CH). Our solution, which stratify the given condition or
constraints is called feasible solution.it more than one feasible solution
 4

will be provides.This algorithm provides a which already feasible and

given me best results then that solution is called optimal solution.it
provides the high degree of stability, increases packet delivery ratio and
reduces routing overhead.

ORGANIZATION OF THESIS
The works reported in the thesis are organized as follows
Chapter 1: Introduction
Chapter 2: Literature Review
Chapter 3: Efficient Prior Path Failure Recovery Based Delegation for
AODV Protocol In The Presence Of Unstable Nodes
Chapter 4: QOS Parameter Based On Best Path Selection (QBP-AODV).
Chapter 5: Resourceful Handoff Mechanism over Cluster Nodes for
Stable Path Implementation in MANET.
Chapter 6: Enhanced Cluster Based Stable Path for Load Balancing and
Reducing Overhead Using Artificial Bee Colony
Chapter 7: Conclusion and Further scope

2 LITERATURE REVIEW

In various routing protocols are installed in Manet there with the

aid of reactive, proactive and hybrid routing protocol.Md. Monzur
Morshed, et. al.,[1] performance analyzed AODV and DSDV protocols
in terms of drop, throughput, delay and jitter. The performance of
DSDV is lesser than AODV in terms of packet release ratio,
communication overhead, delay and jitter. A packet delivery ratio
increases AODV Routing protocol compare to the DSDV table driven
 5

routing protocol. Also, they give the modification in AODV which

improves the execution of AODV in the presence of stable link.
Priyanka Thakkar And Prasanna Shete[2] Proposed Adaptive
Hello Messaging Based AODV Routing Protocol. This method mainly
discussed about hello message and interval time period. Local link
connectivity is discovered based on this hello message and default hello
interval value is assigned 1sec to broadcast the entire neighbor node in
this hello message. This process continuously broadcasting the message,
it leads to unnecessary wastage of high energy and bandwidth. To
overcome this problem, check the event interval is equivalent hello
interval.
Masood Ahmad, et.al.,[3] proposed Honey Bees algorithm based
efficient cluster formation and optimization scheme in MANET. This
algorithm makes utilization of honey bee’s calculation to discover high
adequate neighbor node for initial route discovery based on average
residual power, average signal strength and available of bandwidth and a
path has been established. The available path will be produced reliable
data transmission. Whenever the optimal path retains then the additional
route discovery overhead is removed.
Nihad I. Abbas, et.al.,[4] proposed Fuzzy approach to improving
route stability of the AODV routing protocol. Select the most efficiency,
stable nodes based on route discovery and forwarding the packets
between from source to destination using fuzzy concepts. This technique
is various parameters as node energy level, node mobility and hop count
are assigning the fuzzy logic to compute the rate of the stable node level,
which involves producing the packet delivery ratio, average end to end
delay and route stability.
 6

P. Srinivasan, P. Kamalakkannan [5] proposed algorithm makes

received signal strength depends on link stability and residual energy
level depends on node stability. Using this node stability and link
stability becomes to achieve the reliability factor (RF).Based on this RF
are calculated in an each node to select the max value as the optimal path
for route discovery and forwarding packets.
Sujata V. Mallapur[6] proposed concepts the first candidate node
are chosen relies upon available data transmission and residual
maximum life time. The nature of the connection and the portability is
likewise considered while choosing the stable node. It reduces the packet
dropped ratio and diminishes overhead. It does not provide a sufficient
number of stable nodes for discovery the route path.
Sedrati Maamar, Benyahia Abderezzak proposed [7] predict link
failure in AODV protocol to provide of service in MANET. AODV
includes substitution of existing devices Xp instead of Xi or presenting
new neighbors (Source-Xi-Xp-Destination) along the way or both can be
completed expanding signal quality. It improves the packet delivery ratio
and communication overhead, energy level, load factor, end to end delay
and jitter for selecting the stable neighbor node is not considered.
Mieso K. Denko and Hua Lu proposed [8] node becomes
dynamically change the topology, it affected by the link quality. The
resource not effectively utilized by all the devices such as battery
lifetime, storage capacity, and queue length,etc.In cluster based AODV
routing protocol is a centralized control and shared by the resources all
the nodes.
We propose a self-association based group plan in MANET
utilizing zone-based gathering improve scalability and stability of the
 7

entire network. Load balancing factor for congestion avoidance and

establish a stable route path depends on the behavior of bio-inspired. The
energy level will be reduces the proper utilization of resources. It shows
the result improves the network performance and energy level
reduces.[13]
Honey Bee Optimization (HBO) algorithm is one of powerful
algorithms in an artificial intelligence. It is swarm based Meta heuristic
optimization algorithm inspired by life cycle honey bee algorithm. There
are different honey bee’s algorithms have been developed based on style
building hives and searching and collecting the food etc..It is very
important to know the life cycle of honey bees. Bees in nature look for
food by exploring the neighborhood honey bee of their hives and they
collect and accumulate the food that is nectar. The bee’s information
through waggle dance move with short queue with the movement of ups
and downs. In this waggle dance every movement has different meaning.
QB’S determines the Fitness Value (FV) is energy level and distance
moving in each node and Sufficient Food Resources (SFR) is FV of
cluster member and head involves route discovery and forwarding data
packets. But this paper load balancing factor for congestion avoidance is
not considered.
They feature use to collect food the group of honey bees search
the region looking for the new flower patch. Once a bee finds a food
sources, it tends to hives and inform their mates about the location,
quality and quantity of the food sources.Similary the load balancing in
each CH control and shares the information with all its members more
effectively. As a result the CH maintains a long period of processing
time and produces reliable path.
 8

In this dancing area of hives and number of rounds on left and

right side and jumping ups and downs. Based on this science it will
advertise of the location and enraging the reaming members of their
colony. After the dance some recited members follow the scout bees to
find the food source to collect the more food. This cycle will be repeated
continuously while bees at the hive accumulate the nectar and explore
the new area with a potential food sources. This concept is analysis
suitable for sufficient neighbor node depends on route discovery, it
improves the packet delivery ratio and reduces the routing overhead for
proper route selection.
3. Efficient Prior Path Failure Recovery Based Delegation for AODV
Protocol In The Presence Of Unstable Nodes

MANET is an infrastructure less network in which the packet

delivery involves communication among the neighboring nodes between
from sender to receiver. Each node has limited resources such as battery
lifetime, storage capacity etc. In AODV communicate the neighbor node
travels from sender and receiver, due to the unstable node is affected by
the link, it must be provided by increasing the communication overhead,
end to end delay, jitter and reduces the packet delivery ratio. In Existing
system node failure happens and neighbor node detects the node
unavailability and sends the information the source node to alternative the
path. Source node takes the decision about the alternative path. This
involves communication and transfer overhead between that detects the
failure node and the source node. In our proposed system before the
existing node fails due to energy level or signal stability. The node itself
identifies a trusted neighbor node and delegates all its responsibilities to
 9

stable neighbor node and alternative the path without informing the
source node.
Delegation factor:
1. If node minimum signal strength than computation processing
delegation stable neighbor node. It increases the packet delivery ratio,
2. If node energy level or unavailability than computation processing
delegation stable neighbor node. Normal delegation occurs only when
node becomes unavailable. Thereby reducing the computation speed of
the node, which intern reduces the communication and transfer overhead.
Table 3.1 Simulation parameters PPFRD-AODV
Property Values
Simulation Time 950 Sec
Number of nodes 100
MAC protocol 802.11
Radio Range 250 m
Traffic Source CBR
Routing protocol AODV
Payload Size 512 bytes
Node Deployment Random
Mobility Model Random Way Point
Speed (m/Sec) 2m/Sec
Number of flows 20 flows

100
PACKET DELIVERY RATIO(%)

80

60
AODV
40 PPFRD-AODV
20
Figure 3.1 Number Node Vs Packet Delivery Ratio
0
0 5 10 15 20
NUMBER OF NODES
Figure 3.1 Number Node Vs Packet Delivery Ratio(%)
 10

0.05

END TO END DELAY

0.04
0.03
AODV
0.02
PPFRD-AODV
0.01
0
0 5 10 15 20
NUMBER OF NODES

Figure 3.1 Number Node Vs End to end Delay(Sec)

Table 3.2 Simulation Results of PPFRD – Unstable Node (PDR, End to End Delay)

NUMBER NODE PDR END TO END DELAY

0 91.38 0.020
5 88.22 0.025
10 84.74 0.028
15 76.24 0.034
20 71.67 0.042

4. QOS PARAMETER BASED ON BEST PATH SELECTION (QBP-

AODV).
The path selection is done with both having the minimum
transmission time and maximize throughput. Minimum transmission time
is achieved by estimating and reducing the queuing delay and processing
delay. Maximum throughput estimation is done by analyzing signals and
noise strength. In our proposed system QOS parameter based on best path
selection in AODV MANET. Queue delay is estimated by Bayesian
decision rule methodology by analyzing, packet arrival rate and packet
processing rate. The Hidden Markov model is used to find noise affection
 11

along the path. The selected best path satisfies QOS parameters such as the
increased delivery rate without packet loss and reducing average delay.
Table 4.1 Simulation parameters
Property Values
Simulation Time 1000Sec
Number of nodes 100
MAC protocol 802.11
Radio Range 250 m
Traffic Source CBR
Routing protocol AODV
Payload Size 512 bytes
Node Deployment Random
Mobility Model Random Way Point
Speed (m/Sec) 2m/Sec
Number of flows 20 flows
INCREASE COMMUNICATION

0.25
0.2
0.15 AODV
OVERHEAD

0.1 QOS-AODV
0.05
0
0 5 10 15 20
NUMBER OF NODE

Figure 4.1 Number Node Vs Increase Communication overhead

5. Resourceful Handoff Mechanism over Cluster Nodes for Stable

Path Implementation in MANET

Mobile Ad Hoc Network (MANET), node moves dynamically

without any centralized control. Each node has its own autonomous
behavior to perform the route discovery and packet forwarding. AODV
(Ad hoc on demand Distance Vector) establish the path and there may be
 12

path loss due to rapid node mobility. A node becomes unavailable route
discovery is done once again, but it may lead to routing overhead and
delay in transmission.

100 AODV
Packet Delivery Ratio

80 QOS-AODV
60
40
20
0
0 5 10 15 20
Number of Node

Figure 4.2 Number Node Vs Packet delivery Ratio

Table 4.2 Simulation Results of QOS-AODV – Unstable Node (PDR, Communication

Overhead, and End to End Delay)

NUMBER PDR COMMUNICATION END TO END

NODE OVERHEAD DELAY
0 93.28 0.100 0.017
5 89.93 0.1335 0.021
10 85.67 0.1613 0.025
15 79.42 0.195 0.031
20 73.69 0.2115 0.039

A novel approach dynamic clustering mechanism with handoff technique

is implemented to avoid path loss and establish a stable link between
sources to destination. Dynamic cluster is organized along the path based
on the angular velocity and angular acceleration of mobile nodes.
Whenever a node becomes unavailable, it involves intra (nodes within the
 13

same cluster) and inter (between nodes in different cluster) hand off is
done to maintain the node stability and link stability based on a stable
path is provided. Once creating the link, it Defective signal recovered due
to delay without affecting the transmission. The path stability is
maintained and further route discovery is eliminated, it reduces the
transmission delay and routing overhead.

Dynamic clusters are formed along the routing path in order to

reduce routing overhead and delay. A group of mobile nodes in the
cluster and controls the mobile nodes within the cluster head to form the
cluster structure. A cluster based route discovery maintains a stable path
in which utilizes the resource shared by the energy level, bandwidth and
by improving throughput ratio. The result produces by the Stability and
improving the efficiency of the network as well as reducing the route
discovering overhead.

Table 5.1 Simulation Parameters

Property Values
Simulation Time 1500Sec
Number of nodes 100
MAC protocol IEEE 802.11
Radio Range 250 m 6.
Traffic Source CBR Optimal
Routing protocol AODV
Payload Size 512 bytes path
Node Deployment Random Artificia
Mobility Model Random Way Point
l Bee
Speed (m/Sec) 2m/Sec
Colony
ABC (Artificial Bee’s Colony) Algorithm is a Swarm Intelligence
algorithm, which was produced by karaboga in 2005 [14]. This
calculation is to reproduce the conduct of Scout honey bees (SB) to
 14

recognize the different accessible food resources; Queen Bee (QB) has to
decide the allocation of Foraging bees (FB) to the various available food
resources from the information shared by Scout bees. FB actually collects
the food from the resource to the site.

0.11
ROUTING OVERHEAD

0.108
0.106
0.104
0.102 CAODV
0.1 DHCA-AODV
0.098
0 5 10 15 20
NUMBER OF NODES

Figure 5.1 Number Node Vs Routing Overhead

120
100
Throughput(%)

80 CAODV
60 DHCA-AODV
40
20
0
0 5 10 15 20
Numberof nodes

Figure 5.2 Number Node Vs Throughput

Table 5.2 Simulation Results of DHCA-AODV – Unstable Node (Throughput,

Routing Overhead)
NUMBER NODE THROUGHPUT ROUTING OVRHEAD
0 98.11 0.1024
5 92.23 0.1031
10 89.44 0.1045
15 86.47 0.1053
 15

20 81.58 0.1068

The SB involves a search of initial food sources from different

places and shares the information in available food source. QB receives
the different food resources from QB’S and it determines the fitness value
and sufficient food source of the information available from QB’S and
Deploy the FB’S to collect the food from the best available resource. The
deployment of FB’S is increased and decreased depending on the fitness
value and there is also a search new for possible of food resources by
SB’S by the Waggle dance when there is no sufficient food resource.
6.1. Artificial Bee Colony Algorithm:
Step1: Different SB’S search for different initial food Sources.
Step2: QB’S determines the Fitness Value (FV) and Sufficient Food
Resources (SFR) from the information Available from various SB’S.
Step3: QB’S allocate the FB’S at different food Resources depending on
the Fitness value of Each Resource.
Step 4: If the available resources at a particular site are diminishing,
numbers of FB’S are Decreased and Mapped to the other site.
Step5: Where there is better quality of food a Resource, the number of
FB’S at the better quality Food Resources are increased.
Step6: SB’S search for a new food resource if there is Not Sufficient
Food (NSF) from the Combination Of different site.
Table 6.1. Simulation parameters for OPABC Algorithm
Property Values
Simulation Time 1000Sec
Number of nodes 100
MAC protocol 802.11
Radio Range 250 m
Traffic Source CBR
 16

Routing protocol AODV

Payload Size 512 bytes
Node Deployment Random
Mobility Model Random Way Point
Speed (m/Sec) 2m/Sec
Number of flows 20 flows

The Optimal path concept is deal with the swarm intelligent

behavior of bees Colony Algorithm (BCA). These basic concepts of the
BCA can be used to recognize the stable neighbor node in Mobile ad hoc
network to increase the packet delivery ratio and reduce the routing
overhead.
The performance of the optimal path algorithm is considered by
patch density. A patch density depends on certain qualities of sugar
content. Bee’s communication through this waggle dance, it is direction
of flower patches and distance from the hive. If the food source increases
than the stable neighbor node of the path is also increasing Figure
(6.1).Similarly the Average Residual Energy (ARE) and Average Signal
Strength (ASS) based on the node is calculated to assign Fitness of Stable
Neighbor Node (FSNN). Considering the FSNN value is greater than the
threshold value using this constraint based a node is chosen from the
source to the destination on the path.
Using this FSNN communicate which each other and sharing the
information based on distance and direction of the flow.Nectar density is
usually affected by fluctuations of the delay on the path.As the fluctuation
of the delay increases, then food sources are decreases in the flower
patches figure (6.3). Searching process for route, discovering the new
food sources in fluctuations (variations) of the delay, the stable neighbor
 17

nodes are affected by the path then new route is discovered. If Max Life
of Node is decreased than fitness stable node is decreased Figure (6.2).
When Node becomes unavailable then New FSNN is selected by
means of handoff procedure otherwise Set of link availability connected
to the NN than estimate the link stability by means of signal strength .
Path stability is maintained and further route discovery is eliminated, it
reduces the transmission delay and routing overhead.

1
0.9
0.8
Fitness Stable Node(%)

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
10 20 30 40 50 60 70 80 90 100
Flower Patches (Sugar Content) Variations

Figure 6.1 Flower Patches (Sugar Content) Vs Fitness Stable Nodes

 18

1.2

Fitness Stable Node(%) 1

0.8

0.6

0.4

0.2

0
100 90 80 70 60 50 40 30 20 10
Max lIfe Time Of Node

Figure 6.2 Max Life Time of Node Vs Fitness Stable Nodes

1
0.9
0.8
Fitness Stable Node(%)

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
10 20 30 40 50 60 70 80 90 100
Fluctuation of the Delay

Figure 6.3 Fluctuation of the delay Vs Fitness Stable Node

Table 6.2 Simulation Results of OPABC

 19

FITNESS MAX FITNESS FITNESS

FLOWER FLUCTUATIO
STABLE LIFE OF STABLE STABLE
PATCHES (%) N DELAY (%)
NODE NODE(%) NODE NODE
10 0.188 100 0.999 10 0.93
20 0.222 0.964 20 0.87
90
30 0.268 0.946 30 0.81
80
40 0.367 0.921 40 0.76
70
50 0.444 0.901 50 0.69
60
60 0.583 0.812 60 0.54
50
70 0.641 0.783 70 0.43
40
80 0.733 0.518 80 0.36
30
90 0.862 0.217 90 0.23
20
100 0.91 0.113 100 0.11
10

1.2
Packet Delivery Ratio(%)

1
0.8
OPABC
0.6
DHCA
0.4
PPFRD
0.2
AODV
0
0 20 40 60 80
Nuumber of Nodes

Figure 6.4 Number of nodes in Percentage Vs Packet Delivery Ratio (PDR)

 20

Packet Delivery Ratio is a total number of received data packets by the

destination node and generated data packets by the source node. The Simulation result
of Packet delivery ratio of OPABC when compared to PPFRD, QOS, DHCA and
AODV routing protocol. A stable node increase then the packet delivery ratio also
increases. The AODV, PPFRD and DHCA performances are degraded by unstable
node. Be that as it may, OPABC execution broadly increments 18% than AODV and
roughly 7.2% increments than that of PPFRD and DHCA routing protocol in Figure
(6.4)

70

60
End to End Delay(%)

50
OPABC
40 DHCA
30 PPFRD

20 AODV

10

0
0 20 40 60 80
Number of Nodes

Figure 6.5 Number of nodes in Percentage Vs End to End Delay

End to End Delay is characterized as the normal time taken to deliver the
data packet from source to destination. It also includes the delay caused by route
discovery process (RREQ, RREP, and RERR) and the queue in data packet
transmission. High mobility nodes in the network increases the end-to-end delay
(Figure 5.9).The compared result of OPABC produces 14.4% than AODV and 3.1%,
2.3% of results get better than PPFRD and DHCA respectively in Figure(6.5)
 21

Routing Overhead(%)
8
7 DHCA
6 OPABC
5
4 PPFRD
3
2 AODV
1
0
0 20 40 60 80
Number of Nodes

Figure 6.6 Number of nodes in Percentage Vs Routing Overhead

The Routing Overhead is the link or node is failed, during the transmission the
problem has been happening, again and again finding the route discovery, it gives
more overhead. The simulation of routing overhead the result of the OPABC swarm
intelligent algorithm is better than the comparative performance of PPFRD, DHCA
and AODV under stable node in Figure 6.6.The compared result of AODV produces
16.7% than decreases in OPABC. PPFRD and DHCA routing protocol increases
routing overhead 7.3% compared to the OPABC.

Table 6.3 Simulation Results of OPABC

PACKET
NUMBER OF END TO END ROUTING
S.NO DELIVERY
NODES DELAY (%) OVERHEAD
RATIO

1 10 0.88 60.01 2.2

2 20 0.9 56.33 2.6

3 30 0.92 48.22 3.4

4 40 0.94 35.34 4.2

5 50 0.96 34.12 4.8

6 60 0.98 32.56 5.1

 22

7. CONCLUSION

. In this work investigation have been evaluated improving the

packet delivery ratio and reduces the routing overhead in an OPABC.
Using NS-2 Simulator the various concepts are compared and
implementation that performs as packet delivery ratio and routing
overhead.The packet delivery ratio increases, then the node stability and
link stability to consider as stable path,it provides improving the packet
delivery ratio and reduces the routing overhead. If the number of stable
Nodes is increased then the data delivery also increase.Aim of this work
is choose the best method for route discovery and forwarding data
packets without packet dropped and routing overhead in optimal path.
DHCA, PPFRD and OPABC methods are implemented to reduce the
Packet dropped ratio and overhead in the network.Simulation results
validate the ability of the proposed protocol to successfully packet
delivery ratio and reduces the overhead against the AODV routing
protocol.