Content uploaded by Sasmita Mohapatra
Author content
All content in this area was uploaded by Sasmita Mohapatra on Sep 30, 2019
Content may be subject to copyright.
International Journal of Innovative Technology and Exploring Engineering (IJITEE)
ISSN: 2278-3075, Volume-8 Issue-10, August 2019
2917
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number I8467078919 /2019©BEIESP
DOI: 10.35940/ijitee.I8467.0881019
Abstract: Making MANETs energy efficient has been a biggest
challenge from many years. In this regard in the previous works
routing protocols have been introduced for stabilized clustering
scheme with swarm intelligence where the MANETs are made
energy efficient with well balanced multipath routing protocols.
But in all the proposed protocols the cluster structures are non
overlapping where there is a probability that some of the nodes
may be left without any consideration in the clusters of MANET
as nodes are highly dynamic in nature. These nodes usually do not
belong to any of the clusters thus cannot take part in the routing
process unless cluster splitting takes place. In that case certain
time error may occur if any such kind of node carries any special
information. For that the clusters may be made overlapping but in
that case number of clusters may increase leading to complicated
structure where efficiency may decrease suddenly in case of node
movement or failure. For this a novel clustering and routing
procedure has been introduced in the paper where not only all the
nodes take part in data transmission but also number of clusters
are minimized to improve efficiency. The work has been carried by
using NS-2 simulator.
Index Terms: Energy-efficient, MANET, Clustering,
Non-overlapping, Node-utilization, Stable.
I. INTRODUCTION
Highlight a section that you want to designate with a certain
Mobile AdHoc Networks (MANETs) are self organized
structures which are very important for any wireless
communication structure starting from home to defense. As
MANETs are not fixed to any preapproved structure [1], [2]
so they can have many applications. In the previous work in
[3], [4] routing processes have been introduced where swarm
intelligence with proper cluster maintenance scheme has been
introduced for routing in MANET. For further improvement
in [5], [6], [7] schemes have been introduced where the
routing process is made multi path and more energy efficient
with a well balanced hierarchical clustered MANET structure.
In all the schemes the nodes are given a virtual ID which is
decided according to their residual energy and received signal
strength. A cluster head (CH) has been selected in all the
clusters which takes the major part in transferring the data
from source to destination and also carries highest VID. Also
in the above proposed routing algorithms the clusters of
MANET are considered which are non overlapping. But in all
the cases while doing clustering there is a probability that
some of the nodes are left which could not be considered in
any of the clusters as nodes are highly dynamic in any
MANET structure. These nodes can also carry certain
information or can help in the routing process. But proper
node utilization can’t be done as some of the nodes may not
get a membership as forager or scout in any of the cluster
since they fail to have minimum required residual energy or
received signal strength unless cluster splitting takes place.
Revised Manuscript Received on August 10, 2019
Dr. Sasmita Mohapatra, Associate Professor, Department of
Electronics and Communication Engineering Sapthagiri College of
Engineering, Bengaluru
This problem can be come over if the clusters are made to be
overlapping. Where the distance between such kind of node
and CH reduces and no doubt that node can get cluster
member identity. But this leads to a problem where the
number of clusters increases rapidly and it affects the overall
efficiency of the MANET structure since with increase in
number of clusters the system becomes unstable where the
rapid movement or failure of any node has to be always taken
care to rebuild the clusters. Also by this more updation of the
table is required which consumes more of the band resource.
As a solution to the above mentioned problem the clustering
scheme has been changed a bit which is influenced by
3HBAC (3-HOP BETWEEN ADJACENT CLUSTERS)
scheme as per [9]. To take care of the above problems a new
algorithm has been introduced where-
Care has been taken so that all the newly joining nodes
should be at least declared as cluster member of any
one cluster where a new identity is given to such kind
of nodes as “cluster guest”.
In this way all the nodes in any MANET structure can
take participation in the routing process may be
directly or indirectly.
The overlapping node concept is overruled by which
the MANET is made more efficient.
By avoiding the overlapping cluster concept the CH
changing frequency has been reduced which
increases the stability of the system.
II. PREVIOUS WORK CARRIED OUT
In the previous work carried out several steps has been taken
so that the routing protocol in the MANET can be made very
much energy efficient.In this regard in [3] bee inspired
routing along with clustering is selected as the best routing
process for MANET where each of the nodes according to
their residual energy and received signal strength are given
some virtual ID and classified as cluster head (CH), foragers
and scouts. Here the complete MANET structure is devided
into clusters. In [4] care has been taken for the stability of the
clustering scheme where the change in the CH is reduced by
introducing stable cluster maintenance scheme.
Fig-1: Workflow of Bee-AdHoc-C Network
Routing Protocol for Clustered Bee-Ad Hoc
MANETS with Proper node Utilization
Sasmita Mohapatra
Routing Protocol for Clustered Bee-Ad Hoc MANETS with Proper node Utilization
2918
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number I8467078919 /2019©BEIESP
DOI: 10.35940/ijitee.I8467.0881019
But in both cases the routing process was random as the nodes
were moving randomly outside the cluster in search of the
destination which may increase the traffic and thus the delay
in the routing process can be increased. In [5] the clustering
scheme has been made more systematic by introducing nodes
as Border Cluster Nodes (BCNs) which work as gateways
between adjacent clusters. Here if the nodes don not get the
destination within the cluster then they donnot move
arbitrarily to other cluster rather the process is made
systematic where BCNs are used for inter cluster
communication. But still the process was not fully efficient as
the load was not fully balanced among the clusters and the
shortest route finding was consuming unnecessary time. To
come over this problem a load balanced improved routing
technique was introduced in [6] where care has been taken
that none of the cluster head gets overloaded with more
number of nodes or any cluster head will be with less load.
Here cluster merging and cluster splitting are introduced with
proper selection of CH in each of the cases. Also the shortest
route is found between the source and destination by
calculating the hop count. For further improvement of the
scheme in [7] care has been taken for multipath routing where
the neighboring nodes from adjacent path should not disturb
the routing process.
Fig.2. Functionality of Improvised Bee AdHoc-C
Network
In all the above cases the number of clusters and the minimum
and maximum number of cluster members are predetermined.
But as in any MANET structure there is no central control
over the nodes movement and the nodes highly dynamic to
move arbitrarily so any time new nodes can enter the
MANET. In that case there is a probability that the number of
members in any cluster can increase suddenly and the cluster
has to go for cluster splitting and again should satisfy the
minimum member consideration in the newly formed cluster.
This may lead to certain delay. To come over this problem the
present algorithm is introduced where the newly added nodes
immediately do not bring a condition to split the cluster rather
if they can access any of the nodes of the cluster then they are
considered as cluster guest. This will also help in reducing the
data loss and the efficiency of the system increases. In the
present protocol proper utilization of all the nodes is given
priority.
III. PROPOSED TECHNIQUE FOR EFFICIENT NODE
UTILIZED ROUTING FOR CLUSTERED BEE-ADHOC
MANETS
In all the routing protocols explained above Border Cluster
Nodes are the nodes considered with lowest Virtual ID which
are supposed to be far from the Cluster Heads of adjacent
clusters but should carry a predefined residual energy (RE)
and received signal strength (RSS) to be considered as a
cluster member. Usually they are the common nodes for many
clusters which are adjacent and share the common boundary.
But in this way certain free nodes may not be considered as a
valid cluster member in any of the clusters since their RE and
RSS are very low when they enter the cluster structure
suddenly. In such a case these nodes cannot take any
participation in the routing process. But sometimes if these
nodes also take active participation that can be useful. Though
these nodes are seen to be far from the CH but mostly they are
seen to be very nearer to some other nodes (Foragers, Scouts
or BCNs) of the cluster where it is easy for them to
communicate with these nearby nodes. In that case these
nodes are introduced as “cluster guests” and the nearby nodes
which are already cluster members are considered as APs
(Access Points).
Fig.3. Functionality of Node Utilized Improvised Bee
AdHoc-C Network with cluster guest
Thus in the above figure all the members of any clustered
MANET structure are shown as CH, Foragers, Scouts and
Cluster Guests.
Working of the model
The working of the proposed model is discussed step by step
starting from the formation of cluster to transfer of data from
source to destination.
Cluster Formation
The MANET structure considered here is a liner network
where the x-axis of the working area and the range of
communication decide the number of clusters.
Selection of Cluster Members
The different cluster members are selected according to their
RSS and RE. According to these two values they are assigned
with VID. The node with highest VID is known as CH. The
other nodes are assigned as Foragers and Scouts.
International Journal of Innovative Technology and Exploring Engineering (IJITEE)
ISSN: 2278-3075, Volume-8 Issue-10, August 2019
2919
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number I8467078919 /2019©BEIESP
DOI: 10.35940/ijitee.I8467.0881019
Where Foragers’ VID is set to be greater than the VID of
Scouts. The nodes with lowest VID are taken as BCNs which
are common and stand as gateways to more than one cluster.
Here all the nodes maintain a status table where they maintain
their VID as well as they declare their status as CH, Forager or
Scouts. Here a change has been done where any node which
moves out of the transmission range of all existing cluster
heads but still able to communicate with any of the cluster
member is considered as a “cluster guest” of the specific
cluster. At that time that specific node joins the cluster and
considers the nearest cluster member as its Access Point (AP).
Cluster Maintenance
In this case the number of cluster members for any cluster is
predefined as per [6]. Where if the minimum and maximum
cluster members are not maintained then cluster merging and
cluster splitting takes place and new CH is selected. In
addition to this procedure if at any time any node finds that it
can communicate only with the node with a status of “cluster
guest” it forms a new cluster. Where the member behaves as a
cluster head with all its immediate cluster neighbors as cluster
members. For that new cluster again all the nodes are divided
as Foragers, Scouts and BCNs. Here usually the AP of the
cluster guest is considered as CH and the cluster guest is
considered as Forager. But in this case also the minimum and
maximum number of members in cluster is maintained as per
[6]. Also if any time any CH can find that all its cluster
members can be cluster guests of some other cluster and can
access the cluster head of other clusters at that time it
relinquishes its cluster head role and that CH joins a
neighboring cluster as a cluster guest. In this case the AP for
the CH which is newly turned to cluster guest of any other
cluster becomes one of its former cluster members. By
merging the clusters in this way the number of clusters can
reduce so that the cluster members can be maintained at a
more constant low level during the complete routing process.
Routing process
Once any event occurs and destination is decided the CH
sends a signal to it immediate Forager to search the
destination. In turn the Forager first searches the destination
within all the Foragers and then sends a request signal to all
the neighboring Foragers. If the destination is found among
the neighboring Foragers then an acknowledgement signal
comes from the destination Forger and the shortest route is
decided between the source to destination by calculating
minimum number of hops as stated in [6]. But if the
destination is not found among the Foragers then a request
signal is sent by al the Foragers to their neighboring scouts
and the destination is searched within the cluster. If the
destination is not found within the cluster then an error signal
is sent by the Scouts to Forager and Forager to CH. Then the
CH sends a request signal to the attached BCNs to search the
destination outside the cluster. In turn the BCNs send request
signal to the respective adjacent cluster’s CH and the search
of destination continues in the adjacent cluster. All the
procedure has already been discussed in [6] which may some
times give delay. But in the proposed algorithm since new
mobile nodes are introduced as Cluster Guests so they are also
used in the routing process. Usually the Cluster Guests have
nodes as their Access Points which are either BCNs or nodes
nearer to BCNs. In this case when a destination is not found
within the cluster and any BCN gets a request signal by the
CH then the BCN not only sends a request signal to the
adjacent CH but also it sends a signal to the nearby “cluster
guest”. After this the respective cluster guest tries to find its
destination through its access points and other cluster guests.
If the cluster guests find the destination then it sends a reply to
the BCN in turn BCN sends an acknowledgement signal to the
respective CH of the cluster where source is present. After
that the best route between source and destination is decided
by calculating the minimum hop count using the waggle dance
formula as stated in [6]. By this method, the routing process
can be made more stable and the delay in routing can be
reduced.
IV. SIMULATION MODEL
The simulation model is taken in X-Y coordinator where X-
shows the length and Y- shows the width of the MANET
structure. For the present simulation work, the length has been
selected to 1000m. and the width has been selected to 200m.
The maximum range of communication has been selected to
200m. on X-axis. Thus the complete network area is divided
to 5 clusters. The total number of nodes considered for
simulation is 50-100. The minimum residual energy of any
node to be considered as a cluster member as Forager or Scout
is 10J.
Simulation Model
Simulation Scenario
The simulation for the model has been carried using NS 2.35.
The different simulation criteria are taken as snap shots. Here
each figure specifies different conditions of simulation.
Fig 4 (a) The nodes are distributed randomly in the simulation
area
Simulator
Network Simulator 2.35
NUMBER OF NODES
50-100
AREA
1000m x 200m
COMMUNICATION RANGE
200m
INTERFACE TYPE
Phy/Wireless Phy
MAC TYPE
IEEE 802.11
QUEUE TYPE
Drop tail/Priority Queue
QUEUE LENGTH
50 Packets
ANTENNA TYPE
Omni Antenna
PROPAGATION TYPE
Two Ray Ground
ROUTING AGENT
NU-IBAC, LBEE-IBAC
TRANSPORT AGENT
UDP
APPLICATION AGENT
CBR
INITIAL ENERGY
10 Joules
TRANSMISSION POWER
20mW
RECEPTION POWER
10mW
SIMULATION TIME
50seconds
Routing Protocol for Clustered Bee-Ad Hoc MANETS with Proper node Utilization
2920
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number I8467078919 /2019©BEIESP
DOI: 10.35940/ijitee.I8467.0881019
The above figure shows a condition at the beginning of the
simulation process. Here more than 50 nodes are distributed
in the simulation area randomly.
Fig 4 (b) Destination is identified in the simulation area
The above figure shows a condition where the nodes are
distributed randomly in the simulation area and the
destination is identified
Fig 4 (c) Different clusters are formed
In the above figure, different clusters are formed and nodes are
distributed in different clusters. The nodes are identified as CH,
Foragers and Clusters.
Fig. 4 (d) Routing between source to destination through
different paths
After creation of different clusters the routing paths are decided
between the source to destination in the above figure.
V. SIMULATION RESULTS
The present routing protocol has been named as Node
Utilized Improvised BAC (NU-IBAC). This method is tested
for different parameters to find the working efficiency of any
MANET structure like Packet Delivery Ratio and Route
discovery time in seconds. The performance of the proposed
routing protocol has been compared with the performance of
Load Balanced Energy Enhanced – IBAC (LBEE-IBAC)
discussed in [6].
Packet Delivery Ratio: This shows the ratio between
number of packets delivered to the destination and the packets
generated at the source end (CH).
Fig. 5 (a) Packet Delivery Ratio
As in figure 5(a), PDR is varying in both protocols while with
respect to traffic load. Especially PDR decreases for any
protocol with increase in packet size. PDR in proposed
protocol is better than existing which does not change much
with increase in packet size. Route discovery time: It is the
time taken for a source-generated packet to find route to the
destination. In the present scenario, the CH takes all the
initiative to send the packets from source to destination. Here
it is found by finding the difference between sending time of
forward scout and receiving time of backward scout at CH.
Fig. 5 (b) Route discovery time in sec
International Journal of Innovative Technology and Exploring Engineering (IJITEE)
ISSN: 2278-3075, Volume-8 Issue-10, August 2019
2921
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number I8467078919 /2019©BEIESP
DOI: 10.35940/ijitee.I8467.0881019
Author-1 Photo
As in figure 5(b), route discovery time is increased in both
protocols while increasing the traffic load. But still in the
proposed protocol the route discovery time is reduced to a
greater extend due to the introduction of new nodes in any
cluster as “cluster guests”.
VI. CONCLUSION AND FUTURE SCOPE
The system is made more stable by introducing the new agents
in the cluster as “cluster guest”. The main reason to introduce
these new agents is as MANET is a self-organized structure so
there is always a chance of entering new mobile nodes to the
cluster of any network area. In this case if cluster splitting is
carried as the immediate solution since the cluster structures
have to meet the minimum and maximum number of nodes
condition in the cluster then sometimes the system leads to
instability as well as the delay in routing increases. To come
over this sometimes for a smooth routing process the protocol
can be used and the cluster guests are properly utilized.
But in certain practical cases if many of these nodes are
considered as cluster members there is a probability that the
properly balanced cluster structure may get disturbed leading
to further delay in finding the destination. Further work can be
carried to take care of all these problems and the efficiency of
the routing protocol can be estimated in terms of certain new
parameters as energy efficiency, end to end delay and
throughput etc.
REFERENCES
1. G. O. Young, “Synthetic Cai, Xuelian, YulongDuan, YingHe, JinYang,
Changle Li.: Bee-Sensor-C: an energy-efficient and scalable multipath
routing protocol for wireless sensor net-works. International Journal of
Distributed Sensor Networks. 2015,pp. 1-14 (2015)
2. Saleem, Muhammad, MuddassarFarooq.: Beesensor: a bee-inspired
power aware routing protocol for wireless sensor networks. In:
Workshops on Applications of Evolutionary Computation, pp. 81-90.
Springer, Berlin Heidelberg (2007)
3. SasmitaMohapatra, Siddappa, M.: Bee-Inspired Routing the ultimate
routing process for Energy Efficient MANET. International Journal of
Applied Engineering Research. ISSN 0973-4562. Vol. 10, Number 18,
pp. 38855-38862 (2015)
4. SasmitaMohapatra, Siddappa, M.: Stable Cluster Maintenance Scheme
for Bee-AdHoc-C: An Energy-Efficient and Scalable Multipath Routing
Protocol for MANET. In: Third International Conference On Advances
in Computing, Control and Networking (ACCN ), Bangkok (2015)
5. SasmitaMohapatra, Siddappa, M.: Improvised routing using Border
Cluster Node for Bee-AdHoc-C: An Energy-Efficient and systematic
Routing Protocol for MANETs, In: International Conference On
Advances in Computer Applications (ICACA), IEEE (2016)
6. Mohapatra, Sasmita, and M. Siddappa. "Load-Balanced
Energy-Enhanced Routing Protocol for Clustered Bee-Ad Hoc
MANETs." In Proceedings of First International Conference on Smart
System, Innovations and Computing, pp. 191-202. Springer, Singapore,
2018.
7. Mohapatra S., Siddappa M. (2019) Multi Path Routing in Load
Balanced Energy Enhanced Bee-AdHoc-C MANETs. In: Hemanth J.,
Fernando X., Lafata P., Baig Z. (eds) International Conference on
Intelligent Data Communication Technologies and Internet of Things
(ICICI) 2018. ICICI 2018. Lecture Notes on Data Engineering and
Communications Technologies, vol 26. Springer, Cham
8. Bentaleb, Abdelhak, Abdelhak Boubetra, and Saad Harous. "Survey of
clustering schemes in mobile ad hoc networks." Communications and
Network 5, no. 02 (2013): 8
9. Yu, J. Y., and Peter HJ Chong. "3hbac (3-hop between adjacent
clusterheads): a novel non-overlapping clustering algorithm for mobile
ad hoc networks." In 2003 IEEE Pacific Rim Conference on
Communications Computers and Signal Processing (PACRIM
2003)(Cat. No. 03CH37490), vol. 1, pp. 318-321. IEEE, 2003.
10. Yu, JaneYang, PeterHanJooChong.: A survey of clustering schemes for
mobile ad hoc networks. IEEE Communications Surveys & Tutorials 7,
no. 1, pp. 32-48 (2005)
11. Maan, Fahim, NaumanMazhar.: MANET routing protocols vs mobility
models: A performance evaluation. In: Third International Conference
on Ubiquitous and Future Networks (ICUFN), pp. 179-184. IEEE
(2011)
12. Gopinath, S., Sureshkumar, N., Vijayalakshmi, G., Natraj, N.A.,
Senthil, T., Prabu, P.: Energy Efficient Routing Protocol for MANET,
IJCSI International Journal of Computer Science Issues, Vol. 9, Issue 2,
No 1 (2012)
AUTHORS PROFILE
First Author
Dr. Sasmita Mohapatra received B.E degree in
Electronics and Instrumentation from Utkal
University, Orissa, India in 2001 and M.Tech
degree in Electronics & Communication
Engineering from IASE University of Rajasthan,
India in 2005. She completed her doctoral degree in
Computer Networking in the depaprtment of
Electronics & Communication Engineering from
Sri Siddhartha Academy of Higher Education, Tumkur, India under the
supervision of Dr.M.Siddappa., Dean Academics, SAHE. She is currently
working as an Associate Professor in the department of Electronics and
Communication Engineering at Sapthagiri College of Engineering,
Bangalore, India. She has teaching experience of 15 years and research
experience of 06 years. She published 12 Technical Papers in National,
International Conference and Journals. She has citation index of 14 till 2019
and h-index of 3.Her current research interests include Routing protocols for
Energy Efficient MANETs, Bee Ad-Hoc Network, RF & Antennas &
different communication techniques etc. She has got best paper award at 1st
International Confrence on Smart Systems, Innovations & Computing
(SSIC-2017) at Manipal University Jaipur, Rajasthan, India. She worked as
Assistant Professor in Department of Electronics and Communication
Engineering from 2006 to 2012 at Sapthagiri College of Engineering,
Bangalore. Presently, she is working as Associate Professor, Department of
Electronics & Communication Engineering since 2012 at Sapthagiri College
of Engineering.
