Performance analysis on self organization based clustering scheme for FANETs using K-means algorithm and firefly optimization

International Journal of Informatics and Communication Technology (IJ-ICT)
Vol. 11, No. 2, August 2022, pp. 148~159
ISSN: 2252-8776, DOI: 10.11591/ijict.v11i2.pp148-159  148
Journal homepage: http://ijict.iaescore.com
Performance analysis on self organization based clustering
scheme for FANETs using K-means algorithm and firefly
optimization
Yashu Pulhani, Ankur Singh Kang, Vishal Sharma
Department of Electronics and Communication, University Institute of Engineering and Technology, Panjab University,
Chandigarh, India
Article Info ABSTRACT
Article history:
Received Jan 10, 2022
Revised May 30, 2022
Accepted Jun 10, 2022
With the fast-increasing development of wireless communication networks,
unmanned aerial vehicle (UAV) has emerged as a flying platform for wireless
communication with efficient coverage, capacity, reliability, and its network
is called flying ad-hoc network (FANET); which keeps changing its topology
due to its dynamic nature, causing inefficient communication, and therefore
needs cluster formation. In this paper, we proposed a cluster formation,
selection of cluster head and its members, connectivity and transmission with
the base station using the K-means algorithm, and choice of an optimized path
for transmission using firefly optimization algorithm for efficient
communication. Evaluation of performance with experimental results are
obtained and compared using the K-means algorithm and firefly optimization
algorithm in cluster building time, cluster lifetime, energy consumption, and
probability of delivery success. On comparison of firefly optimization
algorithm with firefly optimization algorithm, i.e., K-means algorithm results
proved than without firefly optimization algorithm, better in terms of cluster
building time, energy consumption, cluster lifetime, and also the probability
of delivery success.
Keywords:
Cluster
Flying ad-hoc networks
Firefly optimization
Glowworm swarm optimization
K-means algorithm
Unmanned aerial vehicle
This is an open access article under the CC BY-SA license.
Corresponding Author:
Yashu Pulhani
Department of Electronics and Communication, University Institute of Engineering and Technology
Panjab University
Chandigarh, India
Email: yash7yashu@gmail.com
1. INTRODUCTION
Technical progress in the area of flying platforms emerged unmanned aerial vehicle (UAV) as a
promising technology. UAVs have various applications in the field of military, navigation, surveillance,
broader supervision, medical supply, control, rescue operations, an inspection of hazardous events, disaster
management, monitoring, relay network, and communication [1]–[7]. Not only do these UAVs have
capabilities in sensing, storage, and processing, which makes it an intelligent system for communication, the
internet of things (IoT) [8]–[11], and the establishment of a smart city. These IoT services include package
delivery, tracking, and search operations. The unique characteristics of UAVs, reliability, survivability,
scalability, flexibility, cost-effectiveness [12], [13] make them an efficient choice for IOT, and communication.
UAVs are commonly called drones. The network of UAVs is called flying ad-hoc network (FANET)
[14]. It is one step up-gradation towards wireless communication technology. Still, a problem in this advanced
technology arises due to its mobile nature of communication that causes a change in its network topology that

Int J Inf & Commun Technol ISSN: 2252-8776 
Performance analysis on self organization based clustering scheme for FANETs using … (Yashu Pulhani)
149
results in problems in their connection maintenance and hence results in inefficient communication. The most
commonly adopted solution to this problem is clustering [15] and opting optimization.
Clustering is an unsupervised technique to divide the network into groups [16]. Each cluster consists
of a cluster head (CH) and cluster members (CMs). The selection of CH plays a vital role in cluster maintenance
and its operation. All the inter and intra operations in clusters, such as routing procedures of the cluster, a
connection of cluster and base station, and removal of the dead node, are performed by the CH. Therefore, the
election of the CH is done with various parameters. For determining the fitness function of a cluster, different
clustering schemes have been proposed by the authors.
Yang [17] proposed a routing mechanism for wireless sensor networks (WSNs) and UAVs. The
network is divided into clusters and sensor nodes. Each UAV knows the position of their CH, and their routing
mechanism is based on ant colony optimization (ACO). The behavior of ants inspires this mechanism in
searching for the shortest path to their destination. Maistrenko et al. [18] proposed a routing mechanism
inspired by the behavior study of ants for optimal pathfinding. This mechanism is inspired by two types of
behaviors of ants—forward and backward. In forward, ants, in search of their food, keep on secreting
pheromones in their way, and backward ants follow the path with the highest pheromones as an optimal path
to their search.
Al-Aboody and Al-Raweshidy [19] proposed a multilevel routing mechanism for the election of CH.
It is of three-level process. In the first level, the CH election is based on the residual energy, distance from the
ground control station and energy. The second level is an election of CH based on the grey wolf optimization
(GWO) algorithm, the number of UAVs in its neighborhood and its residual energy; the third level is the tree-
based approach for CH selection. Khan et al. [20] proposed a clustering scheme for the dynamic nature of
UAVs, causing instability in the routing process of FANET. This paper presents a bio-inspired clustering
scheme, a hybrid of glowworm swarm optimization (GSO) and Krill Herd (K.H.). Cluster head selection is
proposed based on the GSO algorithm and cluster management is proposed using K.H. Genetic operators are
used for determining the position of UAV, path detection based on residual energy, number of neighboring
UAVs, and distance between UAVs. Performance is evaluated in terms of quality-of-service parameters as
cluster building time, energy consumption, cluster lifetime, probability of delivery success with GWO, and
ACO.
Khan et al. [21] proposed that the mobility and changing topology of UAV create communication
issues for efficient communication. GSO is used for cluster formation and maintenance. Cluster head station
and its connectivity with ground control station are maintained using GSO and route mechanism based on
neighborhood, residual energy, and position of UAV.
The above-proposed clustering optimizations algorithms performed well. With motivation from these,
this paper has contributed to: i) we propose a new clustering mechanism in FANET using the K-means
algorithm and firefly optimization algorithm, ii) we propose a K-means algorithm for cluster formation,
selection of CH, CMs, and performing routing operation between nodes and ground control station (GCS), iii)
we propose a novel optimization technique for routing using a firefly optimization algorithm, and iv) we
propose a comparison of performance efficiency of the proposed k-means algorithm and firefly algorithm
(FFA) for routing mechanism in terms of quality of service (QoS) parameters of energy consumption, cluster
building time, cluster lifetime, and probability of delivery success.
The rest of the paper is tabulated as: section 2 explains the k-means algorithm for cluster formation,
cluster management, and routing between nodes. Section 3 describes the optimized technique of routing using
firefly optimization. Section 4 provides an explanation of the evaluation of FANET using k-means and firefly
optimization. Section 5 concludes the paper.
2. K-MEANS ALGORITHM
The K-means is the most straightforward and unsupervised clustering algorithm [22]. It is a method
to divide the network into groups called clusters consists of CH and CMs. This algorithm is a method to group
data into clusters based on Euclidean distance between data members and centroid consists of various steps:
i) from multiple nodes randomly select a UAV called CH of a cluster and ii) calculate the Euclidean distance
of CH with every node. Distance is calculated by (1):
(𝑝𝑘, 𝑞𝑗) = √∑ (𝑝𝑘𝑖 − 𝑞𝑗𝑖)
2
𝑑
𝑖=1 (1)
where, 𝑝𝑘 denotes 𝑘𝑡ℎ
data member and 𝑞𝑗 denotes centroid of cluster j, i denotes the number of data members.
− Nodes with a smaller Euclidean distance from CH become members of a similar cluster.
− Calculate the mean of updated cluster members of a cluster.

 ISSN: 2252-8776
Int J Inf & Commun Technol, Vol. 11, No. 2, August 2022: 148-159
150
− If the calculated mean is same as that of CH of the cluster, then iteration stops and CH gets updated.
− If the difference between mean and CH is there, then number of iterations continues till there is little change
in CH after each iteration, or no change.
Similarly, CH and CMs are elected in each cluster of FANET structure from many UAVs. In
Figure 1, the number of initialized nodes are shown, cluster formation, and selection of CH and CMs take this
node's place using the above-stated standard K-means algorithm. In the Figure 2, initialized nodes are divided
into clusters. Each cluster node is shown in different colors to signify other cluster formations in a region using
a K-means algorithm.
Figure 1. The number of initial nodes
Figure 2. Formation of cluster using K-means algorithm
In Figures 3 and 4, the cluster head is selected based on the above-stated rules of choosing the K-
means algorithm and connected with cluster members. The selection of cluster head is an essential parameter
in cluster operation for data transmission from a cluster to other clusters. During transmission, data from the
source cluster member transmits to that cluster's cluster head. The cluster head of the source cluster sends the
data to the destination cluster head and the cluster head to the respective destination. That is how transmission
in FANET operation takes place. The transmission of data from cluster to base station also similarly takes
place. Every cluster head of the cluster is connected with a base station for efficient communication. Both
Figures 5 and 6 show all cluster heads' connections with the base station for their efficient transmission besides
the effective communication of all clusters with the base station. From cluster formation to connectivity with
the base station, this paper achieves using the K-means algorithm can be seen in Algorithm 1.

151
Figure 3. Selections of CH and CMs
Figure 4. Connectivity of CH and CMs
Figure 5. Connection of all CH of clusters with base station (BS)

 ISSN: 2252-8776
152
Figure 6. Transmission of data from source to BS.
Algorithm 1: Cluster formation and routing using K-means algorithm
Initialize number of nodes, i = 1,2……, n
RTR <= Request to reply
Dest <= To specify destination found
Check pt <= no of times check nodes to reach destination
For (select CH of every cluster)
While (connect every CH with its nodes)
Locate GCS location
For (connect GCS with every CH and other GCS)
Locate radius of GCS and source node position specify number of check points
For 1: n, for all n nodes
Check (distance of source node with every node in its specified range to reach
destination)
If (RTR==1) // connection set up from source to destination
Else
Check (every node in range to reach destination)
If (packet reaches destination)
then Dest = 1;
Else
check till reaches checkpoint
End
Feed the sequence of intermediate nodes from source to destination.
End
End
End
3. FIREFLY OPTIMIZATION ALGORITHM
The firefly optimization algorithm is a subset of the swarm intelligence (SI) algorithm. These
algorithms are inspired from natural phenomena and characters as humans, ant, bee, birds, worms, flies,
termites, and fishes. Inspired by these biological agents, there are multiple optimizations available there:
particle swarm optimization, ant colony optimization, cuckoo search optimization, glowworm swarm
optimization, and firefly optimization [23], [24]. In this paper, we will discuss firefly optimization. Before that,
we will discuss the origin and drawback of optimization that leads to this novel optimization and how it
overcomes its flaw.
The authors developed the firefly optimization algorithm in 2007 and 2008 [25]–[27]. It is inspired by
the natural phenomena of flashing patterns and the behavior of fireflies. It originated to overcome the drawback
of GSO. The flashing behavior of lighting worms inspires GSO [25]–[27]. Each glowworm has its
luminescence called luciferin, which decides its current position. Glowworm with lesser luciferin value moves
towards a brighter one within decision range, changing its position and luciferin level. A higher luciferin value
is closer to an optimal solution that keeps the iteration process on until optimal value. The drawback in this
process is its unconstrained functions in higher dimensions.
Firefly Optimization is inspired by winged insects that produce a bioluminescent flash, which they
use for communication and attracting prey. It was developed at Cambridge University in late 2007 [28], [29].
Yang made three idealized rules for this optimization: i) fireflies are unisex, and they attract one another
regardless of their sex, ii) attractiveness between two fireflies is directly proportional to their brightness and
inversely proportional to the distance between them. As the distance increases, their brightness decreases. Thus,

153
brighter fireflies attract the less bright ones, but if both have the same intelligence, this will lead to the random
movement of both the fireflies, and a landscape of the objective functions determines the brightness of fireflies
[30], [31]. For n fireflies, the brightness of the firefly i is associated with objective function f (x):
𝐼 = 𝑓(𝑥) (2)
It reveals the current position of firefly from the set 𝑥1,𝑥2________𝑥𝑑. Suppose, for two fireflies, m and n namely
with their 𝑥𝑚, 𝑥𝑛 positions respectively. Their attraction depends upon their quantity of brightness, and some
constraints between both of them like the absorption of air and the square of the distance between these two
fireflies. Relative distance between these two fireflies is:
𝐼𝑚𝑛(𝑟𝑚𝑛)=
𝐼𝑛
𝑟𝑚𝑛
2 (3)
or 𝐼𝑚𝑛(𝑟𝑚𝑛) = 𝐼𝑚𝑛𝑒−𝛾𝑟𝑚𝑛
2
(4)
where 𝐼𝑚𝑛 is absolute luminesce of firefly m and firefly n, 𝑟𝑚𝑛 is distance between both fireflies m and n
respectively, 𝛾 is light absortion coefficient. The attraction between fireflies is represented as:
𝛽𝑚𝑛(𝑟𝑚𝑛) = 𝛽0𝑒−𝛾𝑟𝑚𝑛
2
(5)
𝛽0 firefly attraction at r = 0, 𝛾 is light absorption coefficient, in many applications, it is generally taken as
[0.01, 1.00]. Now, firefly n attracted towards firefly m updated its position by the (6):
𝑥𝑛 = 𝑥𝑛 + 𝛽𝑚𝑛(𝑟𝑚𝑛)(𝑥𝑚 − 𝑥𝑛) + 𝛼𝜀𝑛 (6)
where, 𝛽𝑚𝑛(𝑟𝑚𝑛) is an attraction parameter and 𝛼𝜀𝑛 is a randomization parameter [32].
3.1. Firefly routing mechanism
In the firefly routing mechanism, the optimal path for data transmission from a source node to a
destination is selected using the firefly algorithm. The optimal way comprises judging of failure node that is a
node causing delay, more energy consumption, and unsuccessful delivery during transmission by eliminating
a node and selecting a course that comprises maximum efficiency using an algorithm. Artificial neural network
(ANN) is used for optimizing a path of transmission and ANN inspired by the biological nervous system of the
animal brain [33]. In ANN, data transmission takes place by weights, unlike synapses in a biological brain.
This system consists of an input layer, hidden layer, and output layer [33], [34]. The hidden layer is used for
net computation and producing net input applied with activation function to obtain the actual output. These
systems are designed for self-training, implementing and testing processes [35]. This feature of ANN makes it
to help optimize a path for data transmission that includes tracing a faulty node by self-learning capability and
make corrections in it by selecting the optimal route. It then tests the process, again and again, to analyze
whether the process going on should be correct without any amendments. In above-shown Figures 7 and 8, fail
nodes are replaced with optimized paths using the firefly optimization algorithm. In this algorithm, faulty nodes
are traced, and optimization of the path from source to destination is performed using firefly and ANN
techniques of self-training, implementing, and testing process.
Figure 7. Faulty (fail) node is shown with the red
star near BS using ANN
Figure 8. Faulty (fail) node is shown with a red
star near source node using ANN

 ISSN: 2252-8776
154
Algorithm 1: Optimization using firefly
Initialize the empty list of fail UAV
For (data transmission from source to destination)
Check if (fail UAV~=source node & fail UAV~= Destination node)
Then (label it as fail)
fail UAV = fail UAV +1
Update (Transmission path by replacing node)
End
4. RESULTS AND DISCUSSION
We discussed the transmission using K-means an optimized path selection using the firefly
optimization algorithm. Based on optimized path selection, we will evaluate some QoS parameters using
software simulation in MATLAB in terms of cluster lifetime, cluster building time, energy consumption, and
the probability of successful delivery and analysis their results on different area sizes.
4.1. Cluster lifetime
Cluster lifetime is the time of a cluster from its formation to its disposition. That is, its selection as a
cluster head till it reaches its threshold and selection of new cluster head depending on fitness function of
cluster members that complete the requirement of cluster management. In Figures 9 and 10, cluster lifetime is
lesser during transmission using the K-means (without FFA) and improved by selecting the optimized path for
routing using FFA. So, FFA optimization proved better than without using FFA for transmission on different
area sizes.
Figure 9. Cluster lifetime vs no of UAVs in grid size 1000x1000 m2
Figure 10. Cluster lifetime vs no of UAVs in grid size 3000x3000 m2

155
4.2. Energy consumption
Energy is an essential aspect of UAVs. It is an energy consumption from cluster building to its lifetime
for performing all UAV operations, for communication, sensing, route selection, rate of successful packet
deliveries, transmission delays, and collisions during transmission. Transmitters, receivers, and amplifiers
installed also cause energy consumption in the system. On varying area sizes in Figures 11 and 12, energy
consumed during transmission using a K-means algorithm (without FFA) is lesser than the energy consumed
using transmission using the optimized path of the FFA. By increasing the number of UAVs, energy
consumption also increases.
Figure 11. Energy consumption vs no of UAVs in
grid size 1000x1000 m2
Figure 12. Energy consumption vs no of UAVs in
4.3. Cluster building time
Cluster building time is the time taken for cluster formation, selection of CH and its associated CMs.
An increase in the number and energy consumption in UAVs leads to an increase in cluster building time. Low
memory and computation power in UAV also affects cluster building time. In Figures 13 and 14, cluster
building time is lesser during transmission using the K-means (without FFA) and improved by selecting an
optimized path for routing using FFA. So, FFA optimization proved to be better than without using FFA for
transmission on different area sizes.
Figure 13. Cluster building time vs no. of UAVs in
Figure 14. Cluster building time vs no. of UAVs in
4.4. Probability of delivery success
The probability of delivery success comprises of successful transmission of data from source to the
destination node without any likelihood of data drop and node failure cases. The increase in the number of
nodes results in an increase in network density and a decrease in the data drop ratio. Hence, increase in the
probability of delivery success. From Figures 15 and 16, the likelihood of delivery success is more using
optimized transmission of the firefly algorithm.

 ISSN: 2252-8776
156
Figure 15. Probability of successful delivery vs no.
of UAVs in grid size 1000x1000 m2
Figure 16. Probability of successful delivery vs
no of UAVs in grid size 3000x3000 m2
5. CONCLUSION AND FUTURE SCOPE
In this paper, we have discussed UAV and their applications. Network of UAV called FANET, its
changing topology and its effect on communication network calls for the need of clustering. Clustering, using
the K-means algorithm, includes a grouping of nodes, selection of CH and CMs of every cluster, and connecting
all CHs of the cluster with BS for efficient communication. Furthermore, the transmission of the data source
to the destination node using a K-means algorithm, and then selection of optimized path using FFA. Comparing
FFA with without FFA (K-means), results proved that FFA is better in cluster building time, energy
consumption, cluster lifetime, and probability of delivery success.
In the future scope of this paper, work can be done on an optimized path using FFA. By saving an
approach, we can transmit confidential information by using it as a trust management path. Work can also be
done to enhance the optimized way every different transmission path can be selected without any chance of
failure.
ACKNOWLEDGEMENTS
This project was supported by the Department of Electronics & Communication Engineering &
Information Technology for research work under the University Institute of Engineering and Technology
guidelines, Panjab University, Chandigarh, India.
REFERENCES
[1] M. Mozaffari, W. Saad, M. Bennis, and M. Debbah, “Mobile Unmanned Aerial Vehicles (UAVs) for Energy-Efficient Internet of
Things Communications,” IEEE Transactions on Wireless Communications, vol. 16, no. 11, pp. 7574–7589, Nov. 2017, doi:
10.1109/TWC.2017.2751045.
[2] R. Ghanavi, E. Kalantari, M. Sabbaghian, H. Yanikomeroglu, and A. Yongacoglu, “Efficient 3D aerial base station placement
considering users mobility by reinforcement learning,” in 2018 IEEE Wireless Communications and Networking Conference
(WCNC), Apr. 2018, pp. 1–6, doi: 10.1109/WCNC.2018.8377340.
[3] M. Mozaffari, W. Saad, M. Bennis, and M. Debbah, “Unmanned Aerial Vehicle With Underlaid Device-to-Device
Communications: Performance and Tradeoffs,” IEEE Transactions on Wireless Communications, vol. 15, no. 6, pp. 3949–3963,
Jun. 2016, doi: 10.1109/TWC.2016.2531652.
[4] M. Mozaffari, W. Saad, M. Bennis, Y.-H. Nam, and M. Debbah, “A Tutorial on UAVs for Wireless Networks: Applications,
Challenges, and Open Problems,” IEEE Communications Surveys & Tutorials, vol. 21, no. 3, pp. 2334–2360, 2019, doi:
10.1109/COMST.2019.2902862.
[5] M. Mozaffari, W. Saad, M. Bennis, and M. Debbah, “Drone Small Cells in the Clouds: Design, Deployment and Performance
Analysis,” in 2015 IEEE Global Communications Conference (GLOBECOM), Dec. 2014, pp. 1–6, doi:
10.1109/GLOCOM.2014.7417609.
[6] M. Mozaffari, W. Saad, M. Bennis, and M. Debbah, “Efficient Deployment of Multiple Unmanned Aerial Vehicles for Optimal
Wireless Coverage,” IEEE Communications Letters, vol. 20, no. 8, pp. 1647–1650, Aug. 2016, doi:
10.1109/LCOMM.2016.2578312.
[7] Y. Zeng, R. Zhang, and T. J. Lim, “Wireless communications with unmanned aerial vehicles: opportunities and challenges,” IEEE
Communications Magazine, vol. 54, no. 5, pp. 36–42, May 2016, doi: 10.1109/MCOM.2016.7470933.
[8] A. Al-Fuqaha, M. Guizani, M. Mohammadi, M. Aledhari, and M. Ayyash, “Internet of Things: A Survey on Enabling Technologies,
Protocols, and Applications,” IEEE Communications Surveys & Tutorials, vol. 17, no. 4, pp. 2347–2376, 2015, doi:
10.1109/COMST.2015.2444095.
[9] T. Park, N. Abuzainab, and W. Saad, “Learning How to Communicate in the Internet of Things: Finite Resources and
Heterogeneity,” IEEE Access, vol. 4, pp. 7063–7073, 2016, doi: 10.1109/ACCESS.2016.2615643.

157
[10] A. Zanella, N. Bui, A. Castellani, L. Vangelista, and M. Zorzi, “Internet of Things for Smart Cities,” IEEE Internet of Things
Journal, vol. 1, no. 1, pp. 22–32, Feb. 2014, doi: 10.1109/JIOT.2014.2306328.
[11] G. Ding, Q. Wu, L. Zhang, Y. Lin, T. A. Tsiftsis, and Y.-D. Yao, “An Amateur Drone Surveillance System Based on the Cognitive
Internet of Things,” IEEE Communications Magazine, vol. 56, no. 1, pp. 29–35, Jan. 2018, doi: 10.1109/MCOM.2017.1700452.
[12] J. Li, Y. Zhou, and L. Lamont, “Communication architectures and protocols for networking unmanned aerial vehicles,” in 2013
IEEE Globecom Workshops (GC Wkshps), Dec. 2013, pp. 1415–1420, doi: 10.1109/GLOCOMW.2013.6825193.
[13] I. Jawhar, N. Mohamed, J. Al-Jaroodi, D. P. Agrawal, and S. Zhang, “Communication and networking of UAV-based systems:
Classification and associated architectures,” Journal of Network and Computer Applications, vol. 84, pp. 93–108, Apr. 2017, doi:
10.1016/j.jnca.2017.02.008.
[14] İ. Bekmezci, O. K. Sahingoz, and Ş. Temel, “Flying Ad-Hoc Networks (FANETs): A survey,” Ad Hoc Networks, vol. 11, no. 3, pp.
1254–1270, May 2013, doi: 10.1016/j.adhoc.2012.12.004.
[15] S. M. Ali, S. G. Fatima, S. K. Fatima, and S. A. Sattar, “Energy efficient routing in wireless sensor networks based on QoS,”
International Journal of Advanced Research in Engineering and Technology, vol. 10, no. 2, pp. 97–104, Mar. 2019, doi:
10.34218/IJARET.10.2.2019.012.
[16] S. Rana, S. Jasola, and R. Kumar, “A hybrid sequential approach for data clustering using K-Means and particle swarm optimization
algorithm,” International Journal of Engineering, Science and Technology, vol. 2, no. 6, Feb. 2011, doi: 10.4314/ijest.v2i6.63708.
[17] J. Yang et al., “Path planning of unmanned aerial vehicles for farmland information monitoring based on WSN,” in 2016 12th World
Congress on Intelligent Control and Automation (WCICA), Jun. 2016, pp. 2834–2838, doi: 10.1109/WCICA.2016.7578794.
[18] V. A. Maistrenko, L. V. Alexey, and V. A. Danil, “Experimental estimate of using the ant colony optimization algorithm to solve
the routing problem in FANET,” in 2016 International Siberian Conference on Control and Communications (SIBCON), May 2016,
pp. 1–10, doi: 10.1109/SIBCON.2016.7491805.
[19] N. A. Al-Aboody and H. S. Al-Raweshidy, “Grey wolf optimization-based energy-efficient routing protocol for heterogeneous
wireless sensor networks,” in 2016 4th International Symposium on Computational and Business Intelligence (ISCBI), Sep. 2016,
pp. 101–107, doi: 10.1109/ISCBI.2016.7743266.
[20] A. Khan, F. Aftab, and Z. Zhang, “BICSF: Bio-Inspired Clustering Scheme for FANETs,” IEEE Access, vol. 7, pp. 31446–31456,
2019, doi: 10.1109/ACCESS.2019.2902940.
[21] A. Khan, F. Aftab, and Z. Zhang, “Self-organization based clustering scheme for FANETs using Glowworm Swarm Optimization,”
Physical Communication, vol. 36, p. 100769, Oct. 2019, doi: 10.1016/j.phycom.2019.100769.
[22] K. Singh, D. Malik, and N. Sharma, “Evolving limitations in K-means algorithm in data mining and their removal,” IJCEM
International Journal of Computational Engineering & Management ISSN, vol. 12, no. April, pp. 2230–7893, 2011, [Online].
Available: www.IJCEM.org%5Cnwww.ijcem.org.
[23] X. S. Yang and X. He, “Firefly algorithm: recent advances and applications,” International Journal of Swarm Intelligence, vol. 1,
no. 1, p. 36, 2013, doi: 10.1504/IJSI.2013.055801.
[24] I. Fister, I. Fister, X.-S. Yang, and J. Brest, “A comprehensive review of firefly algorithms,” Swarm and Evolutionary Computation,
vol. 13, pp. 34–46, Dec. 2013, doi: 10.1016/j.swevo.2013.06.001.
[25] N. F. Johari, A. M. Zain, M. H. Noorfa, and A. Udin, “Firefly Algorithm for Optimization Problem,” Applied Mechanics and
Materials, vol. 421, pp. 512–517, Sep. 2013, doi: 10.4028/www.scientific.net/AMM.421.512.
[26] X.-S. Yang, “Firefly Algorithm, Lévy Flights and Global Optimization,” in Research and Development in Intelligent Systems XXVI,
London: Springer London, 2010, pp. 209–218.
[27] T. Hassanzadeh and M. R. Meybodi, “A new hybrid approach for data clustering using firefly algorithm and K-means,” in The 16th
CSI International Symposium on Artificial Intelligence and Signal Processing (AISP 2012), May 2012, pp. 007–011, doi:
10.1109/AISP.2012.6313708.
[28] X. S. Yang, “Firefly algorithm, stochastic test functions and design optimisation,” International Journal of Bio-Inspired
Computation, vol. 2, no. 2, p. 78, 2010, doi: 10.1504/IJBIC.2010.032124.
[29] S. Arora and S. Singh, “The Firefly Optimization Algorithm: Convergence Analysis and Parameter Selection,” International
Journal of Computer Applications, vol. 69, no. 3, pp. 48–52, May 2013, doi: 10.5120/11826-7528.
[30] X.-S. Yang, “Firefly Algorithms for Multimodal Optimization,” 2009, pp. 169–178.
[31] N. S. M. Raja, K. S. Manic, and V. Rajinikanth, “Firefly Algorithm with Various Randomization Parameters: An Analysis,” 2013,
pp. 110–121.
[32] D.-K. Bui, T. N. Nguyen, T. D. Ngo, and H. Nguyen-Xuan, “An artificial neural network (ANN) expert system enhanced with the
electromagnetism-based firefly algorithm (EFA) for predicting the energy consumption in buildings,” Energy, vol. 190, p. 116370,
Jan. 2020, doi: 10.1016/j.energy.2019.116370.
[33] J. Zupan, “Introduction to artificial neural network (ANN) methods: what they are and how to use them,” Acta Chimica Slovenica,
vol. 41, pp. 327–327, 1994.
[34] P. Maind, S. B., and Wankar, “Research Paper on Basic of Artificial Neural Network,” International Journal on Recent and
Innovation Trends in Computing and Communication, vol. 2, no. 1, pp. 96–100, 2014.
[35] S. Agatonovic-Kustrin and R. Beresford, “Basic concepts of artificial neural network (ANN) modeling and its application in
pharmaceutical research,” Journal of Pharmaceutical and Biomedical Analysis, vol. 22, no. 5, pp. 717–727, Jun. 2000, doi:
10.1016/S0731-7085(99)00272-1.
BIOGRAPHIES OF AUTHORS
Yashu Pulhani received her first degree from Chitkara University, Department
of Electronics and Communication Engineering, Rajpura, India. She has also Master degree
from UIET, Panjab University, Chandigarh. Her main research interests focus on FANETs,
Optimization algorithms, bio-inspired computing, artificial intelligence, combinatorial
problems, data mining, IoT, and graphs. She can be contacted at email:
yashupulhani67@gmail.com.

 ISSN: 2252-8776
158
Ankur Singh Kang working as Asstt Prof (ECE) at Departement UIET, Faculty
of Engineering and Technology, Panjab University, Chandigarh since 2009. He did his
Bachelors in Technology from GNDU Amritsar in 2007 followed by Masters in Engg in 2009
from UIET Panjab University Chandigarh. After serving Dr. BR Ambedkar National Institute
of Technology Jalandhar in 2009 he joined Deptt UIET, PU SSGRC Centre in the
Departement of ECE. He has 43 publications in International Conferences and Journals of
Repute including Springer, Elseveir & Taylor Francis. He has eight publications in IEEE
International Conferences held in India and abroad. He has been the Reviewer of Several
Science Citation Indexed International Journals: i) IET Signal.Processing, UK, ii)
International Journal of Electronic Letters (Taylor-Francis), iii) International Journal of
Engg, Design & Technology (SAGE Publications), iv) The Computer Journal-OXFORD
Univ Press, UK, v) IEEE Access, vi) IET-Radar, Sonar & Navigation, vii) IET
Communication, viii) Current Analysis on Communication Engineering, Mesford Journals,
Canada, ix) Telecommunication Systems Springer. He participated in India-France
Technology Summit in 2013 and is the Recipient of Research Award with Certificate from
Hon’ble Vice Chancellor. Dr. Arun Kumar Grover in Annual Degree Award Ceremony
presided over by Dr. Harsh Vardhan, Hon’ble Minister of Science &Technology and Earth
Sciences, Govt of India on 8 May 2016 in recognition of published research paper title
Comparative Performance Evaluation of Modified Prototype Filter Bank Multi Carrier
Cognitive Radio under Constraints of Lp, K, N and D in SCI/SCIE indexed journal, The
Computer Journal, Oxford University Press, UK in year 2016. He has a total Experience of
13 years in Teaching, Administration & Research till date. His research areas of interest
include wireless mobile communication & assisted technologies, digital signal processing,
(signal processing in communication), digital communication, cognitive radio, simulation,
modeling & optimization, network management, and future ICT & Information Theory. He
holds the membership of various Societies Miete (New Delhi) Miaeng (Hongkong), Misa
(North Carolina), Miacsit (Singapore) Msdiwc (USA), Miaoe (Austria), Mwaset (Turkey),
Miseis (Canada) Misoc (Switzerland), and Mbsa (Uk). He can be contacted at
Askang_85@pu.ac.in, Askang_85@yahoo.co.in.
Dr. Vishal Sharma is presently working at the Institute of Forensic Science &
Criminology (IFSC), Panjab University, Chandigarh (INDIA). He is currently heading the
Institute at P.U. He has obtained his Ph.D. degree in Physics from Kurukshetra University,
Kurukshetra, India, and Inter-University Accelerator Centre (IUAC-an autonomous centre of
UGC, Govt. of India), New Delhi (INDIA). He is the recipient of the DAE Young scientist
research award in 2011. He is leading a research group (Forensic research and material
research) at Panjab University, Chandigarh, and has 17 years of teaching and research
experience. His current research interest is in the field of Voice Forensics, Machine learning,
Spectroscopy, Chemometrics, Questioned documents, and fingermark sensing. He has filed
02 patents in 2020 to Controller General of Patents, Design & Trade Marks, India. He is the
author of over 110 peer-reviewed scientific papers (all are Scopus and web of science
indexed) with a maximum impact factor of 12.2 with an h index of 22 and around 1250
citations. He has 03 books as Editor (Springer Nature, Elsevier, and RSC) and 08 book
chapters to his credit. He is in the editorial board of Forensic Science International: Reports
and Forensic Science International: Synergy journals of Elsevier. Recently he has been
conferred with the best researcher award by Panjab University under Promotion of University
Research and Scientific Excellence (PURSE), DST Govt. of India grant, and Dr. P.D. Sethi
National Award. He has got India Top Cited Author Award 2019 by IOP Publishing, United
Kingdom (UK) as an author of one of the top 1% of the most-cited papers from India in the
field of Materials (nanophosphor material developed for fingermark sensing), published
across the whole IOP Publishing portfolio in the past three years (2016 to 2018), using
citations recorded in Web of Science. He has guided 04 PhD and 03 post-doctorate students,
about 40 Master thesis, and currently, 05 Ph.D. students and 08 Master students are working
on their thesis under his guidance. Recently he placed in “Top 2% of Researchers in the
World” in the single year impact (2020) list. The list was published by Stanford University,
USA on 19 Oct, 2021. He can be contacted at sharmavishal05@gmail.com.

Performance analysis on self organization based clustering scheme for FANETs using K-means algorithm and firefly optimization

More Related Content

Similar to Performance analysis on self organization based clustering scheme for FANETs using K-means algorithm and firefly optimization (20)

More from IJICTJOURNAL (20)

Recently uploaded (20)

Performance analysis on self organization based clustering scheme for FANETs using K-means algorithm and firefly optimization