Effectual Clustering and Node Placement with Differential Evolution Particle Swarm Optimization using Markov Chain Clustering in FANET
B. Mahalakshmi1, S.Ranjitha Kumari2

1B. Mahalakshmi*,  Research Scholar, School of Computer Studies, Rathnavel Subramaniam College of Arts and Science, Sulur, Coimbatore.
2Dr.S.Ranjitha Kumari,  Associate Professor, School of Computer Studies, Rathnavel Subramaniam College of Arts and Science, Sulur, Coimbatore.
Manuscript received on September 12, 2019. | Revised Manuscript received on October 15, 2019. | Manuscript published on October 30, 2019. | PP: 3927-3934 | Volume-9 Issue-1, October 2019 | Retrieval Number: A1463109119/2019©BEIESP | DOI: 10.35940/ijeat.A1463.109119
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Flying ad hoc network (FANET) comprises of multiple unmanned aerial vehicles (UAVs) which is effectual technology for future generation. In this investigation, the specific way for constructing a FANET topology using clustering technique to achieve end-to-end communication is elaborated. For this purpose, an application that uses the meta-heuristics approach for cluster analysis is anticipated. Specifically, the parameters of differential evolution (DE) and particle swarm optimization (PSO) have gained the attention and extensive popularity in various communities based on its working effectiveness in resolving complex combinational optimization crisis. Thus, hybrid model of DE and PSO based Markov Chain Clustering Model (MCCM) is designed in this investigation to analyse the problems of clustering in FANET and reliability parameters are examined. The proposed (DEPSO-MCM) model is to enhance search capability and to attain superior flexibility in forming nodes cluster. Empirical outcomes demonstrate DEPSOMCM based clustering algorithm attains superior performance in number of epochs to acquire fitness function effectually. The simulation was carried out in NS-2 simulator, the outcomes based on the simulation shows that the proposed method works effectually and shows better trade-off than the existing techniques to provide a meaningful clustering.
Keywords: Unsupervised classification, Particle swarm optimization, Differential evolution, Markov Chain Model, Reliability.