Title: An energy-efficient ensemble clustering based on multiple disjoint and nonlinear structures for wireless sensor network

Authors: S. Sheeja Rani; K. Siva Sankar

Addresses: Department of Computer Science and Engineering, Noorul Islam University, Kanyakumari, India ' Department of Computer Science and Engineering, Noorul Islam University, Kanyakumari, India

Abstract: The evolution of small-scale, cost effective and intelligent sensors with effective communication capabilities has instigated the emergence of wireless sensor networks (WSNs). As the energy of wireless sensor nodes is restricted, efficient use of energy is necessary. Besides, load balancing is regarded as the second issue with larger sensor nodes needed to be addressed. Different clustering algorithms form clusters by imposing different structure on the data. Hence, the performance of single clustering algorithm is insufficient. This issue is addressed by an ensemble of normalised spectral cluster and separation K means clustering algorithm namely normalised spectral cluster and separation K means (NSC-SKM). NSC-SKM determines optimal sensor clusters by using the eigenvector corresponding to the subsequent infinitesimal eigenvalue of the Laplacian, to maintain balance between sensor nodes in the cluster. The energy and load performance of NSC-SKM framework is studied and the proposed framework shows significant performance gains compared to baseline solutions.

Keywords: wireless sensor networks; WSN; load balancing; normalised spectral cluster; separation K means; graph theory; similarity matrix; eigenvalue; Laplacian.

DOI: 10.1504/IJAIP.2024.141522

International Journal of Advanced Intelligence Paradigms, 2024 Vol.29 No.1, pp.1 - 16

Received: 08 Jun 2018
Accepted: 17 Nov 2018
Published online: 23 Sep 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article