Intelligence-Driven Leader Selection in PEGASIS: A Data-Driven Machine Learning Framework for Sustainable and Secure Wireless Sensor Networks
Electronics · 16.04.2026 · 15 (8) · 1686
DOI: 10.3390/electronics15081686 ↗Kurz gefasst
Fachartikel in der Zeitschrift Electronics, online erschienen am 16.04.2026. Verfasst von Abdulla Juwaied und 1 weiteren Autor:innen. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Institute of Applied Computer Science, Lodz University of Technology.
Abstract (Originalsprache)
Energy-efficient routing is critical for extending the operational lifespan of wireless sensor networks (WSNs). While the Power-Efficient Gathering in Sensor Information Systems (PEGASIS) protocol achieves high efficiency through chain-based data aggregation, its standard round-robin leader selection fails to account for dynamic node factors, such as residual energy and historical reliability. This often leads to premature energy depletion and network instability. To address these limitations, this paper proposes K-NN-PEGASIS, a data-driven machine learning framework that utilises a weighted k-nearest neighbours (K-NN) algorithm for intelligent leader selection. By processing a normalised feature vector comprising residual energy, distance to the base station (BS), node degree, and historical performance, the framework adaptively identifies optimal leaders in each round. Simulations conducted in MATLAB for networks ranging from 100 to 1000 nodes demonstrate that K-NN-PEGASIS improves network lifetime by up to 47.3% and reduces total energy dissipation by 52.8% compared to baseline algorithms. Furthermore, the framework provides passive resilience against routing attacks, reducing the selection of malicious leaders by 96% and maintaining a 32.3% higher packet delivery ratio under attack scenarios.
Förderung
- Institute of Applied Computer Science, Lodz University of Technology
Zitieren
APA
Juwaied, A., & Romanowski, A. (2026). Intelligence-Driven Leader Selection in PEGASIS: A Data-Driven Machine Learning Framework for Sustainable and Secure Wireless Sensor Networks. Electronics, 15(8), 1686. https://doi.org/10.3390/electronics15081686
Vancouver
Juwaied A, Romanowski A. Intelligence-Driven Leader Selection in PEGASIS: A Data-Driven Machine Learning Framework for Sustainable and Secure Wireless Sensor Networks. Electronics. 2026;15(8):1686. doi:10.3390/electronics15081686
BibTeX
@article{Juwaied2026,
author = {Juwaied, Abdulla and Romanowski, Andrzej},
title = {Intelligence-Driven Leader Selection in PEGASIS: A Data-Driven Machine Learning Framework for Sustainable and Secure Wireless Sensor Networks},
journal = {Electronics},
year = {2026},
volume = {15},
number = {8},
pages = {1686},
doi = {10.3390/electronics15081686}
}
EUVIMED