Abstract—Wireless Sensor Networks (WSNs) face persistent security challenges owing to their resource-limited nodes and their exposure to sophisticated cyber-attacks. This paper presents and evaluates an ensemble learning-based Intrusion Detection System (IDS) that integrates Particle Swarm Optimization (PSO) for feature selection with a majority voting classification scheme. The proposed framework applies Smote for class balancing and PSO-driven feature selection, followed by weighted majority voting over six base classifiers: Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Logistic Regression (LR), Random Forest (RF), Naïve Bayes (NB), and Decision Tree (DT). Experiments conducted on four benchmark datasets, Wireless Sensor Networks-Dataset (WSN-DS), full name: A Dataset for Intrusion Detection Systems in Wireless Sensor Networks, Network Security Laboratory-Knowledge Discovery and Data Mining (NSL-KDD), University of New South Wales-Network Behavior 2015 (UNSW-NB15), and Internet of Things-23 (IoT-23), yield classification accuracies ranging from 95.73% to 98.23% under 10‒fold cross-validation, with 95% confidence intervals of ±0.053%. The system achieves sub-millisecond inference (0.49 ms), scales linearly with dataset size, and consumes only 108.3 μJ per classification, corresponding to approximately 30 days of battery operation. Comparative experiments confirm that the proposed method achieves a superior balance among accuracy, computational efficiency, and energy awareness relative to deep learning alternatives, making it particularly well suited for deployment on resource-constrained WSN nodes.
 
Keywords—wireless sensor networks, intrusion detection system, ensemble learning, particle swarm optimization, joint voting, feature selection, energy efficiency

Cite: Ouhmi Said, Abdelkarim Ait Temghart, Mbarek Marwan, and Housni Khalid , “Extended Analysis of PSO-Optimized Majority Voting for Wireless Sensor Network (WSN) Intrusion Detection: Cross-Dataset Validation and Deployment Considerations ," Journal of Communications, vol. 21, no. 4, pp. 493-507, 2026.

Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).