A Study on Intelligent and Lightweight Intrusion Detection for Wireless Sensor Networks Using Machine Learning
- Version
- Download 4
- File Size 270.27 KB
- File Count 1
- Create Date 3 January 2026
- Last Updated 3 January 2026
A Study on Intelligent and Lightweight Intrusion Detection for Wireless Sensor Networks Using Machine Learning
Pradeepaa V1, Dr T Deepa2
Pradeepaa V, Research Scholar, KPR College of Arts Science and Research, Coimbatore
Dr T Deepa, Associate Professor and Head, Department of Computer Science, School of Computing Science, KPR College of Arts Science and Research, Coimbatore
Abstract
Wireless Sensor Networks (WSNs) are extensively utilized in vital applications like smart infrastructure, industrial automation, military surveillance, and healthcare monitoring. WSNs are extremely susceptible to security threats and intrusions because of their distributed architecture, wireless connectivity, and deployment in hostile areas. Because WSNs have limited energy, memory and compute capabilities, traditional security techniques are frequently inappropriate [9]. By spotting malicious activity on the network ,intrusion detection systems(IDS) offer an efficient second line of security. Machine Learning (ML) methods have recently demonstrated a great deal of promise for increasing the precision and flexibility of IDS in WSNs. A complete machine learning-based intrusion detection system for wireless sensor networks is presented in this research. To identify routing assaults, denial-of-service attacks, and data integrity breaches, the suggested IDS makes use of supervised and unsupervised learning methods. In comparison to conventional rule-based IDS techniques, simulation results show that the suggested ML-based IDS delivers greater detection accuracy, reduced false alarm rates, and enhanced energy efficiency.
Key Words— Wireless Sensor Networks, Intrusion Detection System, Machine Learning, Network Security, Anomaly Detection
Download