Enhancing intrusion detection in wireless sensor networks using a Tabu search based optimized random forest

Abstract Intrusion detection in Wireless Sensor Networks (WSNs) is an emerging area of research, given their extensive use in sensitive fields like military surveillance, healthcare, environmental monitoring, and smart cities. However, WSNs face several security challenges due to their limited compu...

Full description

Saved in:
Bibliographic Details
Main Authors: Vivek Kumar Pandey, Shiv Prakash, Tarun Kumar Gupta, Priyanshu Sinha, Tiansheng Yang, Rajkumar Singh Rathore, Lu Wang, Sabeen Tahir, Sheikh Tahir Bakhsh
Format: Article
Language:English
Published: Nature Portfolio 2025-05-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-03498-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850243445624406016
author Vivek Kumar Pandey
Shiv Prakash
Tarun Kumar Gupta
Priyanshu Sinha
Tiansheng Yang
Rajkumar Singh Rathore
Lu Wang
Sabeen Tahir
Sheikh Tahir Bakhsh
author_facet Vivek Kumar Pandey
Shiv Prakash
Tarun Kumar Gupta
Priyanshu Sinha
Tiansheng Yang
Rajkumar Singh Rathore
Lu Wang
Sabeen Tahir
Sheikh Tahir Bakhsh
author_sort Vivek Kumar Pandey
collection DOAJ
description Abstract Intrusion detection in Wireless Sensor Networks (WSNs) is an emerging area of research, given their extensive use in sensitive fields like military surveillance, healthcare, environmental monitoring, and smart cities. However, WSNs face several security challenges due to their limited computational capabilities and energy constraints. Their deployment in open, unattended environments makes them especially vulnerable to threats like eavesdropping, interference, and jamming. To address this problem, Random Forest (RF) is a popular machine learning model. The RF model can be tweaked because of its multiple hyperparameters. Tuning these parameters manually is tedious, as the combinations will be exponential. This work presents an enhanced intrusion detection approach by integrating Tabu Search (TS) optimization with a RF classifier. As a result, TS will help RF automatically search optimal hyperparameters and improve the generalization ability. This work integrates the pros of TS with RF. The model was tested on three different datasets, i.e., (a) the WSN-DS dataset, (b) CICIDS 2017, and (c) the CIC-IoT 2023 dataset, which shows better results on different metrics like precision, recall, F1-score, Cohen’s Kappa, and ROC AUC. Detection of Blackhole and Gray Hole attacks also improved, demonstrating the effectiveness of combining metaheuristic optimization with ensemble learning for stronger WSN security.
format Article
id doaj-art-b25f69a5fd6541b49f2a7939c3cdc9d2
institution OA Journals
issn 2045-2322
language English
publishDate 2025-05-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-b25f69a5fd6541b49f2a7939c3cdc9d22025-08-20T02:00:00ZengNature PortfolioScientific Reports2045-23222025-05-0115112110.1038/s41598-025-03498-3Enhancing intrusion detection in wireless sensor networks using a Tabu search based optimized random forestVivek Kumar Pandey0Shiv Prakash1Tarun Kumar Gupta2Priyanshu Sinha3Tiansheng Yang4Rajkumar Singh Rathore5Lu Wang6Sabeen Tahir7Sheikh Tahir Bakhsh8Department of Electronics and Communication, University of AllahabadDepartment of Electronics and Communication, University of AllahabadDepartment of Computer Science, Miranda House, University of DelhiDepartment of Electronics and Communication, University of AllahabadUniversity of South Wales, Llantwit RdDepartment of Computer Science, Cardiff School of Technologies, Cardiff Metropolitan UniversityXi’an Jiaotong-Liverpool UniversityDepartment of Computer Science, Cardiff School of Technologies, Cardiff Metropolitan UniversityDepartment of Computer Science, Cardiff School of Technologies, Cardiff Metropolitan UniversityAbstract Intrusion detection in Wireless Sensor Networks (WSNs) is an emerging area of research, given their extensive use in sensitive fields like military surveillance, healthcare, environmental monitoring, and smart cities. However, WSNs face several security challenges due to their limited computational capabilities and energy constraints. Their deployment in open, unattended environments makes them especially vulnerable to threats like eavesdropping, interference, and jamming. To address this problem, Random Forest (RF) is a popular machine learning model. The RF model can be tweaked because of its multiple hyperparameters. Tuning these parameters manually is tedious, as the combinations will be exponential. This work presents an enhanced intrusion detection approach by integrating Tabu Search (TS) optimization with a RF classifier. As a result, TS will help RF automatically search optimal hyperparameters and improve the generalization ability. This work integrates the pros of TS with RF. The model was tested on three different datasets, i.e., (a) the WSN-DS dataset, (b) CICIDS 2017, and (c) the CIC-IoT 2023 dataset, which shows better results on different metrics like precision, recall, F1-score, Cohen’s Kappa, and ROC AUC. Detection of Blackhole and Gray Hole attacks also improved, demonstrating the effectiveness of combining metaheuristic optimization with ensemble learning for stronger WSN security.https://doi.org/10.1038/s41598-025-03498-3Wireless sensor networks (WSNs)Intrusion detectionWSN-DSTabu searchRandom forestOptimization
spellingShingle Vivek Kumar Pandey
Shiv Prakash
Tarun Kumar Gupta
Priyanshu Sinha
Tiansheng Yang
Rajkumar Singh Rathore
Lu Wang
Sabeen Tahir
Sheikh Tahir Bakhsh
Enhancing intrusion detection in wireless sensor networks using a Tabu search based optimized random forest
Scientific Reports
Wireless sensor networks (WSNs)
Intrusion detection
WSN-DS
Tabu search
Random forest
Optimization
title Enhancing intrusion detection in wireless sensor networks using a Tabu search based optimized random forest
title_full Enhancing intrusion detection in wireless sensor networks using a Tabu search based optimized random forest
title_fullStr Enhancing intrusion detection in wireless sensor networks using a Tabu search based optimized random forest
title_full_unstemmed Enhancing intrusion detection in wireless sensor networks using a Tabu search based optimized random forest
title_short Enhancing intrusion detection in wireless sensor networks using a Tabu search based optimized random forest
title_sort enhancing intrusion detection in wireless sensor networks using a tabu search based optimized random forest
topic Wireless sensor networks (WSNs)
Intrusion detection
WSN-DS
Tabu search
Random forest
Optimization
url https://doi.org/10.1038/s41598-025-03498-3
work_keys_str_mv AT vivekkumarpandey enhancingintrusiondetectioninwirelesssensornetworksusingatabusearchbasedoptimizedrandomforest
AT shivprakash enhancingintrusiondetectioninwirelesssensornetworksusingatabusearchbasedoptimizedrandomforest
AT tarunkumargupta enhancingintrusiondetectioninwirelesssensornetworksusingatabusearchbasedoptimizedrandomforest
AT priyanshusinha enhancingintrusiondetectioninwirelesssensornetworksusingatabusearchbasedoptimizedrandomforest
AT tianshengyang enhancingintrusiondetectioninwirelesssensornetworksusingatabusearchbasedoptimizedrandomforest
AT rajkumarsinghrathore enhancingintrusiondetectioninwirelesssensornetworksusingatabusearchbasedoptimizedrandomforest
AT luwang enhancingintrusiondetectioninwirelesssensornetworksusingatabusearchbasedoptimizedrandomforest
AT sabeentahir enhancingintrusiondetectioninwirelesssensornetworksusingatabusearchbasedoptimizedrandomforest
AT sheikhtahirbakhsh enhancingintrusiondetectioninwirelesssensornetworksusingatabusearchbasedoptimizedrandomforest