Hybrid Intrusion Detection System for RPL IoT Networks Using Machine Learning and Deep Learning

Aug 3, 2026

Upload by Violet

The Internet of Things (IoT) enables smart applications but leaves constrained devices vulnerable to security breaches. The Routing Protocol for LowPower and Lossy Networks (RPL) is valuable for IoT routing, yet it faces major security threats. This paper presents an intrusion detection system based on the ROUT-4-2023 dataset covering Black Hole, Flooding, DODAG Version Number, and Decreased Rank attacks, extracting traffic features and evaluating statistical graphs. Machine learning and deep learning models are compared using confusion-matrix performance and training efficiency, achieving 99% accuracy with Random Forest and 97% F1-score with Transformers.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Abstract

Introduction

This is a peer-reviewed, post-print (final draft post-refereeing) version of the following published document and is licensed under Creative Commons: Attribution-Noncommercial-No Derivative Works 4.0 license: Shahid, Usama ORCID: 0009-0005-6360-333X, Hussain, Muhammad Zunnurain, Hasan, Muhammad Zulkifl, Haider, Ali, Ali, Jibran and Altaf, Jawad (2024) Hybrid Intrusion Detection System for RPL IoT Networks Using Machine Learning and Deep Learning . IEEE Access, 12. pp . 113099-113112 . doi:10.1109/ACCESS.2024.3442529 Official URL: [http://doi.org/10.1109/ACCESS.2024.3442529](http://doi.org/10.1109/ACCESS.2024.3442529) DOI: [http://dx.doi.org/10.1109/ACCESS.2024.3442529](http://dx.doi.org/10.1109/ACCESS.2024.3442529) EPrint URI: [https://eprints.glos.ac.uk/id/eprint/14300](https://eprints.glos.ac.uk/id/eprint/14300) Disclaimer The University of Gloucestershire has obtained warranties from all depositors as to their title in the material deposited and as to their right to deposit such material. The University of Gloucestershire makes no representation or warranties of commercial utility, title, or fitness for a particular purpose or any other warranty, express or implied in respect of any material deposited. The University of Gloucestershire makes no representation that the use of the materials will not infringe any patent, copyright, trademark or other property or proprietary rights. The University of Gloucestershire accepts no liability for any infringement of intellectual property rights in any material deposited but will remove such material from public view pending investigation in the event of an allegation of any such infringement. PLEASE SCROLL DOWN FOR TEXT. This article has been accepted for publication in IEEE Access. This is the author's version which has not been fully edited and content may change prior to final publication. Citation information: DOI 10. 1109/ACCESS.2024.3442529 Date of publication xxxx 00, 0000, date of current version xxxx 00, 0000. Digital Object Identifier 10. 1109/ACCESS.2024.Doi Number Hybrid Intrusion Detection System for RPL IoT Networks Using Machine Learning and Deep Learning Usama Shahid1, Muhammad Zunnurain Hussain2, Muhammad Zulkifl Hasan3, Ali Haider4, Jibran Ali5, Jawad Altaf6 1 School of Business, Computing & Social Sciences, University of Gloucestershire, United Kingdom 2 Department of Computer Science, Bahria University Lahore Campus, Pakistan 3 Faculty of Information Technology, University of Central Punjab, Lahore Pakistan 4 Senior Cyber Security Consultant, Dell SecureWorks USA 5 Deputy Manager Engineering & Operations, Multinet Pakistan 6 National College of Ireland, NCI Corresponding author: Muhammad Zunnurain Hussain (e-mail: [zunnurain.bulc@bahria.edu.pk](zunnurain.bulc@bahria.edu.pk)). ABSTRACT The Internet of Things (IoT) is transforming everyday objects. However, the limited memory, processing power, and network capabilities of its devices make them susceptible to security breaches. The Routing Protocol for LowPower and Lossy Networks (RPL) is a promising IoT protocol but faces significant security challenges. Existing research often focuses on individual attacks, utilizing various mitigation strategies, including machine learning and deep learning for detection. This paper proposes an Intrusion Detection System (IDS) using the ROUT-4-2023 dataset, which encompasses Black Hole, Flooding, DODAG Version Number, and Decreased Rank attacks. The study investigates network traffic features encompassing all four attacks, utilizing statistical information graphs. Additionally, it experiments with various machine learning models and deep learning architectures for comparative analysis, focusing on confusion matrix outcomes and computational efficiency. Results indicate that Random Forest classifier achieves 99% accuracy, while Transformers reach 97% F1-Score with training time of only 16.8 minutes over 5 epochs. INDEX TERMS intru

FAQs

What is the main goal of the proposed work?

To build and evaluate an intrusion detection system for RPL-based IoT networks using machine learning and deep learning.

Which attacks and dataset are used for training and evaluation?

The study uses the ROUT-4-2023 dataset covering Black Hole, Flooding, DODAG Version Number, and Decreased Rank attacks.

How do the best-performing models compare in results?

Random Forest reaches 99% accuracy, while Transformers achieve a 97% F1-score with a training time of 16.8 minutes over 5 epochs.

Aug 3, 2026

Upload by Violet