A Dual-Perspective Self-Supervised IoT Intrusion Detection Method Based on Topology Reconstruction and Feature Perturbation

The rapid growth of the Internet of Things has driven intelligent advancements across various fields. However, the proliferation of IoT devices and diverse network interactions also introduces significant security challenges. As a critical technology for securing IoT, intrusion detection systems aim...

Full description

Saved in:
Bibliographic Details
Main Authors: Ruisheng Li, Huimin Shen, Qilong Zhang
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10902014/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850027839887245312
author Ruisheng Li
Huimin Shen
Qilong Zhang
author_facet Ruisheng Li
Huimin Shen
Qilong Zhang
author_sort Ruisheng Li
collection DOAJ
description The rapid growth of the Internet of Things has driven intelligent advancements across various fields. However, the proliferation of IoT devices and diverse network interactions also introduces significant security challenges. As a critical technology for securing IoT, intrusion detection systems aim to identify potential threats by analyzing network traffic features. Yet, traditional models struggle to capture the complex topological structures in IoT environments, and their training often relies heavily on large amounts of labeled data, making them unsuitable for IoT settings where massive data is continually generated. This paper presents a graph-based framework leveraging self-supervised learning to address these challenges. This framework employs a graph encoder to capture topological information and generate graph embeddings, utilizing topology reconstruction and feature perturbation to create diverse loss graphs that enhance feature learning. Additionally, the custom adaptive cosine loss function dynamically adjusts the loss weights of different samples and utilizes cosine similarity to align the embeddings of real and predicted graphs, thereby maximizing mutual information between local embeddings of real graphs and global summaries. The model was evaluated on four public datasets, and the experimental results demonstrate that the proposed method significantly improves detection performance. It outperforms existing self-supervised graph neural network models, validating the applicability of this approach in settings with unlabeled data.
format Article
id doaj-art-5197b5ef70b446dfbe13d69a9a4389a4
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-5197b5ef70b446dfbe13d69a9a4389a42025-08-20T03:00:01ZengIEEEIEEE Access2169-35362025-01-0113372903730110.1109/ACCESS.2025.354527810902014A Dual-Perspective Self-Supervised IoT Intrusion Detection Method Based on Topology Reconstruction and Feature PerturbationRuisheng Li0Huimin Shen1https://orcid.org/0009-0008-3459-2004Qilong Zhang2https://orcid.org/0009-0007-5774-769XSchool of Artificial Intelligence, Gansu University of Political Science and Law, Lanzhou, ChinaSchool of Cyberspace Security, Gansu University of Political Science and Law, Lanzhou, ChinaSchool of Cyberspace Security, Gansu University of Political Science and Law, Lanzhou, ChinaThe rapid growth of the Internet of Things has driven intelligent advancements across various fields. However, the proliferation of IoT devices and diverse network interactions also introduces significant security challenges. As a critical technology for securing IoT, intrusion detection systems aim to identify potential threats by analyzing network traffic features. Yet, traditional models struggle to capture the complex topological structures in IoT environments, and their training often relies heavily on large amounts of labeled data, making them unsuitable for IoT settings where massive data is continually generated. This paper presents a graph-based framework leveraging self-supervised learning to address these challenges. This framework employs a graph encoder to capture topological information and generate graph embeddings, utilizing topology reconstruction and feature perturbation to create diverse loss graphs that enhance feature learning. Additionally, the custom adaptive cosine loss function dynamically adjusts the loss weights of different samples and utilizes cosine similarity to align the embeddings of real and predicted graphs, thereby maximizing mutual information between local embeddings of real graphs and global summaries. The model was evaluated on four public datasets, and the experimental results demonstrate that the proposed method significantly improves detection performance. It outperforms existing self-supervised graph neural network models, validating the applicability of this approach in settings with unlabeled data.https://ieeexplore.ieee.org/document/10902014/Intrusion detectiongraph neural networksself-supervised learningEE-GraphSAGE
spellingShingle Ruisheng Li
Huimin Shen
Qilong Zhang
A Dual-Perspective Self-Supervised IoT Intrusion Detection Method Based on Topology Reconstruction and Feature Perturbation
IEEE Access
Intrusion detection
graph neural networks
self-supervised learning
EE-GraphSAGE
title A Dual-Perspective Self-Supervised IoT Intrusion Detection Method Based on Topology Reconstruction and Feature Perturbation
title_full A Dual-Perspective Self-Supervised IoT Intrusion Detection Method Based on Topology Reconstruction and Feature Perturbation
title_fullStr A Dual-Perspective Self-Supervised IoT Intrusion Detection Method Based on Topology Reconstruction and Feature Perturbation
title_full_unstemmed A Dual-Perspective Self-Supervised IoT Intrusion Detection Method Based on Topology Reconstruction and Feature Perturbation
title_short A Dual-Perspective Self-Supervised IoT Intrusion Detection Method Based on Topology Reconstruction and Feature Perturbation
title_sort dual perspective self supervised iot intrusion detection method based on topology reconstruction and feature perturbation
topic Intrusion detection
graph neural networks
self-supervised learning
EE-GraphSAGE
url https://ieeexplore.ieee.org/document/10902014/
work_keys_str_mv AT ruishengli adualperspectiveselfsupervisediotintrusiondetectionmethodbasedontopologyreconstructionandfeatureperturbation
AT huiminshen adualperspectiveselfsupervisediotintrusiondetectionmethodbasedontopologyreconstructionandfeatureperturbation
AT qilongzhang adualperspectiveselfsupervisediotintrusiondetectionmethodbasedontopologyreconstructionandfeatureperturbation
AT ruishengli dualperspectiveselfsupervisediotintrusiondetectionmethodbasedontopologyreconstructionandfeatureperturbation
AT huiminshen dualperspectiveselfsupervisediotintrusiondetectionmethodbasedontopologyreconstructionandfeatureperturbation
AT qilongzhang dualperspectiveselfsupervisediotintrusiondetectionmethodbasedontopologyreconstructionandfeatureperturbation