Classification of IoT intrusion detection data based on WGAN-gp and E-GraphSAGE

Chan Wang; Zixian Dong; Wei Hu; Xijun Jin; Xingjie Huang; Jin Pang; Rongliang Shi

doi:10.1117/12.3010362

20 October 2023 Classification of IoT intrusion detection data based on WGAN-gp and E-GraphSAGE

Chan Wang, Zixian Dong, Wei Hu, Xijun Jin, Xingjie Huang, Jin Pang, Rongliang Shi

Proceedings Volume 12814, Third International Conference on Green Communication, Network, and Internet of Things (CNIoT 2023); 1281416 (2023) https://doi.org/10.1117/12.3010362
Event: Third International Conference on Green Communication, Network, and Internet of Things (CNIoT 2023), 2023, Chongqing, China

Abstract

In recent years, the frequency and complexity of IoT network attacks have significantly increased. NIDS, strategically located in IoT network nodes, is an essential tool for monitoring traffic and detecting and mitigating network-based attacks. However, with the significant increase in computer network attacks, many datasets used for training suffer from imbalanced data problems. Therefore, to address the traffic characteristics of IoT networks and the issue of imbalanced data, this paper proposes an intrusion detection method that combines graph neural networks(E-GraphSAGE) and generative adversarial networks. Based on experiments using datasets NF-BoT-IoT, we found that training ML classifiers on datasets balanced with synthetic samples generated by WGAN-gp increased their prediction accuracy to 93.7% .

(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Chan Wang, Zixian Dong, Wei Hu, Xijun Jin, Xingjie Huang, Jin Pang, and Rongliang Shi "Classification of IoT intrusion detection data based on WGAN-gp and E-GraphSAGE", Proc. SPIE 12814, Third International Conference on Green Communication, Network, and Internet of Things (CNIoT 2023), 1281416 (20 October 2023); https://doi.org/10.1117/12.3010362

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available