IoT device identification method based on CNN+SVM

Shan Liu; Ziqiang Zhou; Yao Wang; Shaobo Wang; Shu Yang

doi:10.1117/12.2639694

19 October 2022 IoT device identification method based on CNN+SVM

Shan Liu, Ziqiang Zhou, Yao Wang, Shaobo Wang, Shu Yang

Proceedings Volume 12294, 7th International Symposium on Advances in Electrical, Electronics, and Computer Engineering; 1229437 (2022) https://doi.org/10.1117/12.2639694
Event: 7th International Symposium on Advances in Electrical, Electronics and Computer Engineering (ISAEECE 2022), 2022, Xishuangbanna, China

Abstract

The Internet of Things allows everything to be interconnected, but there is also the risk of malicious devices or vulnerable devices causing damage to the network. In order to avoid this risk, how to accurately identify which devices are connected to the network in a resource-limited IoT environment becomes an urgent problem to be solved. Identification by extracting network traffic features as device fingerprints is a more effective and low-interference method today. This paper proposes a device identification method based on device traffic fingerprints, which utilizes convolutional neural networks (CNN) to extract features from device network traffic, and then uses SVM for device classification. The experimental results show that, compared with the CNN method and the method of manually extracting features, the identification accuracy and stability of the method proposed in this paper are significantly improved.

Citation Download Citation

Shan Liu, Ziqiang Zhou, Yao Wang, Shaobo Wang, and Shu Yang "IoT device identification method based on CNN+SVM", Proc. SPIE 12294, 7th International Symposium on Advances in Electrical, Electronics, and Computer Engineering, 1229437 (19 October 2022); https://doi.org/10.1117/12.2639694

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