An Efficient Intrusion Detection Method Based on Dynamic Autoencoder

The proliferation of wireless sensor networks (WSNs) and their applications has attracted remarkable growth in unsolicited intrusions and security threats, which disrupt the normal operations of the WSNs. Deep learning (DL)-based network intrusion detection (NID) methods have been widely investigate...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:IEEE Wireless Communications Letters Ročník 10; číslo 8; s. 1707 - 1711
Hlavní autori: Zhao, Ruijie, Yin, Jie, Xue, Zhi, Gui, Guan, Adebisi, Bamidele, Ohtsuki, Tomoaki, Gacanin, Haris, Sari, Hikmet
Médium: Journal Article
Jazyk:English
Japanese
Vydavateľské údaje: Piscataway IEEE 01.08.2021
Institute of Electrical and Electronics Engineers (IEEE)
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Predmet:
Artificial neural networks
autoencoder
Computational modeling
Computing costs
Deep learning
Feature extraction
intrusion detection
Intrusion detection systems
Lightweight
lightweight neural network
Machine learning
Model accuracy
Performance evaluation
Robustness (mathematics)
Training
Wireless networks
Wireless sensor networks
ISSN:2162-2337, 2162-2345
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:The proliferation of wireless sensor networks (WSNs) and their applications has attracted remarkable growth in unsolicited intrusions and security threats, which disrupt the normal operations of the WSNs. Deep learning (DL)-based network intrusion detection (NID) methods have been widely investigated and developed. However, the high computational complexity of DL seriously hinders the actual deployment of the DL-based model, particularly in the devices of WSNs that do not have powerful processing performance due to power limitation. In this letter, we propose a lightweight dynamic autoencoder network (LDAN) method for NID, which realizes efficient feature extraction through lightweight structure design. Experimental results show that our proposed model achieves high accuracy and robustness while greatly reducing computational cost and model size.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2162-2337
2162-2345
DOI:10.1109/LWC.2021.3077946