TY  - JOUR
T1  - CNN-Based Physical Layer Authentication against Impersonation Attacks
AU  - Oh, Hanol 
AU  - Yoon, Jihyeon 
AU  - Moon, Jihwan 
AU  - Kim, Taehoon 
AU  - Bang, Inkyu 
JO  - The Journal of Korean Institute of Communications and Information Sciences
PY  - 2025
DA  - 2025/1/1
DO  - 10.7840/kics.2025.50.8.1172
KW  - Physical Layer Authentication (PLA)
KW  - Convolutional Neural Network (CNN)
KW  - Impersonation Attack
KW  - Channel State Information (CSI)
KW  - IQ Imbalance
AB  - In this paper, we propose a convolutional neural network (CNN)-based physical layer authentication (PLA) scheme as a countermeasure against impersonation attacks in WLAN (Wireless Local Area Network). We evaluate the proposed PLA scheme in terms of attack detection rate through experiments. The main contributions are as follows: (1) We implement the Evil Twin attack which is one of the impersonation attacks in WLAN; (2) We implement the channel state information (CSI)-based PLA scheme and perform experiments to measure attack detection rates in both static and dynamic channel environments; (3) We finally propose the CNN-based PLA scheme exploiting the unique IQ imbalance of hardware and evaluate its performance through extensive simulations.