Xu, Yuxuan, Liu, Ming, Peng, Linning, Zhang, Junqing 
ORCID: 0000-0002-3502-2926 and Zheng, Yawen
(2022)
Colluding RF Fingerprint Impersonation Attack Based on Generative Adversarial Network.
In: ICC 2022 - IEEE International Conference on Communications, 2022-5-16 - 2022-5-20.
|
Text
ICC2022_RFFI_Attack.pdf - Author Accepted Manuscript Download (1MB) | Preview |
Abstract
Radio frequency fingerprint (RFF) is an effective way to improve the security of wireless communications. Existing research mainly focused on the classification capability and the robustness of RFFs but overlooked malicious attacks. In this paper, a colluding impersonation attack framework is proposed to emulate the RFF of legitimate users. A colluding attacker is introduced to observe the signal features of the impersonation attacker and the legitimate user and compare their difference. The difference is fed back to the impersonation attacker to help improve its RFF impersonation method. With this idea, the impersonation attack is realized by the Generative Adversarial Network (GAN) structure. The RFF impersonation is formulated as the generator whose objective is to output the signal with RFF similar to the legitimate user, viewed from the colluding attacker's perspective. Simulation results show that the proposed method can effectively impersonate the legitimate user's RFF under the dynamic block fading channel.
| Item Type: | Conference or Workshop Item (Unspecified) |
|---|---|
| Uncontrolled Keywords: | Radio frequency fingerprint (RFF), impersonation attack, general adversarial network (GAN), adversarial machine learning |
| Divisions: | Faculty of Science and Engineering > School of Electrical Engineering, Electronics and Computer Science |
| Depositing User: | Symplectic Admin |
| Date Deposited: | 07 Feb 2022 09:59 |
| Last Modified: | 18 Jan 2023 21:14 |
| DOI: | 10.1109/ICC45855.2022.9838574 |
| Related URLs: | |
| URI: | https://livrepository.liverpool.ac.uk/id/eprint/3148337 |
Dimensions
Dimensions
