Privacy-Preserving Physical-Layer Authentication for Non-Orthogonal Multiple Access Systems Article

Xie, N, Zhang, Q, Chen, J et al. (2022). Privacy-Preserving Physical-Layer Authentication for Non-Orthogonal Multiple Access Systems . 40(4), 1371-1385. 10.1109/JSAC.2022.3143222

cited authors

  • Xie, N; Zhang, Q; Chen, J; Tan, H

fiu authors

abstract

  • This paper addresses the problems of both authentication and privacy in a Non-Orthogonal Multiple Access (NOMA) system by using a privacy-preserving Physical-Layer Authentication (PLA) scheme. The prior scheme has the following limitations: no privacy, low authentication performance, management problem of different secret keys, and constraint to the two-user scenario. In this paper, we propose two privacy-preserving PLA schemes: the Privacy-Preserving PLA (PP-PLA) scheme and the Enhanced Privacy-Preserving PLA (EPP-PLA) scheme, which improve both the privacy and authentication performance as compared to the prior scheme. The fundamental difference between the EPP-PLA and PP-PLA schemes is how to encrypt the source message of a certain user while the other operations of the two proposed schemes keep the same. We provide the theoretical analysis of the proposed schemes in terms of robustness, compatibility, privacy, and security, and derive their closed-form expressions. Moreover, we provide the theoretical comparisons between the proposed schemes and the prior scheme. We optimize the parameters of the proposed schemes to achieve both message-rate fairness and authentication-accuracy fairness. We implement the proposed schemes and conduct extensive performance comparisons through simulations. Experimental results show that the theoretical results perfectly match the corresponding simulation results. Specifically, we take the second user, acting as an administrator for a NOMA system with three users, as an example, where SNR = 16 dB. In comparison with the prior scheme, both proposed schemes have the same compatibility, whereas they improve the robustness by 4.72%. Moreover, the EPP-PLA and PP-PLA schemes improve the security by 32.8%, whereas they improve the privacy by 3000.75% and 1000.75%, respectively.

publication date

  • April 1, 2022

Digital Object Identifier (DOI)

start page

  • 1371

end page

  • 1385

volume

  • 40

issue

  • 4