| With the rapid advancement of wireless communication and ever-increasing mobile applications, securing wireless communication becomes increasingly important and challenging. Compared to a wired network, openness of wireless link makes it vulnerable to eavesdropping, tamper and forgery attacks. To ensure users’ security in a wireless network, it is crucial to accurately authenticate their identities. In recent years, there are increased interests in developing wireless physical layer authentication by exploiting the unique nature of wireless physical layer resources.The physical layer authentication exploits randomness and reciprocity features of wireless fading channel to authenticate users’ identities, and it combines the upper network security protocols to enhance the wireless network security. In this paper, we focus on the following physical layer authentication schemes: PHY-PCRAS and PHY-CRAM without a relay node, PHY-CRAMR and PHY-AUR with a relay node.The main contributions of this paper can be summarized as follows:(1) The performance of PHY-PCRAS with MPSK modulation is investigated in detail. It is shown that when the number of sub-carriers is fixed, better performance can be achieved with the use of higher order modulation schemes, and when the key length is fixed, better performance can be achieved with the use of lower order modulation schemes.(2) The effect of correlation among sub-channels in OFDM transmission on the PHY-PCRAS authentication performance is also investigated. It is shown that the correlation decreases exponentially with the increases of sub-channel space, and the smaller the correlation, the better the authentication performance.(3) Theoretically, we show that the successful authentication rate and false acceptance rate of PHY-CRAM can be calculated by Marcum Q function. With long key, the probability density function of verification metric for each hypothesis can be well approximated by the Rice distribution.(4) The performance of PHY-CRAMR and PHY-AUR is evaluated by ROC, it is shown that the performance in the Rural channel is better than in the Urban channel, and higher SNR results in better performance for both channel models, and the impact of key length on the performance of PHY-CRAMR is not significant, but the performance of PHY-AUR improves with increased key length. |
