Exploiting Channel Gain and Phase Noise for PHY-Layer Authentication in Massive MIMO Systems
Pinchang Zhang, Jun Liu, Yulong Shen, Xiaohong Jiang
Abstract
By exploiting two intrinsic physical (PHY)-layer features in terms of location-specific channel gain and transmitter-specific phase noise, this paper proposes a new PHY-layer authentication scheme for massive multiple-input multiple-output (MIMO) systems. In particular, we apply the linear minimum mean square error technique to estimate the time-varying channel gain and adopt extended Kalman filtering to track the time-varying phase noise. Based on the estimation error covariance matrices of channel gain and phase noise, we then formulate the PHY-layer authentication as a composite hypothesis testing problem. With the help of tools from statistical signal processing, matrix analysis, and composite hypothesis testing, we develop theoretical models to capture the false alarm and detection probability performances of the proposed scheme. Finally, we provide extensive numerical results to validate these theoretical models and to illustrate the efficiency of the proposed authentication scheme.