Secure Hybrid Analog and Digital Beamforming for mmWave XR Communications With Mixed-DAC

Abstract

To achieve a balance between performance and implementation complexity in extended reality (XR)-aided millimeter wave (mmWave) communication, secure hybrid digital and analog (HDA) beamforming with mixed digital-to-analog converters (DACs) is established by partially replacing costly full-resolution DACs with some cheap low-resolution DACs. We focus on secure HDA beamforming for such a system. Furthermore, XR technology is aided to improve the operating efficiency in this complex scenario. First, a closed-form approximation of the average secrecy rate (ASR) is derived. To maximize ASR with partial eavesdropping channel knowledge available, we propose an algorithm of skillfully utilizing the alternating iteration to design beamformers of analog, confidential message (CM) and artificial noise (AN). Given the analog and CM/AN beamformers, the updated AN/CM beamformer is addressed by a gradient descent algorithm. Then, given the beamformers of CM and AN, Dinkelbach and Majorization-Minimization are combined to optimize analog beamformer. Simulation results show that the proposed algorithm achieves much better ASR performance than existing methods in the medium and high signal-to-noise ratio regions.