毫米波下行多用户系统安全混合波束成形算法

毫米波混合波束成形结构受到日益广泛的认可，但缺乏以安全性为着眼点的混合波束成形算法研究，特别是当窃听者具有多用户译码能力时，系统安全性能无法得到保障。针对上述问题，该文基于人工噪声辅助的思想提出一种毫米波下行多用户系统安全混合波束成形算法。首先，将混合波束成形矩阵的模拟部分和数字部分解耦合独立求解，在充分考虑信道特性的基础上，通过最大化用户接收信号能量和迫零思想分别设计有用信号的模拟和数字波束成形矩阵；然后，通过SVD分解设计人工噪声的基带数字预编码矩阵，将人工噪声置于合法用户零空间。仿真结果表明，人工噪声辅助的安全混合波束成形算法有效解决了存在具有多用户译码能力窃听者时系统的安全问题。

Security Hybrid Beamforming Algorithm for Millimeter Wave Downlink Multiuser System

The millimeter-wave hybrid beamforming becomes a widely accepted beamforming method in millimeter-wave systems. However there is almost no hybrid beamforming algorithm based on security. Especially when the eavesdropper has multi-user decoding capability, the system security performance can not be guaranteed. To solve this problem, a security hybrid beamforming algorithm is proposed for millimeter wave downlink multiuser system based on artificial noise. First, the analog part and the digital part of the hybrid beamforming matrix are decoupled. Based on the channel characteristics, the analog and digital beamforming matrices of useful signals are designed by maximizing the user’s received signal energy and Zero-Forcing (ZF). Then, the artificial noise baseband digital precoding matrix is designed by Singular Value Decomposition (SVD), and the artificial noise is placed in null space of the legal users and worsening eavesdropping channel. Simulation results show that the artificial noise-assisted secure hybrid beamforming algorithm solves effectively the security problem of the system when there are multi-user decoding ability eavesdroppers.