一种基于异构随机矢量的波束赋形方法

A BEAMFORMING SCHEME BASED ON HETEROGENEOUS RANDOM VECTOR

为克服迭代波束赋形方法在低信噪比下性能差的问题,针对毫米波MIMO系统提出一种基于异构随机矢量的波束赋形方法。通过在发射机和接收机之间来回发送异构随机矢量,所提出方法能在低复杂度下获得信道自相关矩阵的估计。进而基带处理器可通过奇异值分解的经典幂方法来快速获得近似最优的发送波束赋形矢量。而接收机可通过相同的过程获得对应的接收波束赋形矢量。所提出的方法采用高斯随机变量和正负随机变量所构成的异构随机矢量,能在理论上消除加性高斯白噪声的影响从而获得近似最优等效信道增益。仿真结果表明,所提出的方法有渐进最优的信道增益,在低信噪比下显著优于迭代方法。因此在保密通信等高功率波束训练不允许的场景下具有实用意义。

In order to overcome the bad performance of iterative beamforming scheme at low SNR, a heterogeneous- random-vector-based beamforming scheme for Millimeter-Wave （mm-Wave） MIMO systems is proposed. By transmitting random vectors back and forth between a transmitter （TX） and receiver （RX）, the proposed scheme can obtain the estimation of channel auto-correlation with low complexity operations. Then the baseband processor can carry out the classical power method of singular value decomposition （SVD） to obtain the nearly optimal transmit beamforming vector quickly. And RX can acquire the corresponding receive beamforming vector by the same process. The proposed scheme utilizes the heterogeneous random vector consisting of a Gaussian random variable and a binary random variable, which can reduce the impact of additive white Gaussian noise （AWGN） in theory to achieve nearly optimal equivalent channel gain. From the simulation results, we can see that our scheme has asymptotic optimal channel gain, and outperforms the iterative method at low SNR. Therefore, it is of practical significance in the scenarios where high-power beamforming training secret communication occasion is prohibited.