关键词: 多输人多输出, 正交频分复用, 垂直分层空时码, 几何均值分解, 最小均方误差

Abstract: Considering the error propagation effect and high complexity of Vertical Bell Laboratory Layered Space Time (V-BLAST), a new nonlinear module algebra precoding based on geometric mean decomposition for V-BLAST in MIMO-OFDM downlink systems was proposed. Geometric mean decomposition was used for precoding matrix that has the same equivalent noise gain among each sub-channel; the nonlinear module algebra precoding was used between sub-carrier channels of Orthogonal Frequency Division Multiplexing (OFDM) to eliminate interference from other signals at the transmitter, and could eliminate the error propagation effect of layered space-time codes effectively. At the receiver, minimum mean square error criterion was used. The simulation results show that the proposed method is better than the traditional methods in the performance of system Bit Error Rate (BER), and the complexity of the downlink receiver can be reduced to some extent.

Key words: Multiple Input Multiple Output, Orthogonal Frequency Division Multiplexing, Vertical Bell Labs Layered Space Time, Geometric Mean Decomposition, Minimum Mean Square Error