有限反馈SDMA中基于信干噪比的天线合并算法

多天线广播信道中基于信道量化的MISO-SDMA系统可通过增加用户端天线并采用相关的天线合并技术来有效降低反馈量的需求。该文提出了一种基于信干噪比SINR的天线合并准则，以寻求最优的合并矢量和等效信道量化值。由于综合考虑了接收信道等效功率、量化误差引起的用户间干扰和噪声等因素，改善已有合并算法中仅仅最大化接收端等效信噪比或最小化量化误差的不足。另外，该文从理论上推导分析了在信噪比趋于零和趋于无穷时该算法的极限情形，证明已有的最大比合并和基于量化的合并方法是该文所提算法在极限情况下的特例。计算机仿真表明在相同的反馈量下该文所提算法有着优于二者的性能。

SINR-Based Antenna Combining for Limited Feedback SDMA

The channel feedback requirement of the MISO-SDMA using Zero-Forcing (ZF) precoding in multi- antenna broadcast channels can be considerably reduced by employing more antennas at user terminals and using the related antenna combining technique. In this paper, a combining scheme based on SINR maximization is proposed to jointly design the optimal combining vector and the quantization of the effective channel vector. Different from the existing combining schemes, the proposed scheme takes into account of both the effective received SNR and the inter-user interferences from the channel quantization. Also, the existing schemes of MRC and quantization-based combining (QBC) are proved that the particular cases if the SNR tends to zero and to infinite, respectively. Simulation results show that the proposed scheme outperforms the above two algorithms in terms of sum capacity, with the same feedback bits.