摘要
分布式大规模多输入多输出(MIMO)系统中,利用时分双工模式的信道互易特性,发送端的接入节点可以根据估计得到的上行信道状态信息来进行下行发送预编码的设计.然而,完整的通信信道还包含收发两端的射频电路.而射频增益的失配,破坏了通信信道的互易性,降低了系统的性能.本文研究分布式大规模MIMO系统中的互易性校准问题,接入节点和用户均配置多根天线,系统采用块对角化预编码.分析了在射频增益失配时系统的可达速率,推导了总体最小二乘算法的优化目标函数表达式.同时为了避免总体最小二乘算法需要特征值分解所带来的较高复杂度,提出了迭代坐标下降校准算法.理论分析和仿真结果表明,本文所提迭代坐标下降校准算法基本达到了总体最小二乘算法的性能,收敛速度快,大大降低了实现的复杂度.
Large-scale multiple-input multiple-output（MIMO） is a promising technique for next generation wireless communication systems.Thanks to channel reciprocity,the time-division duplexing operation is usually used for distributed large-scale MIMO systems to perform downlink precoding by uplink channel estimation.However,the entire channel is composed of not only the wireless propagation channel but also the transceiver radio frequency circuits at both sides of the link.The channel reciprocity is often jeopardized by nonsymmetric transceiver radio frequency circuits.This paper is focused on the reciprocity calibration for the distributed large-scale MIMO systems with block diagonalization precoding,where both access points（APs） and user equipment（UE） have multiple antennas.Theoretical analysis and simulation results show that radio frequency mismatches at the UEs still cause severe interstream interference and significant performance loss with perfect partial calibration at the APs.Then,in order to avoid using an eigenvalue decomposition operation of the total least squares method to perform the full calibration,an iterative algorithm named the iterative coordinate descent method is proposed to significantly reduce the complexity and essentially achieve the performance of the total least squares method.
出处
《中国科学：信息科学》
CSCD
北大核心
2016年第1期61-74,共14页
Scientia Sinica(Informationis)
基金
国家重点基础研究发展计划(973计划)(批准号:2013CB336600)
国家自然科学基金(批准号:61221002
61271205)
国家高技术研究发展计划(863计划)(批准号:2014AA01A706)
江苏省普通高校研究生科研创新计划(批准号:KYLX15 0075)资助项目
关键词
互易性校准
大规模MIMO
坐标下降
总体最小二乘
块对角化
reciprocity calibration
large-scale MIMO
coordinate descent
total least squares
block diagonalization
