大规模MIMO下最优预编码选择策略研究

大规模多入多出(Massive MIMO)系统中,随着天线数的增加,线性预编码算法的性能逐渐趋于最优,选择合适的线性预编码对系统性能具有重要的意义.针对发射端信道状态信息(Channel State Information at Transmitter,CSIT)不完美的Massive MIMO系统,推导出了迫零(Zero-Forcing,ZF)和最大比传输(Maximum Ratio Transmission,MRT)这两种常见预编码方案在向量归一化方式下的下行可达和速率下界,并给出了证明.随后对两种下界进行了分析,提出了一个关于系统用户数的阈值,当系统用户数和阈值的大小关系不同时,两种预编码性能的优劣关系也不相同.根据分析结果,文章进一步提出了一种以系统中用户数为参量的预编码选择策略,可以保证不论用户数如何变化,系统都能选择出更优的那一个预编码算法来对信号进行预处理.分析的有效性和方案的可靠性通过仿真得到了验证.     

莱斯信道下分布式大规模MIMO系统基站选择算法的研究

该文考虑一种分布式大规模MIMO系统,假设基站端与用户之间的信道为莱斯信道,研究了该系统中基站选择的算法。首先给出了系统采用匹配滤波和迫零预编码时,用户下行可达速率的闭式表达式,并分析了系统的功率效率性能。然后基于此闭式表达式,以最大化系统的频谱效率为目标,提出了基于增量选择和基于用户优先级的基站选择算法。这两种算法只需要系统获取基站端与用户之间的信道统计特征信息,从而有效降低了系统开销。仿真结果表明,所提出的两种基站选择算法性能仍能逼近最优算法。特别地,当采用匹配滤波预编码且基站端天线数趋于无穷时,基于用户优先级的基站选择算法优于基于增量选择的算法。     

多天线两小区蜂窝系统的低开销线性干扰对齐算法

在蜂窝系统中,由于干扰的存在,用户性能受到影响,特别是对于小区边缘用户,其通信质量较差。干扰对齐作为一种能够消除干扰、提高系统容量的技术,近年来受到广泛关注。针对多天线两小区蜂窝系统的边缘用户,本文给出了一种系统开销小、需要天线数少的线性干扰对齐算法。该算法利用发送端预编码矢量消除小区内干扰,利用接收端干扰抑制矢量消除小区间干扰。采用本文算法后,在每个小区有 个用户、基站有 根发送天线、用户有 根接收天线的情况下,只需 和 就可以实现上下行的干扰对齐,整个系统可以达到 的自由度,并且在下行链路中不需要小区间反馈,而在上行链路中只需要较小的小区间反馈。仿真结果表明,本文提出的算法能够以较小的开销实现比以往的算法更好的性能。      

TDD大规模MIMO系统中两种新的下行预编码方案

针对时分双工大规模MIMO系统,该文提出两种新的下行链路预编码方案,即新最大比合并(NMRC)和新迫零(NZF)方案,通过理论分析得到两种新预编码方案下目标用户和非目标用户的下行链路信干噪比和频谱效率的表达式,并与传统的迫零(ZF)和最大比合并(MRC)预编码的下行链路信干噪比、频谱效率和误码率性能进行比较。仿真结果表明,所提出的NMRC和NZF预编码与传统的MRC和ZF预编码相比,在不增加运算复杂度的前提下可以获得更好的性能,而且新的预编码方案在提高目标用户性能的同时,对其他非目标用户的性能也有一定的改善。     

Performance analysis of ZF receivers with imperfect CSI for uplink massive MIMO systems

We consider the uplink of massive multiple-input multiple-output systems in a multicell environment. Since the base station (BS) estimates the channel state information (CSI) using the pilot signals transmitted from the users, each BS will have imperfect CSI in practice. Assuming zero-forcing method to eliminate the multi-user interference, we derive the exact analytical expressions for the probability density function of the signal-to-interference-plus-noise ratio, the corresponding achievable rate, the outage probability, and the symbol error rate (SER) when the BS has imperfect CSI. An upper bound of the SER is also derived for an arbitrary number of antennas at the BS. Moreover, we derive the upper bound of the achievable rate for the case where the number of antennas at the BS goes to infinity, and the analysis is verified by presenting numerical results.     

Analysis of Tomlinson–Harashima Precoding in Multiuser MIMO Systems With Imperfect Channel State Information

Tomlinson–Harashima precoding (THP) is a successive pre-equalization technique that arranges all the receivers into a group of layers and then suppresses the interlayer interference in sequence at the transmitter. There are two basic THP structures according to the positions of the diagonal weighted filter—decentralized at the receivers or centralized at the transmitter—which are denoted dTHP or cTHP, respectively. In this paper, the investigation of THP in downlink multiuser multiple-input–multiple-output (MIMO) systems with imperfect channel state information (CSI) is performed. The interference power caused by imperfect CSI is first derived, and then, the comparison between the two basic THP structures and the linear zero-forcing (ZF) precoding is carried out in terms of the interference power and the system capacity. Furthermore, the cases with different CSI errors among users are considered, through which the independence of the interference power among layers or subchannels is presented, including the cases with single-antenna and multiantenna receivers. Analytical and simulation results indicate that with imperfect CSI, THP leads to less interference power than linear-ZF, and dTHP leads to larger system capacity than cTHP and linear-ZF.      

基于非理想CSI的下行MU-MIMO鲁棒波束成形

在下行多用户多入多出（MU-MIMO）系统中,基站（BS）所获得的非理想信道状态信息（CSI）会导致频分双工（FDD）系统预编码性能变差.现有的MU-MIMO鲁棒预编码算法虽然可以对抗非理想CSI所导致的系统性能损失,但其只考虑其中一种或两种信道误差的鲁棒性,因此系统性能提升有限.本文通过建立包含信道估计误差、量化误差和延时误差的联合信道误差模型,推导出具有集中式特性的基于最小均方误差（MMSE）的鲁棒波束成形矩阵的闭式解;随后将这种信道条件应用到分布式通信系统,并推导出具有分布式特性的基于信号泄露的MMSE的鲁棒波束成形矩阵的闭式解.数值分析表明,本文所提的集中式和分布式MU-MIMO波束成形算法,与只考虑量化误差的鲁棒MMSE算法相比,具有更优的系统和速率与误码率,补偿了上述三种信道误差所导致的预编码性能损失.     

Adaptive bit allocation scheme in multi-cell cooperative block diagonalization systems with delayed and limited feedback

In multi-cell cooperative multi-input multi-output(MIMO) systems,base station(BS) can exchange and utilize channel state information(CSI) of adjacent cell users to manage co-channel interference.Users quantize the CSIs of desired channel and interference channels using finite-rate feedback links,then BS can generate cooperative block diagonalization(BD) precoding matrices using the obtained quantized CSI at transmitter to supress co-channel interference.In this paper,a novel adaptive bit allocation scheme is proposed to minimize the rate loss due to imperfect CSI.We derive the closed-form expression of rate loss caused by both channel delay and limited feedback.Based on the derived rate loss expression,the proposed scheme can adaptively allocate more bits to quantize the better channels with smaller delays and fewer bits to worse channels with larger delays.Simulation results show that the proposed scheme yields higher performance than other allocation schemes.     

基于大规模天线的多用户 MISO 下行链路频谱效率分析简

在大规模天线时分双工通信系统中,分析了多用户MISO下行链路的频谱效率。假定用户数固定且基站天线数M无限增大,通过理论推导分析发现,当基站发送功率减小到单天线基站的1/M时,随着M的增加系统频谱效率趋于一个恒值;并且简单的预处理即可消除用户间干扰和快衰落的影响。虽然多小区系统会受导频污染的影响,致使小区间干扰不能被完全消除,但是降低后的发送功率可使系统由干扰受限转变为噪声受限。采用迫零单位预编码与采用最大比发送单位预编码的系统频谱效率极限表达式相同。最后给出采用空时分组码的多用户下行链路的频谱效率下限值,并通过仿真分析证实了以上结论的正确性。     

大规模MIMO时分双工系统的鲁棒预编码设计

对于采用大规模MIMO技术的时分双工系统,天线互易误差会破坏上下行信道互易特性,大幅降低预编码算法下行传输性能。由于实际系统难以完全消除天线互易误差,该文以最大化各用户平均信泄噪比为目标,根据天线互易误差的统计特性,设计了对该误差具有鲁棒性的线性预编码算法。同时为了进一步降低用户接收端的等效噪声功率,该文还将该线性鲁棒预编码算法扩展为基于矢量扰动的非线性鲁棒预编码算法,并通过减格辅助技术降低其扰动矢量求解复杂度,使其更适用于大规模MIMO系统应用。计算机仿真结果表明在存在基站天线互易误差条件下,该文所提出的线性与非线性鲁棒性预编码算法的性能均优于传统预编码算法的性能。     

统计和瞬时CSI混合的3D MIMO预编码

本文针对分别具有统计信道状态信息（Channel State Information,CSI）和瞬时CSI的用户,研究了两类用户在三维多输入多输出（Three-Dimension Multiple-Input Multiple-Output,3D MIMO）系统中的下行链路传输模式,解决了混合利用统计和瞬时CSI的下行预编码的设计问题。利用3D MIMO信道的克罗内克积的性质,提出了一种用于最小化基站总发射功率的扩展迫零预编码的方法,并分别求解其最优水平预编码矩阵和最优垂直预编码矩阵。仿真结果表明,该方案具有更低的复杂度和良好的速率。      

大规模MIMO系统的预编码算法综述

相比于传统多输入多输出（MIMO）系统,大规模MIMO的天线数量大幅增加,使得系统的容量提升、误比特率下降,但也造成预编码矩阵维度升高,算法复杂度、系统成本及实现难度增大。将大规模MIMO系统主要采用的预编码技术分为线性和非线性两个部分,对两者进行了归纳和对比,并着重介绍了几种经过简化的线性预编码算法和几种比较典型的非线性预编码算法,指出因为非线性算法的复杂度很高,故未来大规模MIMO系统的预编码应当以线性算法为主。     

Degrees of freedom of MIMO interference channel with distributed space-time interference alignment

In the context of K-user MIMO interference channel (IC),achievable degrees of freedom (DoF) were investigated with distributed space-time interference alignment (DSTIA).By precoding with distributed current and outdated channel state information at the transmitters (CSIT),new tradeoff regions between achievable DoF and CSI feedback delay/frequency were achieved for MISO system.The impact of the number of transmit antennas on achievable DoF in the MISO system was analyzed,revealing that DoF results approach to the outer bound as the number of transmit antennas increases.Further,the impact of the number of receive antennas on achievable DoF was characterized,deriving the range of CSI feedback delay that preserves achievable DoF in the MIMO system.Theoretical and numerical analyses show that,the proposed DSTIA scheme can achieve better sum-DoFs by eliminating inter-user interference perfectly,tighten the gap between achievable DoF and outer bound,as well as improve the achievable rate of the system.     

BER and Capacity Improvement for Multiuser MIMO System by MRT Precoding Technique

In recent years several of endeavors have been conducted in various papers to improve capacity and bit error rate performance of multiuser MIMO system. Every effort tried to increase gain of the individual stream between base station and users’ antennas while reducing the interference among multiple users and interference between streams of single user itself. All these papers adopted precoding techniques for beam forming of individual stream. Various papers have compared precoding techniques such as DPC, BD-ZF, BD-MMSE, BD-QR-MRL, and BD-QR-LRL. Block diagonalization technique is capable to reduce multiuser interference, but not able to combat interference between antennas of single user itself. BD combines with other techniques to overcome this problem. This paper also proposes a precoding technique called modified maximum ratio transmission, which is able to deal with the problem without much complexity. Simulation is done to compare all precoding technique with MRT. The paper tests the precoding techniques on the parameters of BER and capacity, to achieve optimum receiver performance. BER curve and capacity curve are plotted to conclude the paper.

     

Interference alignment scheme for massive MIMO system

The performance of massive multiple-input multiple-output （MIMO） system is limited by pilot contamination. To reduce the pilot contamination, uplink and downlink precoding algorithms are put forward based on interference alignment criterion. In the uplink receiving processing, the target function aligns the pilot contamination and the interference signals to the same null space and acquires the maximal space degree of the desired signals. The uplink receiving precoding matrix is solved on maximal signal to interference plus noise ratio （SINR） criterion considering the impact of the pilot contamination on channel estimations. The uplink receiving precoding matrix is used as the downlink transmitting precoding matrix. Exploiting the channel reciprocity, it is proved that, if the uplink receiving precoding matrix achieves maximal S1NR, the identical precoding matrix can be used in the downlink transmission and acquires maximal signal to leakage plus noise ratio （SLNR）. Simulations show that the spectrum efficiency of the proposed algorithm can reach about 1.5 times higher than that of popular matched filtering （MF） precoding algorithm, and about 1.1 times higher than multi-cell minimum mean square error （MMSE） precoding algorithm. The performance of the proposed algorithm can be improved approximately linearly with the increasing of the number of antennas.     

Efficient multi-user MIMO precoding based on GSVD and vector perturbation

We present an efficient precoding method by using generalized singular value decomposition (GSVD) and vector perturbation in the downlink of multi-user MIMO. Linear precoding/decoding matrices are explicitly computed through the GSVD algorithm, avoiding any matrix inversion or squaring operation. Accordingly, the computational complexity can be drastically reduced compared to the conventional iterative algorithms. In addition, by considering the inter-user interference, we derive a new perturbation vector that minimizes system-wide mean-squared-error of detected symbols. We also show that the proposed precoding scheme is suitable to update the precoding weights efficiently when the channel is newly updated, rather than recomputing the precoding weights for all users from scratch.     

Low Complexity Iterative-QR (IQR) Precoder Design Based on Structure Blocks for Massive MIMO System

Massive multiple-input multiple-output (MIMO) technology is a promising technique having a high spectral efficiency for future wireless systems. Counterintuitively, the practical issues of transmitted signals are being attenuated by fading, propagation limitations, and implement non-linear precoding are solved by enlarging system dimensions. However, the computational complexity of precoding grows with the system dimensions. The existence block diagonalization (BD) precoding, which completely pre-cancels the multiuser interference is very complicated to implement with the use of a large number of BS antennas, since it considers full multiplexing order. Motivated by the high performance of the BD and generalized for the case when the users have multiple antennas, we propose a structure blocks based on iterative QR decompositions (IQRDs) to compute the precoding scheme. The proposed BLIQR-based precoder designed partitioned the channel matrix into capable square-wise blocks matrix and the IQRDs are applied to the blocks channel matrix. The channel matrix is partitioned such that it can fulfill the multiplexing order for the use in Massive MIMO. The computational complexity of the proposed design is effectively reduced and the sum-rate performance is improved, especially in large number of BS antennas. The performance of the proposed scheme achieves a good trade-off between throughput and computational complexity.     

大规模MIMO系统中基于权重二对角迭代的低复杂度预编码

大规模MIMO系统中由于系统下行链路的迫零（zero forcing,ZF）预编码中存在大矩阵求逆运算,随着用户数与天线数的增加,其复杂度随之增加。为了降低复杂度,提出了一种基于雅克比（Jacobi）迭代算法的改进预编码算法,用下二对角矩阵作为迭代矩阵,并且将迭代结果与上一步迭代结果进行权重相加来加速迭代。根据大规模MIMO系统信道矩阵的对角占优特性,将矩阵求逆的诺依曼近似的第一项作为迭代的初始值进一步加速迭代。相比于传统迫零预编码方案,提出的方案可以降低一个量级的算法复杂度,并且保证了预编码方案的性能。     

Transmission Scheme for 2D Antenna Array MIMO Systems with Limited Feedback

Three-dimensional (3D) multiple-input multiple-output (MIMO) systems exploit spatial richness and provide another degree of freedom to transmit signals and eliminate spatial interference. Currently, however, there is no 3D codebook for two-dimensional (2D) antenna array MIMO systems with limited feedback. In this paper, based on the existing 2D codebook, we present a limited feedback and transmission scheme for 2D antenna array MIMO systems. In this scheme, the mobile station (MS) has imperfect channel knowledge, and the base station (BS) only acquires partial information relating the channel instantiation. MS must feed back two channel state information (CSI) instances, i.e., the horizontal and vertical CSIs. After receiving the two CSI instances, the BS interpolates a new vertical precoding vector using the vertical CSI. Then, the BS re-constructs a 3D beamforming vector using horizontal and vertical precoding vectors and compensates the reported horizontal channel quality indicator. System level simulation is employed, and the simulation results show that the proposed method improves the system spectral efficiency and the cell-edge SE significantly.     

基于泊松点过程分布的多蜂窝协作系统中干扰对齐技术研究

在多蜂窝MIMO(Multiple-Input Multiple-Output)协作通信系统中,该文研究了基站站点服从泊松点过程(PPP)分布时,协作基站(BSs)和用户对采用干扰对齐技术的中断概率和网络吞吐量性能,推导了完全信道状态信息(CSI)和部分CSI两种情况下的上述性能的解析表达式,并分析了系统性能与协作参数的单调关系。仿真分析发现:在完全CSI情况下,网络吞吐量随着基站密度、协作基站数、天线数的增加而增加;在部分CSI情况下,由于综合考虑了信道训练和有限反馈带来的资源开销以及量化CSI引入的信道失真,存在一个使网络吞吐量最大的最优协作基站数,当天线数较少或用户移动速度较小时,应有较多的基站参与协作,当天线数较多或用户移动速度较大时,应适当减少协作基站数。