MIMO中继系统中基于不完全信道状态信息的预编码算法研究
发布时间：2019-01-04 16:58
【摘要】：多天线中继系统将多输入多输出(Multiple-Input Multiple-Output,MIMO)技术与中继技术的优势结合,可明显提升系统覆盖范围与频谱利用率。预编码技术作为能进一步提升MIMO中继系统性能的关键技术,一直被学术界广泛关注。预编码设计时需要获取信道状态信息(Channel State Information,CSI),但由于信道估计方法的局限性以及反馈链路的时延性,理想的CSI很难得到。本文基于不完全CSI,以改善系统误码率为优化目标,分别对MIMO系统、单用户MIMO中继系统、多用户MIMO中继系统、MIMO双向中继系统的预编码算法进行了研究。主要研究工作如下:(1)针对MIMO系统,考虑信道存在估计误差及反馈延迟的不完全CSI情况,研究了基于最小均方误差(Minimum Mean Squared Error,MMSE)准则的预编码算法。介绍了求解约束优化问题时采用的拉格朗日乘子法和无约束优化问题时采用的求导法,以及通过交替迭代实现联合优化的联合迭代法,为MIMO中继系统的预编码算法研究打下基础。(2)针对单用户MIMO中继系统,考虑信道存在估计误差及反馈延迟的不完全CSI情况,提出了基于MMSE准则的预编码算法。考虑发射端和中继节点存在功率约束的情况,通过修正因子的引入,采用拉格朗日乘子法和KKT准则将求解三节点矩阵的优化问题转化为存在不等式约束的多元矩阵方程组问题,最后采用联合迭代法交替更新得到各节点矩阵的最优解。仿真结果显示,与未综合考虑各个节点的预编码算法相比,所提算法能明显改善单用户MIMO中继系统的误码率性能。(3)针对多用户MIMO中继系统,在信道存在估计误差及反馈延迟的不完全CSI条件下,提出了基于MMSE准则的预编码算法。不同于单用户MIMO中继系统的联合求解方案,多用户MIMO中继系统预编码算法将约束优化问题拆分为三个包含矩阵变量的子凸优化问题并分别求解。因为发送端的用户配有单天线且服从独立分布可先确定发送端预编码矩阵;中继预编码矩阵的求解问题可转化为SDP半正定规划问题,利用CVX工具箱求解;接收端处理矩阵利用线性搜索法求解。最后联合迭代得到最优解。仿真结果表明,与未考虑中继节点预编码的算法相比,所提算法能使多用户MIMO中继系统获得更优的误码率性能。(4)针对MIMO双向中继系统,在信道存在估计误差及反馈延迟的不完全CSI条件下,以MMSE为设计准则,提出了基于矩阵分解法和联合迭代法的两种预编码算法。矩阵分解法通过矩阵分解和不等式理论,将矩阵优化问题转化为矩阵内的标量元素优化问题,采用二分法求解。联合迭代法交替更新直至算法收敛,能得到中继预编码矩阵、接收端处理矩阵的最优解。与已有算法的系统性能仿真对比表明,所提两种算法能有效提升MIMO双向中继系统的误码率性能。
[Abstract]:The multi-antenna relay system combines the advantages of multi-input and multi-output (Multiple-Input Multiple-Output,MIMO) technology with relay technology, which can obviously improve the coverage and spectral efficiency of the system. As a key technology to further improve the performance of MIMO relay systems, precoding technology has been widely concerned by academia. It is necessary to obtain channel state information (Channel State Information,CSI) in precoding design. However, due to the limitation of channel estimation method and the delay of feedback link, the ideal CSI is difficult to obtain. Based on incomplete CSI, the precoding algorithms of MIMO system, single user MIMO relay system, multiuser MIMO relay system and MIMO bidirectional relay system are studied. The main research works are as follows: (1) for MIMO systems, a precoding algorithm based on the minimum mean square error (Minimum Mean Squared Error,MMSE) criterion is studied for the incomplete CSI with channel estimation error and feedback delay. The Lagrange multiplier method and the derivation method used in solving constrained optimization problems are introduced, and the joint iteration method for joint optimization by alternating iteration is introduced. It lays a foundation for the research of precoding algorithm of MIMO relay system. (2) for single-user MIMO relay system, considering the incomplete CSI with channel estimation error and feedback delay, a pre-coding algorithm based on MMSE criterion is proposed. Considering the existence of power constraints at the transmitter and relay nodes, the Lagrange multiplier method and KKT criterion are used to transform the optimization problem of the three-node matrix into the problem of multivariate matrix equations with inequality constraints. Finally, the joint iteration method is used to alternately update the optimal solution of each node matrix. Simulation results show that the proposed algorithm can significantly improve the BER performance of single-user MIMO relay system compared with the pre-coding algorithm without comprehensive consideration of each node. (3) for multi-user MIMO relay system, A precoding algorithm based on MMSE criterion is proposed under the condition of incomplete CSI with channel estimation error and feedback delay. Different from the joint solution scheme of single-user MIMO relay system, the multi-user MIMO relay system precoding algorithm divides the constrained optimization problem into three subconvex optimization problems containing matrix variables and solves them separately. Because the user of the transmitter has a single antenna and the independent distribution of the transmitter can determine the precoding matrix of the transmitter, the solution of the relay precoding matrix can be transformed into a SDP semi-definite programming problem, and the solution can be solved by using the CVX toolbox. The receiver processing matrix is solved by linear search method. Finally, the optimal solution is obtained by joint iteration. Simulation results show that the proposed algorithm can make the multi-user MIMO relay system obtain better BER performance than the algorithm without considering the relay node precoding. (4) for the MIMO bidirectional relay system, Under the condition of incomplete CSI with channel estimation error and feedback delay, two precoding algorithms based on matrix decomposition method and joint iteration method are proposed using MMSE as the design criterion. By matrix decomposition and inequality theory, the matrix optimization problem is transformed into the scalar element optimization problem in the matrix, and the bisection method is used to solve the problem. The joint iterative method alternately updates until the algorithm converges, and the optimal solution of the relay precoding matrix and the receiver processing matrix can be obtained. Compared with the existing algorithms, the simulation results show that the proposed two algorithms can effectively improve the BER performance of MIMO bidirectional relay systems.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN919.3
本文编号：2400550
[Abstract]:The multi-antenna relay system combines the advantages of multi-input and multi-output (Multiple-Input Multiple-Output,MIMO) technology with relay technology, which can obviously improve the coverage and spectral efficiency of the system. As a key technology to further improve the performance of MIMO relay systems, precoding technology has been widely concerned by academia. It is necessary to obtain channel state information (Channel State Information,CSI) in precoding design. However, due to the limitation of channel estimation method and the delay of feedback link, the ideal CSI is difficult to obtain. Based on incomplete CSI, the precoding algorithms of MIMO system, single user MIMO relay system, multiuser MIMO relay system and MIMO bidirectional relay system are studied. The main research works are as follows: (1) for MIMO systems, a precoding algorithm based on the minimum mean square error (Minimum Mean Squared Error,MMSE) criterion is studied for the incomplete CSI with channel estimation error and feedback delay. The Lagrange multiplier method and the derivation method used in solving constrained optimization problems are introduced, and the joint iteration method for joint optimization by alternating iteration is introduced. It lays a foundation for the research of precoding algorithm of MIMO relay system. (2) for single-user MIMO relay system, considering the incomplete CSI with channel estimation error and feedback delay, a pre-coding algorithm based on MMSE criterion is proposed. Considering the existence of power constraints at the transmitter and relay nodes, the Lagrange multiplier method and KKT criterion are used to transform the optimization problem of the three-node matrix into the problem of multivariate matrix equations with inequality constraints. Finally, the joint iteration method is used to alternately update the optimal solution of each node matrix. Simulation results show that the proposed algorithm can significantly improve the BER performance of single-user MIMO relay system compared with the pre-coding algorithm without comprehensive consideration of each node. (3) for multi-user MIMO relay system, A precoding algorithm based on MMSE criterion is proposed under the condition of incomplete CSI with channel estimation error and feedback delay. Different from the joint solution scheme of single-user MIMO relay system, the multi-user MIMO relay system precoding algorithm divides the constrained optimization problem into three subconvex optimization problems containing matrix variables and solves them separately. Because the user of the transmitter has a single antenna and the independent distribution of the transmitter can determine the precoding matrix of the transmitter, the solution of the relay precoding matrix can be transformed into a SDP semi-definite programming problem, and the solution can be solved by using the CVX toolbox. The receiver processing matrix is solved by linear search method. Finally, the optimal solution is obtained by joint iteration. Simulation results show that the proposed algorithm can make the multi-user MIMO relay system obtain better BER performance than the algorithm without considering the relay node precoding. (4) for the MIMO bidirectional relay system, Under the condition of incomplete CSI with channel estimation error and feedback delay, two precoding algorithms based on matrix decomposition method and joint iteration method are proposed using MMSE as the design criterion. By matrix decomposition and inequality theory, the matrix optimization problem is transformed into the scalar element optimization problem in the matrix, and the bisection method is used to solve the problem. The joint iterative method alternately updates until the algorithm converges, and the optimal solution of the relay precoding matrix and the receiver processing matrix can be obtained. Compared with the existing algorithms, the simulation results show that the proposed two algorithms can effectively improve the BER performance of MIMO bidirectional relay systems.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN919.3
【参考文献】
相关期刊论文 前2条
1 尤肖虎;潘志文;高西奇;曹淑敏;邬贺铨;;5G移动通信发展趋势与若干关键技术[J];中国科学:信息科学;2014年05期
2 张彬;贺志强;牛凯;;MIMO中继系统中的线性鲁棒收发机联合设计[J];北京邮电大学学报;2010年02期
，本文编号：2400550
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/2400550.html
