A Unified Framework for Optimizing Linear Nonregenerative Multicarrier MIMO Relay Communication Systems
A Unified Framework for Optimizing Linear Nonregenerative Multicarrier MIMO Relay Communication Systems
复制标题
DOI：
10.1109/tsp.2009.2027779
复制
发表时间：
2009-12
影响因子：
5.4
通讯作者：
Y. Rong;Xiaojun Tang;Y. Hua
中科院分区：
文献类型：
--
作者：
Y. Rong;Xiaojun Tang;Y. Hua
在本文中，我们开发了一个统一的框架，线性非再生多载波多输入多输出（MIMO）中继通信的直接源-目的地链路的情况下。这个统一的框架分类最常用的设计目标，如最小均方误差和最大互信息分为两类：舒尔凹和舒尔凸函数。证明了在Schur-convex目标函数下，最优源预编码矩阵和中继放大矩阵联合对角化源-中继-目的信道矩阵，将多载波MIMO中继信道转化为并行单输入单输出（SISO）中继信道.而对于舒尔凸目标，这种联合对角化发生在源预编码矩阵的特定旋转之后。在确定了源和继电器矩阵的最优结构之后，线性非再生继电器设计问题归结为所得到的SISO继电器通道之间的功率负载问题。我们表明，这种功率加载问题可以有效地解决了交替技术。数值算例验证了该框架的有效性。
In this paper, we develop a unified framework for linear nonregenerative multicarrier multiple-input multiple-output (MIMO) relay communications in the absence of the direct source-destination link. This unified framework classifies most commonly used design objectives such as the minimal mean-square error and the maximal mutual information into two categories: Schur-concave and Schur-convex functions. We prove that for Schur-concave objective functions, the optimal source precoding matrix and relay amplifying matrix jointly diagonalize the source-relay-destination channel matrix and convert the multicarrier MIMO relay channel into parallel single-input single-output (SISO) relay channels. While for Schur-convex objectives, such joint diagonalization occurs after a specific rotation of the source precoding matrix. After the optimal structure of the source and relay matrices is determined, the linear nonregenerative relay design problem boils down to the issue of power loading among the resulting SISO relay channels. We show that this power loading problem can be efficiently solved by an alternating technique. Numerical examples demonstrate the effectiveness of the proposed framework.