MIMO双向中继GSVD线性预编码的功率优化算法  

Power allocation under GSVD linear precoding in the MIMO two-way relay network

作　　者：刘圣恩 肖霖[1] 杨鼎成[1] 谭安平[1] 

机构地区：[1]南昌大学信息工程学院,江西南昌330031

出　　处：《西安电子科技大学学报》2015年第5期161-167,共7页Journal of Xidian University

基　　金：国家自然科学基金资助项目(61340025)

摘　　要：为优化多输入多输出双向中继系统的功率分配,应用广义奇异值分解方法对角化一对用户到中继的信道,并划分成两种传输模式:单向和双向中继传输模式.针对划分后的空间子信道采取矩阵置换的方法对齐中继节点处的链路信号.此外,相比传统的每个节点总功率约束,提出约束单链路下3节点总功率的方法来最大化系统平均总速率,获得功率优化的封闭解并降低计算的复杂度.仿真结果显示,所提功率算法具有和全局优化算法非常相近的性能,且在30dB处,天线规格从3-5-3到4-6-4再到5-7-5,能够带来近2bit/(s·Hz-1)的递增增益.In order to optimize the power allocation of the MIMO two-way relay system,we diagonalize the channels between users and relay using the generalized singular value decomposition（GSVD）method,and divide them into two transmission models：one-way relay model and two-way relay model.For the subchannels,a scheme of permutation is proposed to align the data streams at the relay node.Moreover,we propose the method for the power constraint of three nodes under each data stream to maximize the average sum-rate of the system instead of the constraint of each node＇s total power and get the close-form optimal solution which can reduce the complexity of the algorithm.Finally,simulation results demonstrate that the performance of the scheme we use is very close to that by the global optimization algorithm,and that with the antennas specification from 3-5-3to 4-6-4and to 5-7-5,the scheme can bring about approximate2bit/（s·Hz-1）increased gain when the SNR is 30 dB.

关 键 词：多输入多输出 双向中继 对角化 广义奇异值分解 功率优化 

分 类 号：TN925[电子电信—通信与信息系统]