大规模MIMO系统中基于Lanczos方法的低复杂度预编码  

Low Complexity Precoding Based on Lanczos Method in Massive MIMO System

作　　者：曹海燕[1] 周冬 方昕[1] 王秀敏 CAO Haiyan;ZHOU Dong;FANG Xin;WANG Xiumin(Institute of Communications Engineering,Hangzhou Dianzi University,Hangzhou 310018,China;Institute of Information Engineering,China Jiliang University,Hangzhou 310018,China)

机构地区：[1]杭州电子科技大学通信工程学院,杭州310018 [2]中国计量大学信息工程学院,杭州310018

出　　处：《计算机工程》2019年第2期87-91,共5页Computer Engineering

基　　金：国家自然科学基金(61501158;61379027);浙江省自然科学基金(LY14F010019;LQ15F01004)

摘　　要：针对大规模MIMO系统中因基站天线数与用户数过大导致迫零(ZF)预编码矩阵求逆复杂度较高的问题,提出一种基于迭代子空间投影算法的Lanczos方法低复杂度预编码方案。根据大规模MIMO系统信道矩阵具有对角占优特性,将信道大矩阵求逆诺依曼级数的第1项作为迭代的初始值,从而加快算法的收敛速度,使得ZF预编码的复杂度从O(K^3)降低到O(K^2)。仿真结果表明,该算法以较快的收敛速度逼近传统ZF预编码方案的信道容量与误码率性能。Aiming at the problem that the computational complexity of the required matrix inversion of Zero Forcing(ZF) precoding matrix is very high due to the large number of antennas and users in large-scale MIMO systems,a low complexity ZF precoding scheme based on Lanczos Method(LM) is proposed.In the proposed iterative algorithm,according to the channel matrix of massive MIMO systems which has diagonally dominant characteristics,the first term of the Neumann series of the inverse of the channel large matrix is taken as the initial value of the iteration,that speeds up the convergence rate of the algorithm,and the complexity of ZF precoding can be reduced from O(K 3) to O(K 2).Simulation results show that the algorithm approaches the channel capacity and Bit Error Rate (BER) of traditional ZF precoding scheme with fast convergence speed.

关 键 词：MIMO系统 ZF预编码 LANCZOS方法 低复杂度 诺依曼级数 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]