IQ不平衡OFDM系统两种逼近最大似然性能的亚最大似然补偿算法  

Two quasi-ML compensation methods for OFDM systems with IQ imbalances

作　　者：童娟娟 束锋[1,2,3] 李隽[1] 余诗尧 赵军辉[4] 

机构地区：[1]南京理工大学电子工程与光电技术学院,南京210094 [2]东南大学移动通信国家重点实验室,南京210096 [3]南京理工大学近程高速目标探测技术国防重点学科实验室,南京210094 [4]北京交通大学电子与信息学院,北京100044

出　　处：《高技术通讯》2014年第9期910-914,共5页Chinese High Technology Letters

基　　金：国家自然科学基金(61271230;61472190);中央高校基本科研业务费专项资金(30920130122004);东南大学移动通信国家重点实验室(2013D02)资助项目

摘　　要：针对在同相正交IQ不平衡的正交频分复用(OFDM)系统中用传统的最大似然(ML)补偿算法需要二维复向量的搜索,从而导致计算复杂度较高的问题,提出了两种逼近ML算法性能的补偿算法,即一维实向量迭代搜索算法和二维实向量迭代搜索算法,它们通过在空间中迭代搜索一维或二维实向量来降低计算复杂度。仿真结果表明,上述改进的亚ML补偿算法的性能优于现有的频域和时域最小二乘补偿方法,逼近ML算法,其计算复杂度明显低于传统的ML算法,误比特性能与ML算法相当。Based on the study of the problem that for orthogonal frequency division multiplexing （OFDM） systems with inphase and quadrature （IQ） imbalances, the conventional maximum likelihood （ML） compensation method has the higher computational complexity since it adopts exhaustive searching in space of two-dimensional complex vectors, two quasi-ML compensation methods, the 1D vector searching and the 2D vector searching, were proposed based on the conventional ML compensation method, which reduce the complexity by iterative searching in space of one or two-dimensional real vectors. The simulation results show that the performance of the proposed quasi-ML compensation methods outperform the existing compensation methods such as the time-domain least square （LS） and the frequency-domain LS, and are close to the conventional ML but the computational complexity is obviously lower than it.

关 键 词：正交频分复用(OFDM) IQ不平衡 最大似然(ML)算法 亚最大似然算法 补偿 

分 类 号：TN929.53[电子电信—通信与信息系统]