一种新的半盲联合信道数据估计算法  

A novel semi-blind joint channel parameters and data estimator

作　　者：白栋[1] 刘红[2] 梁庆林[1] 

机构地区：[1]北京大学电子学系卫星与移动通信实验室,北京100871 [2]北京邮电大学宽带通信网络实验室,北京100876

出　　处：《通信学报》2004年第11期43-48,共6页Journal on Communications

基　　金：国家自然科学基金资助项目(90204001)

摘　　要：提出一种新的半盲联合信道估计与数据检测器,通过将递归-EM信道估计算法嵌入Turbo均衡器中,根据接收信号进行逐符号信道估计,不但能够适应时变信道,而且有效提高了联合信道数据估计收敛速度。仿真表明本算法对初始信道响应估计精度要求很低,从而大大降低了对训练序列长度的要求,使用很短的训练序列就能够达到较好性能,有效提高了频谱利用率。A novel semi-blind joint channel parameters and data estimator was presented in this article. Recursive expectation-maximization (EM) algorithm and Turbo equalization were combined to estimate unknown time-invariant/variant channel parameters symbol by symbol. The recursive times needed for converging are much decreased. Simulation results show that the proposed method works well with rough initialization of channel parameters obtained from very short train sequence, which lead to high spectrum efficiency.

关 键 词：信号检测 递归-EM算法 TURBO均衡 半盲联合信道数据估计 渐近ML估计 

分 类 号：TN911.5[电子电信—通信与信息系统]