中文摘要：

TDLTE-A信道估计的经典算法有最小平方（LS）算法,线性最小均方误差（LMMSE）算法以及基于变换域的离散傅里叶变换（DFT）算法.LS算法不需要知道信道的先验信息,但是对噪声没有做任何处理.LMMSE算法利用了信道的自相关矩阵,估计性能比LS要优良,但是计算较为复杂,不利于在实际系统中实现.DFT算法则是在LS算法估计结果的基础上,进行变换域处理.但是DFT算法存在较为明显的“地板效应”.然而,提出通过改进TD-LTE-A系统的参数配置,可以有效地改善DFT算法的估计性能,使其性能接近于理想信道估计,并且通过仿真比较了DFT估计与其他信道估计算法在信道响应值的均方误差和信道冲击响应值的差异.

英文摘要：

LS algorithm,I.MMSE algorithm and transform-based DFT algorithm are classical algorithm in TD-LTE-A downlink channel estimation. LS algorithm do not require prior information of channel, while it does not process the noise that mixed in channel. LMMSE algorithm utilize the auto correlation matrix of the channel, which makes its property better than LS algorithm, but it＇s not so easy to be implemented in real world due to its complexity. Meanwhile,the DFT algorithm needs to be processed in frequency domain based on the result of LS estimation. Under this circumstance,there is so called ＂floor effect＂ exist in IDTF algorithm. In this paper, we present a new method , namely to modify the parameter of TD-LTE-A system, to improve the estimation property of DFT algorithm. In addition,we compare the mean square error and the value of channel impulse response of DFT algorithm with other channel estimation algorithm through simulation.