中文摘要：

LTE系统是当前国内外研究的热点之一，也是当下高铁主要应用的场景，其中，信道估计算法是影响系统性能的主要因素之一。针对高铁LTE下行链路，研究了大区制宏基站3个站点6个RRU基于OFDM的信道估计算法，在高铁快速场景下，在莱斯多径的现实模型中，使用传统的MMSE算法性能较差，对于这种情况，提出了基于延时功率谱的LMMSE信道估计算法，利用时频域滤波和多次频域测量做频偏校准。针对这些改进算法，分别在理想情况和实际情况下进行了仿真实验。结果表明，提出的信道估计算法不仅有较低的复杂度，而且比传统的MMSE算法性能要高，是一种很有效的适合当前高铁场景的信道估计算法。

英文摘要：

LTE system is a hot research topic now, and is mainly applied in high-speed rail. The channel estimation algorithm is one of the main factors that affect the throughput. For LTE downlink, this paper studies the channel estimation algorithm based on the district system macro base station which includes three sites and six radio remote units. In the high speed scene, the performance of the traditional MMSE algorithm is bad in the Race channel model. In the result, we propose an LMMSE channel estimation algorithm based on delay power spectrum. In addition, we use frequency domain filtering and multiple fre- quency measurement to calibrate the frequency offset. The simulation experiments are carried out under ideal conditions and ac- tual conditions respectively. The results show that the proposed channel estimation algorithm not only has low complexity, but has about 11dB gain than the traditional algorithm in performance. It is a very effective channel estimation algorithm.