文档介绍：摘要随着移动通信技术的发展,人们对高速移动下通信业务的需求迅速增加,为了保证高速数据传输的可靠性,需要对时变信道下的OFDM系统信道估计技术进行研究。与时不变信道不同的是,在时变信道下进行信道估计时,每条径的时域冲激响应不再是常数,这使得在单个OFDM符号里需要估计的未知量个数大于所观察数据的个数。此外,接收的导频数据还会受到子载波间干扰的污染,从而导致时变信道估计的难度增大。,现有的插值方法并没有很好地消除ICI及噪声的影响。为此,本文提出了一种基于时域插值的改进算法,通过在信道多径时延扩展是采样时间整数倍的慢时变信道下设置一个合适的能量门限,消除由于ICI干扰及噪声产生的假径,增加了插值的精度,从而提高了信道估计的性能。,现有的估计方法通常只是利用了导频的信息,而忽略了发送的数据信息。为此,本文提出了一种LS估计与SAGE信号均衡相结合的迭代估计算法。首先通过LS估计获得初步信道信息,再经过SAGE信号均衡获得发送的频域信息,并将其反馈应用于LS估计中。仿真结果验证了算法的有效性。尽管对时变信道下的信道估计进行了一定的研究,但诸如线性模型和基扩展模型(BEM)的研究都还只停留在初步阶段,如何进一步提高时变信道估计的精度以及降低其复杂度需要继续研究。关键词:时变信道信道估计ICI时域插值SAGEAbstractWith the development of mobile communication technology, the demand munication services under the high-speed mobile environment increases rapidly,toguarantee the reliability of high-speed data transmission, channel estimation must bestudied in time-variant channel for OFDM systems. On contrary tothat in time-invariant channel, the implus response of each path in time domainis not constant in time-variant channel, which makes the number of parameters to be estimated are larger than that of number of the observation date in one OFDM symbol. In addition, the pilots received are smeared byinter-carrier interference (ICI), increasing the difficulty of channel estimation. So the research on channel estimation method is particularly existing interpolation methods could not thoroughly eliminate ICI and noise at thenormalized Doppler FrequencyShiftis less than slowly time-variant channel. To solve this problem, an improved time-domain interpolation algorithm is is performed by setting an appropriate energy threshold to eliminate the fake path generated by ICI interference and noise in the casethe multipath delay spreadis an integer multiple of the sampling duration. As a result, it improves the performanceofchannel estimation, for it increases the accuracy