自适应OFDM系统中基于LS估计的信道预测算法分析

摘要

信道预测技术可以为自适应系统提供准确及时的信道状态信息.针对现有的信道预测算法都是假设已知当前及以前的信道信息来预测未来的信道状态信息而忽略信道估计的影响,提出了一种基于LS信道估计的时变多径信道的时域预测方法.该方法采用自适应OFDM系统模型,以当前接收的导频符号序列为观测值进行LS信道估计得到信道的频域值,然后通过IDFT变换到时域,结合AR模型来预测未来信道衰落系数.仿真结果表明,与传统的频域预测算法相比,该算法有效地降低了最小均方误差,能很好地满足自适应OFDM系统中时变多径信道的要求.%Channel prediction technology can provide adaptive system for accurate and timely channel state informa-tion(CSI). According to the existing channel prediction algorithm assuming that all known current and former channel information to predict the future channel state information and ignoring the influence of channel estimation, a time-varying multipath channel time-domain prediction algorithm based on LS estimate was introduced. This method adopt the adaptive OFDM system model, with the current receiving pilot symbol sequence using LS channel estimation for channel frequency domain value, and then transformed to time domain through IDFT, combined with AR model to predict future channel fading coefficient. The simulation results show that compared with the traditional frequency domain prediction algorithm, the algorithm has effectively reduced the minimum mean square error and can be satisfied with the adaptive OFDM system time-varying multipath channel requirements.