基于数据依赖的叠加训练序列信道估计

通过采用依赖于信息数据的叠加式周期训练序列发送机制,考虑到信道存在未知直流偏移的情况,提出了一种估计频率选择性信道的新算法。推导出信道估计的均方误差公式,获得了最优的训练序列所应满足的条件。理论分析和仿真结果表明,该算法大大提高了信道估计和信号检测的性能,从而验证了该算法的有效性和可靠性。

Channel Estimation Using Data-dependent Superimposed Training

In this paper, in consideration of the possibility of having an unknown DC offset at the receiver in the frequency-selective channel,a new channel estimation algorithm based on data-dependent superimposed training is presented, and a closed-form solution for the channel estimation variance is derived. The characteristics that a training sequence should have in order to obtain an optimal channel estimation performance are also derived here. Theoretic analysis and simulation results show that the proposed algorithm offers quite good estimation performance and BER characteristics, so the method is effective and reliable.