论文部分内容阅读
在移动通信中,信道增益噪声方差和Doppler频移等参数的估计具有重要意义.然而移动台的相对运动使得传输信道变成时变信道,这降低了参数估计的准确性.因此本文研究一般时间选择性信道的参数估计方差界推导,即有限样本Cramér-Rao下界(CRLB),推导时考虑Doppler扩展、方向性散射以及加性噪声的联合影响.这些界对于评估相应信道参数估计器的性能极为有益.为了处理信道功率谱密度(power spectrum density,PSD)边界的奇异性,本文提出了一种近似谱密度方案,解决了非近似方案无法适应PSD边界不可微的缺陷.数值计算结果表明,只要样本数满足一定的条件,近似方案可以获得与非近似方案逼近的CRLB计算结果,但是前者扩展了计算方法对于Doppler频移范围的适应性.
In mobile communication, the estimation of channel gain noise variance and Doppler frequency shift is of great significance.However, the relative motion of mobile station makes the transmission channel become time-varying channel, which reduces the accuracy of parameter estimation.Therefore, The Cramér-Rao lower bound (CRLB), which is derived from the variance estimation of the selective channel parameter estimation, takes into account the combined effects of Doppler spreading, directional scatter and additive noise, which are extremely useful for evaluating the corresponding channel parameter estimators In order to deal with the singularity of the power spectral density (PSD) boundary, an approximate spectral density scheme is proposed to solve the problem that the non-approximate scheme can not adapt to the non-differentiable PSD boundary. The numerical results show that, The number of samples satisfies certain conditions. The approximation scheme can obtain the CRLB calculation results approaching the non-approximation scheme, but the former expands the adaptability of the calculation method to the Doppler frequency shift range.