存在噪声和未建模动态系统传递函数的渐近无偏估计

存在未建模动态时，如何去除系统传递函数估计中噪声引起的偏差是系统辨识关心的重要问题。本文将我们已提出的ＢＥＬＳ方法进行推广得出：当系统未建模动态可用－ＦＩＲ模型描述时，应用本文的方法可得到系统传递函数的无偏辨识，而不需知道有关噪声和未建模动态概率分布函数的先验知识。

Asymptotically Unbiased Estimation of Transfer Functions with Colored Noises and Unmodeled Dynamics

In the presence of the unmodeled dynamics,one and important concern is how to remove the noise-induced bias in the estimate of the transfer function An extension to the newly es tablished BELS method ismade in this paper When the unmodeled dynamics is decribed by a finite im pulse response (FIR)model,It can achieve a unbiased estimate of the transfer function without the priori knowledge about the probability density functions(PDFs)of the noise and the dynamics.