循环谱密度估计的最大似然滤波器法

Estimation of cyclic spectra using maximum likelihood filters

针对周期平稳随机过程的循环谱密度的估计过去主要采用时域平均周期图法和频域平滑周期图法，特别当短数据时，上述方法的分辨率很低，方差极大，提出了循环谱密度估计的最大似然滤波器法，并作了改进．该方法不仅具有很高的分辨率，而且估计的方差也很小，具有较好的估计性能．

The conventional estimation methods of cyclic spectra for cyclostationary processes are temporally smoothed cyclic periodogram and spectrally smoothed cyclic periodogram. In the case of short record data, both methods have low resolution and bad reliability. This paper uses maximum likelihood filters with the modified analysis effective bandwidth to estimate cyclic spectra. Fine performances such as resolution and reliability can be obtained with this method.