Blind channel identification using lower-order statistics
暂无分享,去创建一个
Non-Gaussian statistical signal processing is important when the signals deviate from the ideal Gaussian model. One of the most important non-Gaussian distributions is the family of the alpha-stable (0 less than (alpha) less than 2) distributions. They are proven to be effective in modeling the impulsive signal environments both in theory and in practice. This paper presents a brief introduction to the alpha-stable distributions, their applications to signal processing using fractional lower-order statistics, and the alpha-spectrum, a new spectral analysis tool for blind channel identification in impulsive environments.
[1] S. Cambanis,et al. Linear Problems in Linear Problems in pth Order and Stable Processes , 1981 .
[2] C. L. Nikias,et al. Signal processing with fractional lower order moments: stable processes and their applications , 1993, Proc. IEEE.
[3] Chrysostomos L. Nikias,et al. The robust covariation-based MUSIC (ROC-MUSIC) algorithm for bearing estimation in impulsive noise environments , 1996, IEEE Trans. Signal Process..
[4] Christian Houdré,et al. Disjointness results for some classes of stable processes , 1993 .