Smoothed Wigner-Ville parametric modeling for the analysis of nonstationary signals
Smoothed Wigner-Ville parametric modeling for the analysis of nonstationary signals
复制标题
DOI：
10.1109/iscas.1989.100401
复制
发表时间：
1989-05
期刊：
影响因子：
--
通讯作者：
E. Velez;R. Absher
中科院分区：
文献类型：
--
作者：
E. Velez;R. Absher
Wigner-Ville分布（WVD）具有良好的分辨率，是分析非平稳信号和时变系统的有效工具。然而，当应用于多分量信号时，它会产生称为交叉项的干扰项，这会妨碍信号分量的正确识别。可以通过平滑WVD（SWVD）来显著衰减交叉项。另一个问题是需要大量的存储空间。WVD的参数化建模可用于有效的数据简化和改进的频率分辨率。然而，交叉项在建模后仍然存在，并且通常比信号分量更强。通过估计SWVD的参数模型，可以更清楚地表示信号分量的演化和更低的模型阶数。分析信号的使用允许每个信号分量由单个复极点表示，因此仅需要通常AR模型阶数的一半。进一步的数据压缩是通过对AR参数进行多项式拟合来获得的。
The Wigner-Ville distribution (WVD) is a useful tool in the analysis of nonstationary signals and time-varying systems due to its excellent resolution. When applied to multicomponent signals, however, it produces interference terms known as crossterms, which can hinder the correct identification of signal components. It is possible to attenuate the crossterms significantly by smoothing the WVD in time (SWVD). Another problem is the large amount of storage needed. Parametric modeling of the WVD can be used for efficient data reduction and improved frequency resolution. Crossterms, however, are still present after modeling, and are often stronger than the signal components. By estimating a parametric model of the SWVD, a clearer representation of the signal components' evolution and a lower model order are possible. Use of the analytic signal allows each signal component to be represented by a single complex pole, thus requiring only half the usual AR model order. Further data compression is obtained by polynomial fitting to the AR parameters.>