基于LPFT时频滤波器的WVD交叉项抑制方法

WVD cross term suppression using the LPFT-based time-frequency filtering

本文提出一种抑制Wigner-Ville分布(WVD)交叉项的新方法。利用局部多项式傅里叶变换(LPFT)构建时频滤波器,确定自项支撑区域,再利用此滤波器对WVD进行处理,从而达到抑制交叉项的目的。通过3dB信噪比的分析可以看出,LPFT具有比短时傅里叶变换(STFT)更好的时频聚集性,因此基于LPFT的时频滤波器能更有效抑制WVD中的交叉项干扰,同时又能保留WVD的高时频聚集性。通过与Chio-Williams分布、径向高斯核函数时频分布、基于STFT时频滤波器的交叉项抑制方法的比较,验证了该方法对抑制多分量信号及非线性调频信号的交叉项以及噪声干扰的有效性,显示了该方法在保持高时频聚集性,抑制交叉项干扰,以及抑制噪声干扰方面的优势。

In this paper,a new method is proposed to suppress the cross terms in the WVD.First,the time-frequency filtering based on the local polynomial Fourier transform（LPFT） is constructed,then the filtering is used to determine the auto term region and suppress cross terms in the WVD.As shown in the 3dB SNR analysis,the LPFT can provide better resolution than the STFT,therefore,the LPFT-based time-frequency filtering can effectively suppress cross terms in the WVD,as well as maintain the high resolution characteristic of the WVD.Compared with the Chio-Williams distribution,the radially Gaussian kernel time-frequency representation,and the WVD cross term suppression method using the STFT-based filtering,the proposed method proves its advantages in cross term suppression,as well as in noise suppression,for multicomponent or nonlinear FM signals.