典型非线性固定核函数时频分布的性能分析

Performance Analysis on Typical Nonlinear Fixed Kernel Function Time Frequency Distribution

传统的傅立叶变换无法满足对非线性、非平稳信号的分析,时频分析旨在构造一种时间和频率的密度函数,能够反映非平稳信号的时变特征,是非平稳信号分析的有力工具。文章讨论了魏格纳-威尔分布、伪平滑魏格纳-威尔分布及乔伊-威廉斯分布三种固定核函数时频分析方法的性能特征,进行计算机仿真,最后总结这三种分布在时频分辨率、抗干扰能力和信号适用性方面的优缺点。

Traditional Fourier transform can＇t analyze non-linear non-stationary signals. In contrast, time-frequency analysis intends to constructs two-dimensional space of time and frequency so as to represent time-varying characteristics of non-stationary signals. The paper launches an investigation on the performance of three typical nonlinear time frequency analysis methods： Wigner-Ville distribution, Smoothed pseudo Wigner-Ville distribution and Choi-Williams distribution. The simulation is conducted on the computer. The advantages and disadvantages of the three methods are summarized in terms of time-frequency resolution, anti interference ability, and signal application.