基于线性正则变换的时频信号分离方法

Novel Method for Time-Frequency Signal Separation Based on the Linear Canonical Transform

时频信号分析是当今信号处理领域研究的一个热点问题,各种时频分布函数得到了广泛的研究和应用。线性正则变换是一种重要的时频分析工具。文中研究了线性正则变换与传统时频分布函数的关系;并基于这些关系,提出了一种新的时频信号分离方法,能够把在时频面上互不重叠、但在时域和频域均存在较强耦合的多分量合成信号有效地分离。仿真实例表明了该方法的正确性和实用性。

Time-frequency signal analysis is a hot research topic in signal processing domain at present. A number of time-frequency distributions have been developed and used to analyze time-frequency signal. The linear canonical transform with multi-parameter is a powerful time-frequency analysis tool. It is the generalization of the Fourier transform and the fractional Fourier transform. In this paper, the relations between the linear canonical transform and time-frequency distributions are obtained. Based on these relations, a novel method for time-frequency signal separation is proposed. This method is very effective to separate components from a time-frequency signal, whose components interact in the time and the frequency domains and does not overlapped with each other in the time-frequency plane.The simulation results illustrate the validity and practicability of the proposed method.