基于瞬时参数估计的信号分量提取

具有非线性时频关系的多分量信号频谱占有较宽的频带且相互交叠,针对此类信号分量提取难的问题,提出了一种基于瞬时参数估计的信号分量提取算法。首先,采用FMmlet变换将被分析信号在时频空间内展开;然后,根据时频面内能量峰脊与瞬时频率的对应关系,应用谱峰检测法估计出所需提取分量的瞬时频率;按照最小二乘法滤波原理估计该分量的瞬时幅值,继而按照信号瞬时模型恢复重构出对应的信号分量。理论分析和仿真试验表明,该方法能从时频相交或不相交的多分量复杂时变信号中提取出所需信号分量,且失真度小。

时频面上基于瞬频估计的信号提取方法

针对多分量非平稳信号的噪声抑制和信号分离问题,提出了一种新的时频滤波法。采用FMmlet自适应分解将被分析信号在时频空间内展开,得到无交叉项干扰、时频聚集性很强,且能反映信号的线性或非线性结构的时频特性的时频分布。根据时频面内能量脊与瞬时频率的对应关系,采用谱峰检测和时频加窗轮流迭代的方法,估计出信号分量的瞬时频率。在瞬时频率精确估计的基础上,设计适当的时频滤波函数,对被分析信号的Wigner-Ville分布(WVD)进行时频加窗处理,得到单分量信号的修正WVD,然后采用WVD反变换进行信号分量的时域重构,达到从非平稳信号中分离有用分量的目的。理论分析和仿真实验表明,利用该方法从复杂非线性时变信号中提取出的信号分量失真度小。该方法在非平稳信号的深层特征提取中具有良好的应用前景。

FM^mlet变换在信号分离中的应用

具有非线性时频关系的多分量信号的频谱占有较宽的频带、并且各信号分量的频谱常常互相交叠,因之难以用常规的滤波方法将各信号分量分离,将信号与背景噪声分离亦非易事。由于FM^m let变换的原子能匹配信号的线性和非线性时变成分,因此本文提出采用FM^m let变换进行信号分离。实际算例结果印证了本文方法的有效性。

APPLICATION OF FM^mlet TRANSFORM TO SIGNAL SEPARATION

Since the multicomponent signal with nonlinear time-frequency structures occupies a wide frequency band, and the spectral contents may alias, it is therefore difficult to separate the signal components and to separate the signal from Ijackground noise. In this paper, a new signal separation method using FMmlet transform is proposed by taking the advantage that the atoms of FMm let transform can match both the linear and nonlinear time-varying structures. Theoretical predictions and numerical experiments show the feasibility of the methodology advocated.

Subspaces of FM^mlet transform

The subspaces of FMmlet transform are investigated. It is shown that some of the existing transforms like the Fourier transform, short-time Fourier transform, Gabor transform, wavelet transform, chirplet transform, the mean of signal, and the FM-1let transform, and the butterfly subspace are all special cases of FMmlet transform. Therefore the FMmlet transform is more flexible for delineating both the linear and nonlinear time-varying structures of a signal.

Properties and convergence analysis of FM^mlet transform

The linearity,time shifting,time scaling,and time inversion properties of FMmlet transform are proved,and the frequency shifting property of one of the subspaces of FMmlet transform,namely the chirplet transform is presented.Moreover,it is proved that in the process of FMm let based atomic signal decomposition,the residual signals decay exponentially.