应用希尔伯特黄变换的单矢量传感器多目标分辨研究

将希尔伯特黄变换与矢量信号处理相结合应用到水声领域,与传统方法相比有许多优势。在希尔伯特黄变换中本征模态函数是基于序列数据的局部时间尺度特征而得出。每个本征模态函数可以看作是信号中一个固有振荡模态,其对应于某个目标的固有振荡,希尔伯特黄变换用经验模态分解的方法将信号分解为多个本征模态函数,可以较好地将不同目标的能量分离。该变换得到的瞬时频率、瞬时能量和瞬时方位具有清晰的物理意义,信号的时间-频率-方位-能量的四维分布为多目标信号处理提供了丰富的信息量。与传统处理方法相比不仅有较好的目标分辨效果,而且适用于非线性和非平稳信号的处理。

Separation of multiple targets using a single vector sensor based on Hilbert-Huang transform

An application of Hilbert-Huang transform combining with the vector signal processing in underwater acoustics is introduced.The technique has many advantages in comparison with traditional methods.In Hilbert-Huang transform,the intrinsic mode function is derived from the data by their characteristic time scales.Each IMF is an intrinsic oscillatory mode,and it can be identified as the intrinsic oscillatory of one target.Hilbert-Huang transform decomposes the data into IMFs to better separate the energy of different targets.The instantaneous frequency,instantaneous energy and instantaneous azimuth obtained by HHT can be defined everywhere.The amplitude-frequency-azimuth-time distribution of the signal can provide rich information of targets.In comparison with traditional methods,the new method not only gives better result,but also fit for nonlinear and non-stationary signal processing.