激光探测水下振动特征的多尺度变换与时频分析

Multi-scale Transform and Time-frequency Analysis of Underwater Vibration Characteristic for Laser Detection

由于水下振动特征极其微弱,利用激光相干探测水表面信息,采用传统频谱分析的方法从水面信息反演出水下振动的特征是一项艰苦的工作。故在传统频谱分析的基础上,采用短时傅里叶变换和小波多尺度变换联合,可有效地增强从水表面信息中获取水下振源振动特征。通过对1 000 Hz振源放置水下80 cm处振动实验的分析。结果表明,水下振源的振动特征得到增强,有利于目标特征的提取。

Due to the extremely weak underwater vibration characteristics, with the water surface information detected through laser coherence method, it is a hard work to obtain the underwater vibration characteristics from the water surface information by using traditional spectral analysis. Therefore, on the basis of the traditional spectrum analysis, the underwater vibration characteristics obtained from water surface information can be enhanced effectively with short-time Fourier transform （STFT） and wavelet multi-scale transform. Through analyzing the vibration experiment of a vibration source with 1 000 Hz placed underwater 80 cm, the results show that the underwater vibration characteristics are enhanced, which is useful to extract the target characteristics.