小波变换对突变信号峰值奇异点的精确检测

ACCURATE DETECTION FOR PEAKVALUE SINGULAR POINTS OF THETRANSIENT SIGNAL WITH WAVELETTRANSFORMS

研究了小波变换对突变信号峰值奇异点的精确检测机理和方法,采用了离散小波变换的直接算法,避免了塔式算法在本检测方法中的某些局限性。通过模拟算例和应用实例的验证,证明即使在有严重噪声干扰的情况下,该方法也很容易实现对突变信号峰值奇异点的准确检测和精确定位,具有相当高的定位精度(其误差不大于1个采样点)和分析精度(不存在累积误差),同时具有较高的实时性。

The principium of the accurate detection for the peak value singular points of the transient signal with wavelet transforms is researched. The direct algorithm of discrete wavelet transforms is applied to detect the location of peak value singular points of the transient signal, thereby avoiding the limitation of the pyramids algorithm within this accurate detection. Some examples are presented which can demonstrate superiority of the method. This method can detect accurately and localize precisely to the peak value singular points of the transient signal in the serious noise signal, and has the quite high locating precision (its error is not more than one sampling point) and analyzing accuracy (no accumulative error). The real-time of this method is excellent, and the real-time detection to the peak value singular points of the transient signal can be achieved using this method.