阶跃型奇异点的小波检测  

Detection of Step Singularity Based on Wavelet

作　　者：刘密歌[1] 李小斌[2] 

机构地区：[1]西安文理学院,陕西西安710065 [2]西安电子科技大学,陕西西安710071

出　　处：《计算机仿真》2010年第5期314-317,共4页Computer Simulation

基　　金：西安文理学院专业技术人员科研项目(KYC200724)

摘　　要：在检测实际系统信号问题上,阶跃型奇异点是一类重要的信号奇异点,针对检测和定位在许多实际问题上都有重要的意义。根据阶跃型奇异点的特征,证明了信号的阶跃奇异点与信号小波变换的最值有关,如果适当选择小波基函数,那么信号的阶跃奇异点将对应于信号小波变换的最值点。据此提出了一个利用小波变换最值的阶跃奇异点检测方法,并给出了小波基选择条件。最后利用上述方法对大量仿真信号及实际音频信号上的阶跃奇异点进行检测,结果表明研究有方法不仅有效可靠,而且具有计算简单,定位准确的特点。Step singularity is an important singular point in signal, for which detection and localization have significance. According to the characteristic of step singularity, it is showed that the step singularity in signal is related to the maximum absolute value of real wavelet transform. By using the wavelet base choosed carefully, the step singular point in signal exactly corresponds to the maximum absolute value point of the wavelet transform. As a resuh, a novel method for detecting the step singularity in signal is developed and choice condition of wavelet base is presented. Finally, by using the methed for detecting step singular points in many emulational signals and audio signal, the results not only show that the method can identify the sigularity from signal successfully but also have characters of simple computation and exact localization.

关 键 词：阶跃型奇异点 小波变换 最值点 检测方法 

分 类 号：TN911.6[电子电信—通信与信息系统]