中文摘要：

主要研究小波变换的分形方法在分析鼾音信号中的应用问题。首先分析信号波形特点并检测其瞬态脉冲。通过对频谱分形特性的动态分析,及时诊断病变信号的危害程度。仿真实验结果表明,针对典型的非平稳鼾音信号,与传统的傅氏变换方法相比,所提出的方法具有明显的优点。

英文摘要：

This paper addresses the problem of the analysis of snoring signal with fractals based on wavelet transform.Analysing characteristics of snores signal and detecting its instantaneous pulse are firstly presented.The status is that the variation of snores signal is diagnosed through dynamically analysing its spectrum fractals characteristics.The simulation experiment results show that the proposed method of snores signal based on wavelet transform is obviously advantageous over traditional Fourier transform for analysing the typical non-stable signal.