蝙蝠回声定位信号的小波分析

【作者】 秦琴；

【导师】 王双维；

【作者基本信息】 东北师范大学 ， 电路与系统， 2003， 硕士

【摘要】 蝙蝠回声定位的研究在国际上一直是热点和重点，因为蝙蝠特殊的生态位及声行为的独特性，决定了蝙蝠回声定位行为的生态学研究具有重要意义。目前对蝙蝠回声定位信号的研究方法主要局限于传统的傅立叶分析。而小波分析则突破了传统的傅立叶分析方法，利用其在时域和频域上同时具有的良好的局部性，成为信号分析的一个有利的工具。本文的工作就是将两者结合在一起，为蝙蝠回声定位信号领域的研究提供一种新的、有效的方法。 本文将离散小波变换应用于蝙蝠回声定位叫声的分析中，利用小波的伸缩和时移特性，通过对小波尺度的适当离散，解决了传统语谱分析的分辨率固定、信息冗余量大的问题。通过MATLAB仿真证明，小波变换分析结果不仅没有丢失原有信息，而且能够更加直观、准确地刻画信号的特性，提取信号的特征参量。从分析结果中我们能够发现一些蝙蝠回声定位信号的新的特征信息，这对于进一步寻找蝙蝠新的生态学特征有着重要意义。因此，小波分析是分析蝙蝠回声定位叫声的一种有效的新途径，有其广泛的应用价值。更多还原

【Abstract】 The study of the echolocation calls of bat is hotspot and emphases in the world. The speciality of the ecological niche and the calls of bat make it very important to study the echolocation calls of bats. Presently the way of study the echolocation calls of bat is localizated in traditional Fourier analysis. Wavelet analysis because of its good retractility has become an advantaged tool in signal analysis field. We provide a new valid way for studying the echolocation calls of bat by using this tool.This paper applies discret wavelet transform (DWT) to the analysis of echolocation calls of several bats. Taking advantage of retractility and time-delay characteristics of wavelet, this paper solves the problems of fixed resolving power and large redundance of the information caused by traditional sonogram analysis by choosing the scale of wavelet correctly. The computations of the signals were made on a computer in a MATLAB environment. The results of DWT can describe the character of the signal more concisely and more precisely than traditional Fourier analysis. We can find some new character of the echolocation calls of bat, and it is very important for search for new character of bats in animal ecology . DWT is testified to be a new effective way in analyzing the echolocation signals of bats.更多还原