基于连续小波变换的简正波检测与提取技术  

Normal modes detection and extraction based on wavelet analysis

作　　者：林京[1] 张仁和[1] 李整林[1] 张爱民[1] 

机构地区：[1]中国科学院声学研究所声场声信息国家重点实验室,北京100080

出　　处：《应用声学》2005年第4期221-226,共6页Journal of Applied Acoustics

摘　　要：宽带声源在远距离传播中的频散过程反映着海洋环境特性的重要信息,准确提取这一过程中各简正波的有关信息是有效实现海洋声场反演的重要前提。小波变换虽然已被用于获取宽带声源的时频分布,但不具备适用于不同信号的柔性。在此背景下,本文介绍了一种自适应基小波方法,可适应不同形式的爆炸声信号,从而有效实现了频散现象的获取。在此基础上,本文同时建立了各号简正波的重构方法,经仿真测试,重构出的简正波与实际波形几乎相同。The dispersion of a broadband acoustic source carries important information about the properties of the acoustic field. An accurate estimate for each mode is the basis to achieve an efficient inversion for the acoustic field. The usage is lack of sufficient flexibility although continuous wavelet transform has been used to obtain time-frequency distributions for explosions. An adaptive method is introduced to improve the flexibility of the basic wavelet under this background, which could provide a clear display of the dispersion phenomena for different explosions. Based on that, an efficient method is proposed to extract the normal modes. This method shows good performance by using a simulated record of explosion propagation.

关 键 词：简正波 连续小波变换 提取技术 检测 远距离传播 环境特性 海洋声场 时频分布 小波方法 频散现象 重构方法 仿真测试 自适应 声信号 信息 声源 宽带 

分 类 号：TN911.7[电子电信—通信与信息系统] P733.21[电子电信—信息与通信工程]