摘要
对水声信号的识别,提出了高频小波能量法,并构造了一个自动识别系统模型。系统对信号进行小波分解后,取高频子空间能量作为特征向量值进行信号的识别。高频小波能量法可采取小波包与小波两种变换,通过对两种方法的比较显示:高频小波包能量法构造的识别系统能够更好地识别低信噪比水声信号。
To recognize underwater acoustic signals, high frequency wavelet energy method is presented and a recognition system is designed. The system pick-up high fi-equency energy as eigenvector with wavelet transform. Both wavelet pocket and wavelet transform in the system are taken. The result in comparison between two methods show that system made of high fi＇equency wavelet-pocket energy method can recognize lower-SNR better.
出处
《计算机工程与设计》
CSCD
北大核心
2006年第12期2187-2188,2195,共3页
Computer Engineering and Design
基金
广东海洋大学自然科学基金项目(2005)
关键词
水声信号
小波
小波能量
特征向量
识别系统
underwater acoustic signal
wavelet
wavelet energy
eigenvector
recognition system
