摘要
应用非线性小波去噪技术对小波包能量法进行改进,提出了一种新的方法———去噪小波包能量法。该算法对信号小波去噪后,再用小波包能量法提取信号的特征。文中研究了不同的小波去噪方法对水声信号分类识别率的影响,在实测信号样本集上用BP神经网络进行了识别实验。结果显示了算法的有效性。
Improved the wavelet-packet energy method with nonlinear wavelet denoising technique and presented a new method:denoising wavelet-packet energy method.The arithmetic denoised the signal and then contracted the features of it with wavelet packet energy method.Different wavelet-based denoising standard were studied and their influence to recognition rate were given.Experiments with BP neural network showed the validity of this arithmetic.
出处
《微机发展》
2005年第4期21-23,共3页
Microcomputer Development
基金
广东省科技攻关项目(A1020103)
关键词
非线性小波去噪
小波包能量法
分类识别
神经网络
水声信号
nonlinear wavelet denoising
wavelet-packet energy method
classification
neural network
underwater acoustic signal
