The Neural Network Adaptive Filter Model Based on Wavelet Transform

摘要：

Due to the problem that the noise in the noisy signal can not be predicted in many practical fields, we have proposed an adaptive filter based on wavelet transform method. As the adaptive filter has the characteristic of eliminating noise no use to predict the priori knowledge of the noise in the signal, we have taken the signal after the first wavelet threshold denoising as the main input of the adaptive filter, meanwhile taken the wavelet reconstruction coefficients after the second wavelet transform as the reference input of the adaptive filter. And a neural network adaptive filter model based on wavelet transform is constructed. The model has applied the Hopfield neural network to implement the adaptive filtering algorithm LMS, so as to improve the computation speed. The simulation results show that the neural network adaptive filter model based on wavelet transform can achieve the best denoising effect.

关键词： Wavelet Transform Denoising Hopfield Neural Network Adaptive Filter Model Weight

作者: XIAO Qian GE Gang WANG Jianhui

作者单位: Key Laboratory of Process Industry Automation, Northeastern University, Ministry of Education, Sheny State Key Laboratory of Rolling and Automation, Northeastern University, Shenyang 110004, China

会议类型: 国际会议

会议名称: 2009 Ninth International Conference on Hybrid Intelligent Systems(第九届混合智能系统国际会议 HIS 2009)

会议地点: 沈阳

会议语种:英文

页码: 1-6

在线出版日期: 2009-08-12（万方平台首次上网日期，不代表论文的发表时间）