首页 | 本学科首页   官方微博 | 高级检索  
     

一种新的自适应消噪方法
引用本文:司锡才,张雯雯,李利,柴娟芳. 一种新的自适应消噪方法[J]. 宇航学报, 2008, 29(6): 2013-2018. DOI: 10.3873/j.issn.1000-1328.2008.06.062
作者姓名:司锡才  张雯雯  李利  柴娟芳
作者单位:哈尔滨工程大学信息与通信工程学院,哈尔滨 150001
基金项目:收稿日期:20070703; 修回日期:20081006
摘    要:基于小波变换的自适应消噪方法为雷达信号的滤波提供了一种可行的办法。考虑到离 散小波变换(DWT)不具有平移不变性,而静态小波变换(SWT)又不能很好地分析信号的高频 部分,提出了一种新的自适应消噪方法。它根据CWT的提升方法,得到静态小波包的提升 实现方法,并设计出适合本系统的确定最优小波包分解树的相应步骤,利用引入了更多动量 因子的权系数迭代公式对各子带进行自适应匹配,并将匹配结果二次自适应,得到拟合的原 信号。仿真中将其与其它两种基于小波的自适应方法进行了比较,表明该方法可在适当减小 计算量的同时,进一步改善系统的滤波性能。

关 键 词:自适应滤波  静态小波包变换  提升框架  平移不变  
收稿时间:2007-07-03

A New Method of Adaptive Noise
SI Xi-cai,ZHANG Wen-wen,LI Li,CHAI Juan-fang. A New Method of Adaptive Noise[J]. Journal of Astronautics, 2008, 29(6): 2013-2018. DOI: 10.3873/j.issn.1000-1328.2008.06.062
Authors:SI Xi-cai  ZHANG Wen-wen  LI Li  CHAI Juan-fang
Abstract:The method of adaptive denoising based on wavelet transform provides a feasible solution for radar signal filtering,seeing that the discrete wavelet transform(DWT) has not the characteristic of translation invariance of the wavelet coefficients,and the static wavelet transform(SWT) can not effectively analyze the high-frequency part of signals.This paper proposes a new adaptive denoising method.It educed the lifting method of static wavelet packet based on the lifting method of DWT,designed the corresponding steps of determining the optimal wavelet packet trees which is Suitable for this system,carried on adaptive matching to each sub-bands using the Weighting coefficient iterative formula which has more momentum factors.Finally,it took the second adaptive filter of the matching results to acquire the fitted signal.Compared with the other two adaptive methods based on wavelet transform,simulation results show that the method can further improve the filtering performance and properly decrease the calculation as well.
Keywords:Adapted filter  Stationary wavelet packet transform  Lifting scheme  Translation invariance  
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《宇航学报》浏览原始摘要信息
点击此处可从《宇航学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号