基于EEMD的多传感器自适应噪声滤波技术
Adaptive noise filtering technology of multi-sensor based on EEMD
吕艳新 1金珊 2陈一伟 1卫东 1石红 1黄伟杰1
作者信息
- 1. 生态环境部核与辐射安全中心,北京100082
- 2. 中国核电工程有限公司,北京100840
- 折叠
摘要
针对多传感器采集所得被动信号中存在噪声使得其应用受到限制对的问题,提出一种基于集合经验模态分解(EEMD)的多传感器信息融合自适应噪声滤波技术.首先,构建一个自适应噪声滤波系统.采用EEMD方法对多传感器信号进行分解,结合各传感器的置信度对分解后的信号进行加权,融合多传感器的信息,得到自适应滤波器的参考噪声信号.将参考信号输入自适应滤波器,采用最小均方(LMS)算法,使得自适应滤波器输出信号在LMS意义下,抵消噪声信号,从而得到最终的滤波信号.最后,将该方法应用于实测两类目标的多传感器信号,得到了较为满意的降噪效果.
Abstract
Aiming at the problem that passive signals acquired by multi-sensor exist noise which limit its applications,an adaptive noise filtering technology of multi-sensor information fusion based on ensemble empirical mode decomposition (EEMD )is proposed.Firstly,an adaptive noise filtering system is constructed.Then,to decompose the signals of multi-sensor by EEMD,to weight the decomposed signals by multi-sensor integrated support degree,the reference noises were gained by multi-sensor information fusion.To put reference noises into adaptive noise canceling system,making the output signals of adaptive noise canceling implement to cancel the noises of signals by LMS,in order to get the final denoising signals.Finally,the proposed method is applied to multi-sensor signals of two targets,and obtains satisfactory denoising results.
关键词
集合经验模态分解/自适应/多传感器/最小均方/滤波Key words
EEMD/adaptive/multi-sensor/LMS/filtering引用本文复制引用
出版年
2024
NETL
NSTL
万方数据