一种基于声阵列信息融合及改进EEMD的信号降噪方法
发布时间：2018-09-19 12:19
【摘要】：针对声阵列多通道信号的去噪问题,提出一种基于多传声器信息融合辅助的改进总体平均经验模态分解(Ensemble Empirical Mode Decomposition,EEMD)的被动声信号去噪方法。对标准EEMD进行改进,通过多通道信号频谱分析,选取多传声器信号最小有效频率作为各通道信号EEMD分解的筛选截止频率,采用改进的EEMD算法将原始信号快速分解为完备的IMF分量,有效抑制了模态混叠现象并提高信号分解效率;引入声阵列时延矢量封闭准则(Time Delay Vector Close Rule,TDVCR)概念,结合多传声器数据一致性融合及信号相关性理论,对各IMF分量进行相应的权重计算,再由已确定权值对各IMF分量进行加权重构得到去噪信号;最终通过半实物仿真试验以及同传统EMD去噪的比较验证了该算法在多通道信号去噪中的有效性和实用性。
[Abstract]:A passive acoustic signal de-noising method based on multi-microphone information fusion assisted by improved population average empirical mode decomposition (Ensemble Empirical Mode Decomposition,EEMD) is proposed for the de-noising of acoustic array multi-channel signals. The standard EEMD is improved. The minimum effective frequency of the multi-microphone signal is selected as the screening cutoff frequency of the EEMD decomposition of each channel signal through the multi-channel signal spectrum analysis. The improved EEMD algorithm is used to decompose the original signal into complete IMF components quickly, which effectively reduces the mode aliasing phenomenon and improves the signal decomposition efficiency. The concept of acoustic array time-delay vector closure criterion (Time Delay Vector Close Rule,TDVCR) is introduced. Combining the data consistency fusion of multi-microphone and the theory of signal correlation, the corresponding weight of each IMF component is calculated, and then the de-noising signal is obtained by the weighted reconstruction of each IMF component with the determined weight value. Finally, the effectiveness and practicability of the algorithm in multi-channel signal denoising are verified by the hardware-in-the-loop simulation experiment and the comparison with the traditional EMD de-noising algorithm.
【作者单位】： 南京理工大学机械工程学院;贵州大学智能信息处理研究所;
【基金】：国家自然科学基金(61263005)
【分类号】：TB535
本文编号：2250105
[Abstract]:A passive acoustic signal de-noising method based on multi-microphone information fusion assisted by improved population average empirical mode decomposition (Ensemble Empirical Mode Decomposition,EEMD) is proposed for the de-noising of acoustic array multi-channel signals. The standard EEMD is improved. The minimum effective frequency of the multi-microphone signal is selected as the screening cutoff frequency of the EEMD decomposition of each channel signal through the multi-channel signal spectrum analysis. The improved EEMD algorithm is used to decompose the original signal into complete IMF components quickly, which effectively reduces the mode aliasing phenomenon and improves the signal decomposition efficiency. The concept of acoustic array time-delay vector closure criterion (Time Delay Vector Close Rule,TDVCR) is introduced. Combining the data consistency fusion of multi-microphone and the theory of signal correlation, the corresponding weight of each IMF component is calculated, and then the de-noising signal is obtained by the weighted reconstruction of each IMF component with the determined weight value. Finally, the effectiveness and practicability of the algorithm in multi-channel signal denoising are verified by the hardware-in-the-loop simulation experiment and the comparison with the traditional EMD de-noising algorithm.
【作者单位】： 南京理工大学机械工程学院;贵州大学智能信息处理研究所;
【基金】：国家自然科学基金(61263005)
【分类号】：TB535
【相似文献】
相关期刊论文 前1条
1 林向真,明代都;SF-10深海复合棒振子的设计及其布阵应用[J];声学学报;1979年03期
相关会议论文 前1条
1 陈宝柱;;水声阵元域标准测试数据库构建研究[A];中国声学学会水声学分会2013年全国水声学学术会议论文集[C];2013年
相关博士学位论文 前1条
1 蒋佳佳;水声阵列探测算法及系统关键技术研究[D];天津大学;2014年
相关硕士学位论文 前1条
1 李会超;声阵列超指向性测量机理及方法研究[D];青岛理工大学;2016年
，本文编号：2250105
本文链接：https://www.wllwen.com/guanlilunwen/gongchengguanli/2250105.html
