利用EMD滤波的特征提取方法

详细信息    查看官网全文

• 英文论文题名：Feature extraction method by the EMD filter
• 论文作者：孙微 ; 黄强 ; 王方勇
• 英文论文作者：SUN Wei ; HUANG Qiang ; WANG Fang-yong ; Hangzhou Applied Acoustics Reasearch Institute ; Science and Technology on Sonar Laboratory
• 年：2016
• 作者机构：声呐技术重点实验室;
• 论文关键词：特征提取 ; 经验模态分解 ; 概率密度函数 ; 能量熵
• 英文论文关键词：Feature extraction ; empirical mode decomposition ; probability density function ; energy entropy
• 会议召开时间：2016-08-21
• 会议录名称：2016’中国西部声学学术交流会论文集
• 语种：中文
• 分类号：TB56
• 学会代码：SXJS
• 会议名称：2016’中国西部声学学术交流会
• 会议地点：中国四川成都
• 主办单位：中国声学学会微声学分会、中国声学学会物理声学分会、陕西省声学学会、浙江省声学学会、山东省声学学会、上海市声学学会、黑龙江省声学学会、水声技术重点实验室、重庆市声学学会、西安市声学学会、四川省声学学会
• 学会名称：《声学技术》编辑部
• 页数：5
• 文件大小：563k
• 原文格式：D
• 会议级别：全国

摘要

本文介绍了一种主动目标回波特征提取方法,该方法首先通过对目标回波进行经验模态分解(EMD),得到回波的多个本征波动模态(IMF)。然后估计每个本征模态的概率密度函数,计算各个本征模态的概率密度函数与输入回波信号的概率密度函数的相似度,通过相似度测量结果提取出信号的主模态。将提取出来的主模态转化成能量特征向量,由能量特征向量计算出重建信号的能量熵,实现对目标的分类识别。通过仿真实验和对实测数据进行处理,验证了利用EMD滤波提取本征模态能量熵可以进行有效的目标分类识别。
This paper describes an active target echo feature extraction method by which the target echo Empirical Mode Decomposition(EMD),to obtain a plurality of intrinsic mode function(IMF). The paper estimated the probability density function(pdf) of each extracted mode,calculating the similarity between the pdf of the input signal and that of each mode. The relevant modes being selected by the similarity measure, which is used for signal partial reconstruction. The extracted relevant modes are converted into energy feature vectors,calculating energy entropy of the reconstructed signal by it, to achieve targets classification and recognition. In this paper,numerical simulation and experimental result showed the effectiveness of EMD filter and intrinsic mode energy entropy.

引文

[1][梁国龙.回波信号瞬时参数序列分析及其应用研究[D].哈尔滨:哈尔滨工程大学,1997.Liang Guolong.A Study on Instantaneous Parameter Sequence Analysis of Echoes and Its Application[D].Harbin:Harbin Engineering University,1997.
    [2]JC Cexus,AO Boudraa.Teager-Huang analysis applied to sonar target recognition[J].World Academy of Science,Engineering and Technology,2007;1(2):321-325.
    [3]Ali Komaty,Abdel-Ouahab Boudraa,Benoit AUGIER,Delphine DARE.EMD-Based Filtering Using Similarity Measure Between Probability Density Functions of IMFs[J].IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT,2014;63(1):27-34.
    [4]D Trad,T Al-Ani,M Jemni.A feature extraction technique of EEG based on EMD-BP for motor imagery classification in BCI[C].2015 5th International Conference on Information&Communication Technology and Accessibility(ICTA).Marrakech:IEEE,2015;1-6.
    [5]Li Lin,Ji Hongbing.Signal feature extraction based on an improved EMD[J].Measurement,2009;42(5):796–803.
    [6]胡桥,郝保安,吕林夏,陈亚琳,孙起.一种新的水声目标EMD能量熵检测方法[J]鱼雷技术,2007;15(6):9-12.Hu Qiao,Shao Baoan,Lv Linxia,Chen Yalin,Sun Qi.Target Detection Method of Underwater Acoustic Signals Based on Energy Entropy of Empirical Mode Decomposition[J].TORPEDO TECHNOLOGY,2007;15(6):9-12.