基于STFT和EMD的多普勒信号分离算法  

A Doppler Signal Separation Algorithm Based on STFT and EMD

作　　者：杨彦利[1] 马德 权建峰[2] YANG Yanli;MA De;QUAN Jianfeng(Tianjin Key Laboratory of Optoelectronic Detection Technology and Systems,Tianjin Polytechnic University,Tianjin 300387,China;Xi'an Institute of Electromechanical Information Technology,Xi'an 710065,China)

机构地区：[1]天津工业大学光电检测技术与系统重点实验室,天津300387 [2]西安机电信息技术研究所,陕西西安710065

出　　处：《探测与控制学报》2018年第4期122-125,132,共5页Journal of Detection & Control

基　　金：国家自然科学基金项目资助(61401305)

摘　　要：针对近场区多普勒信号的识别问题,提出了基于短时傅里叶变换(STFT)和经验模态分解(EMD)的多普勒信号分离算法。该算法利用STFT时频分布,通过极值滤波法并结合EMD算法将回波信号分解成若干个窄带子信号,实现了对多普勒信号的分离和多普勒信号瞬时频率的估计。实测验证表明,该算法能够将频率成分相近的多普勒回波信号分解成若干个窄带的子信号,有助于实现对近场区多普勒信号的提取和识别。To solve the near-field Doppler signal identification problem,a Doppler signal separation algorithm based on short-time Fourier transform(STFT)and empirical mode decomposition(EMD)was proposed.By using the time-frequency distribution of STFT,this algorithm decomposed the Doppler echo signal into several narrow-band sub-signals through the extreme filtering method combined with the EMD.Then,the Doppler signals separation and the instantaneous frequency estimation of Doppler signals were achieved.The experimental results showed that the proposed algorithm can decomposed the Doppler echo signals with similar frequency components into several narrow-band sub-signals,which was helpful to extract and recognize Doppler signals in the near-field region.

关 键 词：多普勒信号检测 体目标效应 瞬时频率 短时傅里叶变换 经验模态分解 

分 类 号：TN911.72[电子电信—通信与信息系统]