摘要
针对MEMS陀螺信号,提出了一种基于变分模态分解(VMD)和样本熵(SE)的随机噪声抑制方法。首先采用VMD算法自适应地将原始信号分解为固有模态函数(IMF)的集合,然后针对不同的固有模态序列建立基于样本熵理论的信号组分筛选标准,从而将其划分为低频有效信息IMFs、信息和噪声混合IMFs和高频噪声IMFs。舍弃高频噪声IMFs,并利用软区间阈值降噪方法实现对混合分量的进一步处理,最后通过重构得到最终的信号。对一组真实的MEMS陀螺静态漂移输出数据进行实验分析,比较结果表明该算法的去噪性能优于同为模态分解的EMD去噪方法。
For MEMS gyro signals,a random noise suppression method based on variational mode decomposition(VMD)and sample entropy(SE)was proposed.First,the VMD algorithm was used to adaptively decompose the original signal into a set of intrinsic mode functions(IMF),and then to establish a signal component screening standard based on the sample entropy theory for different intrinsic modal sequences,thereby dividing it into low-frequency effective information IMFs,information and noise mixed IMFs and high-frequency noise IMFs three different parts.Abandon high-frequency noise IMFs,and use soft interval threshold denoising method to achieve further processing of the mixed components,and finally obtain the final signal through reconstruction.An experimental analysis of a set of real MEMS gyroscope static drift output data was performed.The comparison results show that the algorithm s denoising performance is better than the modal decomposition EMD denoising method.
作者
刘洋
李杰
张德彪
冯凯强
鲁正隆
LIU Yang;LI Jie;ZHANG De-biao;FENG Kai-qiang;LU Zheng-long(Key Laboratory of Instrumental Science and Dynamic Testing of Ministry of Education,North University of China,Taiyuan 030051,China)
出处
《仪表技术与传感器》
CSCD
北大核心
2021年第6期90-94,共5页
Instrument Technique and Sensor
基金
国家自然科学基金面上项目(61973280)
中国博士后科学基金(2019M661069)。
关键词
MEMS陀螺仪
去噪算法
变分模态分解
信号重构
区间阈值降噪
样本熵
MEMS gyroscope
denoising algorithm
variational mode decomposition
signal reconstruction
interval threshold denoising
sample entropy
作者简介
刘洋(1995-),硕士研究生,主要研究方向为组合导航算法、微系统集成。E-mail:lylyly357@163.com;通信作者:李杰(1976-),教授,博士,主要从事捷联惯导和智能信息处理等方面的研究。E-mail:lijie@nuc.edu.cn。
