脉冲星信号的经验模态分解模态单元比例萎缩消噪算法  

Pulsar signal denoising method based on empirical mode decomposition mode cell proportion shrinking

作　　者：王文波[1,2] 张晓东[3] 汪祥莉[4] 

机构地区：[1]武汉科技大学信息与计算科学系,武汉430065 [2]遥感科学国家重点实验室,北京100101 [3]国家海洋局第二研究所,卫星海洋环境动力学国家重点实验室,杭州310012 [4]武汉理工大学计算机科学与计算学院,武汉430063

出　　处：《物理学报》2013年第6期526-534,共9页Acta Physica Sinica

基　　金：国家自然科学基金(批准号:41071270;11201354);测绘遥感信息工程国家重点实验室开放基金(批准号:11R01);遥感科学国家重点实验室开放基金(批准号:OFSLRSS201209);中央高校基本科研业务费专项基金(批准号:2012-IV-043);湖北省自然科学基金(批准号:2010CDB03305);武汉市晨光计划(批准号:201150431096)资助的课题~~

摘　　要：针对脉冲星信号的消噪问题,提出了一种基于模态单元比例萎缩的经验模态分解(EMD)消噪方法.利用经验模态分解将含噪脉冲星信号分解为一组内蕴模态函数(IMF),将IMF中两个过零点间的部分定义为模态单元,以模态单元为基本单位构造最优比例萎缩因子,对IMF中的每个模态单元进行比例萎缩去噪,进而建立基于模态单元比例萎缩的脉冲星信号滤波模型.对含噪脉冲星信号进行了消噪实验分析,实验结果表明,与小波硬阈值消噪法、比例萎缩小波消噪法和基于模态单元阈值的EMD消噪法相比,该方法可以更有效地去除脉冲星信号中的噪声,同时更好地保留了原信号中的有用细节信息.In order to improve the denoising quality of the pulsar signal, an empirical mode decomposing method （EMD） of pulsar signal denoising based on mode cell proportion shrinking is proposed. Firstly, the pulsar signal is decomposed into a series of intrinsic mode functions （IMF）, and the part between the two adjacent zero-crossing within IMF is defined as a mode cell. Then, the optimal proportional shrinking factor is constructed by treating mode cell as the basic unit of analysis. Finally, the all mode cells within IMF are denoised by proportion shrinking, and the mode cell proportion shrinking denoising model is established. The experimental results show that compared with the two EMD denoising algorithms based on coefficient threshold and mode cell threshold, the proposed method can more effectively remove the pulsar signal noise, with better preserving the useful detail information in the original signal.

关 键 词：经验模态分解 脉冲星信号 模态单元比例萎缩 消噪 

分 类 号：P145.6[天文地球—天体物理]