基于经验模态分解空域相关滤波的脉冲星信号消噪  

Pulsar signal denosing based on empirical mode decomposition spatial correlation filter

作　　者：赵攀[1] 王文波[1] 龚邦明[1] 常毓禅 张童硕 

机构地区：[1]武汉科技大学理学院,武汉430065 [2]中国人民大学财政金融学院,北京100872 [3]电子科技大学数学科学学院,成都611731

出　　处：《中国科技论文》2015年第5期592-596,607,共6页China Sciencepaper

基　　金：国家自然科学基金资助项目(11201354)

摘　　要：针对脉冲星信号信噪比极低,且信号中有色噪声成分远大于高斯白噪声的特点,提出了一种基于经验模态分解(empirical mode decomposition,EMD)的脉冲星信号空域相关滤波消噪方法。首先建立脉冲星信号的分形高斯噪声模型,根据双对数功率谱密度估计分形高斯噪声的Hurst参数值;然后,结合脉冲星累积信号的窗口辐射特性和分形高斯噪声经EMD分解后的噪声分布特点,估计脉冲星信号各层内蕴模态函数(intrinsic mode function,IMF)中噪声的方差;最后,以IMF噪声方差为基础,建立EMD空域相关滤波方法并应用于脉冲星累积轮廓消噪。实验结果表明,与经典的平移不变小波阈值消噪法、平移不变小波空域相关滤波消噪法以及EMD模态单元比例萎缩消噪法相比,改进的EMD空域相关滤波消噪法在抑制脉冲信号中的噪声和保留更多的微脉冲特征细节信息方面更加有效。Because of extremely low signal noise ratio（SNR）of pulsar signals and predominant colored noises,a new spatial correlation de-noising approach based on empirical mode decomposition（EMD）is improved.Firstly,we build the fractional Gauss noise model of pulsar signal to achieve Hurst parameters of fractional Gauss noise according to double logarithmic power spectrum density.Then,we estimate noise variance of intrinsic mode function（IMF）according to the windows radiation characteristics of cumulative signal and the noise distribution characteristics of decomposed fractional Gauss noise by EMD.Finally,EMD spatial correlation filtering method is established and applied to the pulsar cumulative profile de-noising.Experiments show that the proposed approach can preserve more details compared with wavelet threshold de-noising,wavelet spatial correlation filtering de-noising and EMD mode cell proportion shrinking de-noising.

关 键 词：脉冲星 经验模态分解(EMD) 消噪 分形高斯噪声 有色噪声 空域相关滤波 

分 类 号：TN911.4[电子电信—通信与信息系统]