摘要
为实现低信噪比情况下微弱的船舶轴频电场信号的有效检测,提出了一种结合经验模态分解(empirical mode decomposition,EMD)和四阶混合累积量对角切片滑动功率谱的方法。首先,利用EMD将信号自适应地进行子带分解,对得到的本征模态函数(intrinsic mode functions,IMF)采用相关系数准则进行筛选;然后,利用高阶累积量可抑制高斯色噪声的特性,计算各有效IMF分量的四阶混合累积量对角切片的功率谱,并进行了多子带中的滑动检测。实测数据处理结果表明:该方法具有较好的应用价值。
In order to implement effective detection of weak ship shaft-rate electric field signal under low signal to noise ratio （SNR）, a method combining empirical mode decomposition （EMD） with sli- ding power spectrum of fourth-order mixed cumulant diagonal slice is proposed. Firstly, EMD method is employed to adaptively decompose the signal into a set of intrinsic mode functions （IMFs）, from which the valid ones are selected according to correlation coefficient criterion. Then, by exploiting the property of higher order cumulant which can suppress Gaussian colored noise, power spectrum of fourth-order mixed cumulant diagonal slice of selected IMFs is caleulated, which is then used for slid- ing detection in multiple sub-bands. The result of processing practical data illustrates that this method is of great value in application.
出处
《海军工程大学学报》
CAS
北大核心
2015年第6期21-26,共6页
Journal of Naval University of Engineering
基金
国家自然科学基金资助项目(51109215)
关键词
轴频电场
经验模态分解
四阶混合累积量
滑动功率谱
信号检测
shaft-rate electric field
empirical mode decomposition
fourth order mixed cumulant
sliding power spectrum
signal detection
