Abstract: According to the characteristic that short-time strong disturbance in signals can lead to obvious abnormity in instantaneous amplitudes and frequencies, an HHT-based method to eliminate the disturbance is proposed. The signal with short-time strong disturbance is decomposed into a series of Intrinsic Mode Functions（IMF） and a residue by the Empirical Mode Decomposition（EMD）. The instantaneous amplitudes and frequencies of each IMF are calculated. In abnormal sections, they are fitted according to the data in normal sections, replacing the original ones with the fitted data. A new set of IMF is reconstructed by using the processed instantaneous amplitudes and frequencies. For the residue, abnormal fluctuations are directly removed. A new signal with the short-time strong disturbance eliminated is reconstructed by superposing all the new IMF and the residue. Results of numerical simulation and measured signal verify the feasibility of the method.

Key words: short-time strong disturbance, Hilbert-Huang transform, empirical mode decomposition, instantaneous amplitude, instantaneous frequency

摘要： 针对信号中含有的短时强干扰会引起瞬时振幅和瞬时频率明显异常的特点，基于Hilbert-Huang变换（HHT）提出一种消除方法。利用经验模态分解（EMD）将存在短时强干扰的信号分解成本征模函数（IMF）和残余项；计算每阶IMF的瞬时振幅和瞬时频率，在异常区段，将其根据正常区段的数据进行拟合，用拟合后的数据代替原数据。将处理后的瞬时振幅和瞬时频率重构得到一组新的IMF，对残余项，直接去除异常波动。将所有新的IMF和残余项重新组合，所得信号就消除了干扰的影响。数值仿真和实测数据处理结果表明了该方法的可行性。

关键词: 短时强干扰, Hilbert-Huang变换, 经验模态分解, 瞬时振幅, 瞬时频率