基于经验模态分解的大地电磁资料人文噪声处理

摘要

将经验模态分解(Empiricalmodedecomposition，EMD)方法应用到大地电磁资料的人文噪声处理中，根据人文噪声的不同来源和特征，提出基于EMD的时空滤波器或硬(软)阈值对噪声进行抑制的方法。给出经验模态分解去噪方法的原理和步骤，并对实测大地电磁信号中常见的脉冲干扰、矩形干扰和周期正弦噪声等人文干扰进行消噪处理。研究结果表明：本文提出的噪声改正方法是有效的，突出了有用信号的信息，改善了受干扰大地电磁数据的质量。%The empirical mode decomposition (EMD) method was applied to eliminate the human noise of magnetotelluric(MT) data. Considering the statistic feature and different sources of human noise, some methods using the time-space filters or threshold method to suppress the noise were proposed based on EMD. The principle and steps of method were given, and some human noises, such as impulse jamming, rectangle disturbing and sine wave noise, were processed for the actual MT data. The results show that noise has successfully suppressed and the useful information about MT data is enchanced. The quality of MT data is improved greatly.