摘要
大地电磁(MT)勘测方法因存在多种人文噪声使得某些频段受噪声影响显得更为明显,提出了一种基于频率切片小波变换时频分析的时频域MT信号去噪新方法。给出了方法原理和实现步骤,先对MT信号进行频率切片小波变换,得到全频带下的时频分布,在此基础上对时频谱进行时频阈值滤波和逆变换,重构分离出去噪后的MT信号。仿真实验和实测信号分析结果表明,该方法可有效消除大地电磁信号中的强噪声;去噪后计算的响应参数曲线的突变点得到了有效抑制,曲线变得平滑、连续。
Magnetotelluric （MT） survey methods always affected by a variety of artificial noises, which is more obvious in some frequency bands. In order to solve this problem, an improved denosing method for magnetotelluric data in time-frequen- cy domain was proposed based on time-frequency analysis of the frequency slice wavelet transform （FSWT）. The principle and implementation steps of the method were presented. Firstly, the MT signal was processed with the FSWT to get its time- frequency distribution for full frequency bands. Then the derived time-frequency spectrum was applied with the threshold fil- tering and the denoised signal was reconstructed by the inverse FSWT of the processed time-frequency distribution. The sim- ulation experiment and the measured signal analysis show that the proposed method can effectively eliminate the strong noise of MT signal. After denoising, the mutation points of the calculated response parameters curve were suppressed and parame- ter curves become smooth and continuous.
出处
《石油物探》
EI
CSCD
北大核心
2016年第6期904-912,共9页
Geophysical Prospecting For Petroleum
基金
国家自然科学基金项目(41304098)
湖南省教育厅重点项目(16A146)共同资助~~
关键词
频率切片小波变换
时频分析
大地电磁信号
去噪
frequency slice wavelet transform, time-frequency analysis, magnetotelluric （MT） signal, denosing
