Abstract:PreviousNext No AccessSEG Technical Program Expanded Abstracts 2017Power frequency noise suppression of magnetotelluric data based on frequency-domain sparse decompositionAuthors: Guang LiJingtian TangCong ZhouXiao XiaoShangshi WenGuang LiCentral South UniversitySearch for more papers by this author, Jingtian TangCentral South UniversitySearch for more papers by this author, Cong ZhouCentral South UniversitySearch for more papers by this author, Xiao XiaoCentral South UniversitySearch for more papers by this author, and Shangshi WenCentral South UniversitySearch for more papers by this authorhttps://doi.org/10.1190/segam2017-17638915.1 SectionsAboutPDF/ePub ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InRedditEmail Abstract Power frequency noise is one of the most common noises during the acquisition of magnetotelluric signal. Though there are trap circuits designed in most of the acquisition devices to suppress the power frequency noise, however, the fundamental frequency of the power frequency noise will fluctuate with the changing of load current, while the center frequency of the trap circuits are fixed, hence the magnetotelluric data are still seriously disturbed by the power frequency noise. Methods to isolate and remove the power frequency noise are typically based on remote reference processing, time-series editing, or data segments selecting based on statistic parameters. However, the amplitude of the odd harmonic signal is obviously decreased with the increasing of frequency, so it is difficult to identify power frequency noise from time domain accurately. Practical experience shows that it is difficult to choose an excellent remote reference site, when the magnetotelluric data are disturbed by the power frequency noise, remote reference processing is usually rarely work. Since the power system is fixed, the power frequency noise usually appear in the whole acquisition process, therefore data segments selecting also cannot suppress the noise. This paper proposes an alternative approach for power frequency noise removal based on frequency domain sparse decomposition. Firstly, the Fourier transform of the acquired magnetotelluric signal is performed. Subsequently, a redundant dictionary which matches with the power frequency noise but insensitive to the useful signal is designed, power frequency noises are separated by using of the dictionary and a greedy algorithm called IOMP. Finally, the frequency domain data are performed an inverse Fourier transform. Examples from Lu-Zong ore district illustrate that this scheme can effectively suppress the power frequency noise. This paper has been withdrawn from the Technical Program and will not be presented at the 87th SEG Annual Meeting. Keywords: frequency-domain, signal processing, fourier, magnetotelluricPermalink: https://doi.org/10.1190/segam2017-17638915.1FiguresReferencesRelatedDetails SEG Technical Program Expanded Abstracts 2017ISSN (print):1052-3812 ISSN (online):1949-4645Copyright: 2017 Pages: 6093 publication data© 2017 Published in electronic format with permission by the Society of Exploration GeophysicistsPublisher:Society of Exploration Geophysicists HistoryPublished Online: 17 Aug 2017 CITATION INFORMATION Guang Li, Jingtian Tang, Cong Zhou, Xiao Xiao, and Shangshi Wen, (2017), "Power frequency noise suppression of magnetotelluric data based on frequency-domain sparse decomposition," SEG Technical Program Expanded Abstracts : 1246-1251. https://doi.org/10.1190/segam2017-17638915.1 Plain-Language Summary Keywordsfrequency-domainsignal processingfouriermagnetotelluricPDF DownloadLoading ...

Frequency-domain Blind Deconvolution Based on Negentropy Using Standard Symmetric Power Distribution

An improvement of blind deconvolution based on MIR to the nonstationary seismic data

Blind Deconvolution of Seismic Data Using F-Divergences

Spectral decomposition using deconvolutive short time Fourier transform spectrogram

Blind Restoration of Poststack 3D Seismic Data

A Frequency-Domain Seismic Blind Deconvolution Based on Gini Correlations

Combining Synchrosqueezed Wavelet Transforms and Empirical Mode Decomposition for Filtering Seismic Random Noise

Negative entropy-based Bussgang seismic blind de-convolution

Bayesian sparse inversion based on nonstationary deconvolution using hyperbolic smoothing methods

Deghosting for Variable Depth Streamer Based on Super-Gaussianity

Random Noise Suppression Using Normalized Convolution Filter

Blind deconvolution of seismic data based on Kagan divergences

Deblending by Iterative Orthogonalization and Seislet Thresholding

A negative entropy-based Bussgang seismic blind de-convolotion

Seismic Blind Deconvolution Method Based on Generalized Gaussian Distribution

Spectral decomposition based on multiple windows reassigned spectrum

Seismic Data Denoising Based on Sparse and Redundant Representation

Nonstationary Sparse-Reflectivity Inversion Using Nonconvex Constraint in Frequency Domain

Power Frequency Noise Suppression of Magnetotelluric Data Based on Frequency-Domain Sparse Decomposition

A Blind Seismic Deconvolution Method Based on Particle Swarm Optimization

Sparse Multichannel Blind Deconvolution of Seismic Data via Spectral Projected-Gradient