Abstract
Eliminating noise signals of the magnetotelluric (MT) method is bound to improve the quality of MT data. However, existing de-noising methods are designed for use in whole MT data sets, causing the loss of low-frequency information and severe mutation of the apparent resistivity-phase curve in low-frequency bands. In this paper, we used information entropy (IE), the Lempel–Ziv complexity (LZC), and matching pursuit (MP) to distinguish and suppress MT noise signals. Firstly, we extracted IE and LZC characteristic parameters from each segment of the MT signal in the time-series. Then, the characteristic parameters were input into the FCM clustering to automatically distinguish between the signal and noise. Next, the MP de-noising algorithm was used independently to eliminate MT signal segments that were identified as interference. Finally, the identified useful signal segments were combined with the denoised data segments to reconstruct the signal. The proposed method was validated through clustering analysis based on the signal samples collected at the Qinghai test site and the measured sites, where the results were compared to those obtained using the remote reference method and independent use of the MP method. The findings show that strong interference is purposefully removed, and the apparent resistivity-phase curve is continuous and stable. Moreover, the processed data can accurately reflect the geoelectrical information and improve the level of geological interpretation.
Highlights
The magnetotelluric (MT) method, an electromagnetic exploration method proposed in the early 1950s [1,2], measures the orthogonal electric-magnetic fields at the Earth’s surface to obtain the distribution of the underground geoelectric structure
The information entropy (IE)-Lempel–Ziv complexity (LZC) characteristics were developed to analyze the essential features of MT in combination with the fuzzy c-means (FCM) clustering algorithm for MT signal-noise identification or with the selective use of matching pursuit (MP) as the de-noising algorithm
The proposed method was processed for time-series and it was composed of three steps: characteristic extraction (IE and LZC), clustering analysis (FCM), and the de-noising algorithm (MP)
Summary
The magnetotelluric (MT) method, an electromagnetic exploration method proposed in the early 1950s [1,2], measures the orthogonal electric-magnetic fields at the Earth’s surface to obtain the distribution of the underground geoelectric structure. The emerging developments in MT signal-noise identification over recent years have presented a new processing mode for MT signal-noise separation [14,15] These methods distinguish whether the signal is contaminated by noise based on several characteristic parameters extracted from the signal. The IE-LZC characteristics were developed to analyze the essential features of MT in combination with the fuzzy c-means (FCM) clustering algorithm for MT signal-noise identification or with the selective use of matching pursuit (MP) as the de-noising algorithm. The denoised data from the proposed method closely resembled the original undisturbed data in terms of the essential characteristics, and the geoelectric structure information of the measured site was accurately reflected in the results
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
