Long Short-Term Memory Recurrent Network Architectures for Electromagnetic Field Reconstruction Based on Underground Observations

• Published in: Atmosphere Vol. 15; no. 6; p. 734
• Main Authors: Tian, Yixing, Xie, Chengliang, Wang, Yun
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.06.2024
• Subjects: Algorithms; Ambient noise; Background noise; Cosmic rays; Data processing; Datasets; Deep learning; Dictionaries; Earth science; electromagnetic field; Electromagnetic fields; Fourier transforms; Geophysical data; Geophysics; Laboratories; Long short-term memory; LSTM; Methods; Morphology; neural network; Neural networks; Noise levels; Noise reduction; noise suppression; Observational learning; R&D; Reconstruction; Recurrent neural networks; Research & development; Signal processing; Signal to noise ratio; Theoretical analysis; Time series; underground laboratory; Wavelet transforms; France

Deep underground laboratories offer advantages for conducting high-precision observations of weak geophysical signals, benefiting from a low background noise level. Enhancing strong, noisy ground electromagnetic (EM) field data using synchronously recorded underground EM signals, which typically exh...