DOE OSTI · 3030373
Model Card for WaveDP
Abstract
WaveDP was trained using both earthquake signals and pure noise waveforms from the Stanford Earthquake Dataset (STEAD). The model achieved a median signal-to-noise ratio (SNR) improvement of 7 dB. Because the model uses seismograms as direct input, the model is four times faster than an existing model for signal denoising that uses spectral representations of seismograms as input.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Chai, Chengping [Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)] (ORCID:0000000267926014), Rose, Derek [Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)] (ORCID:0000000252174374), Stewart, Scott [Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)] (ORCID:0000000343205818), Martindale, Nathan [Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)] (ORCID:0000000250365433), Adams, Mark [Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)] (ORCID:0000000254142800), Torres Polanco, Anibely [Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)] (ORCID:0000000152182174), Linville, Lisa [Sandia National Laboratories (SNL), Albuquerque, NM, and Livermore, CA (United States)], Stanley, Christopher [Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)] (ORCID:0000000242267710), Bingham, Philip [Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)] (ORCID:0000000346166084). 2025-12-01. Model Card for WaveDP. https://doi.org/10.2172/3030373
Cite the original work for its findings. Save a collection to share your selection of sources.