English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This article presents a weakly supervised machine learning method, which wecall DAS-N2N, for suppressing strong random noise in distributed acousticsensing (DAS) recordings. DAS-N2N requires no manually produced labels (i.e.,pre-determined examples of clean event signals or sections of noise) fortraining and aims to map random noise processes to a chosen summary statistic,such as the distribution mean, median or mode, whilst retaining the trueunderlying signal. This is achieved by splicing (joining together) two fibreshosted within a single optical cable, recording two noisy copies of the sameunderlying signal corrupted by different independent realizations of randomobservational noise. A deep learning model can then be trained using only thesetwo noisy copies of the data to produce a near fully-denoised copy. Once themodel is trained, only noisy data from a single fibre is required. Using adataset from a DAS array deployed on the surface of the Rutford Ice Stream inAntarctica, we demonstrate that DAS-N2N greatly suppresses incoherent noise andenhances the signal-to-noise ratios (SNR) of natural microseismic icequakeevents. We further show that this approach is inherently more efficient andeffective than standard stop/pass band and white noise (e.g., Wiener) filteringroutines, as well as a comparable self-supervised learning method based onmasking individual DAS channels. Our preferred model for this task islightweight, processing 30 seconds of data recorded at a sampling frequency of1000 Hz over 985 channels (approx. 1 km of fiber) in $<$1 s. Due to the highnoise levels in DAS recordings, efficient data-driven denoising methods, suchas DAS-N2N, will prove essential to time-critical DAS earthquake detection,particularly in the case of microseismic monitoring.

DAS-N2N: Machine learning Distributed Acoustic Sensing (DAS) signal denoising without clean data