Unsupervised CP-UNet Framework for Denoising DAS Data with Decay Noise

2025-02-19

AI Summary

Distributed acoustic sensing (DAS) data are severely corrupted by random and impulsive noise, while existing supervised denoising methods rely heavily on high-quality ground-truth labels—a major bottleneck in practice. To address this, we propose CP-UNet, an unsupervised deep learning framework. It innovatively integrates a Context Pyramid module with a Connection module, replaces Batch Normalization with Layer Normalization to enhance training stability and convergence speed, and employs Huber loss to formulate an end-to-end unsupervised learning paradigm. Evaluated on both synthetic and field-recorded 2D DAS datasets, CP-UNet consistently outperforms conventional methods and state-of-the-art unsupervised models. It effectively suppresses decay-type noise while preserving signal fidelity and reconstructing structural details. The framework offers a lightweight, label-free, and deployable solution for low-SNR DAS applications.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Computer Vision: Learning & Optimization for CV

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Distributed acoustic sensor (DAS) technology leverages optical fiber cables to detect acoustic signals, providing cost-effective and dense monitoring capabilities. It offers several advantages including resistance to extreme conditions, immunity to electromagnetic interference, and accurate detection. However, DAS typically exhibits a lower signal-to-noise ratio (S/N) compared to geophones and is susceptible to various noise types, such as random noise, erratic noise, level noise, and long-period noise. This reduced S/N can negatively impact data analyses containing inversion and interpretation. While artificial intelligence has demonstrated excellent denoising capabilities, most existing methods rely on supervised learning with labeled data, which imposes stringent requirements on the quality of the labels. To address this issue, we develop a label-free unsupervised learning (UL) network model based on Context-Pyramid-UNet (CP-UNet) to suppress erratic and random noises in DAS data. The CP-UNet utilizes the Context Pyramid Module in the encoding and decoding process to extract features and reconstruct the DAS data. To enhance the connectivity between shallow and deep features, we add a Connected Module (CM) to both encoding and decoding section. Layer Normalization (LN) is utilized to replace the commonly employed Batch Normalization (BN), accelerating the convergence of the model and preventing gradient explosion during training. Huber-loss is adopted as our loss function whose parameters are experimentally determined. We apply the network to both the 2-D synthetic and filed data. Comparing to traditional denoising methods and the latest UL framework, our proposed method demonstrates superior noise reduction performance.

Problem

• Unsupervised denoising of DAS data
• Suppressing erratic and random noises
• Enhancing signal-to-noise ratio

Innovation

• Unsupervised CP-UNet for denoising
• Context Pyramid Module integration
• Huber-loss for model optimization