Spatial Deep Deconvolution U-Net for Traffic Analyses with Distributed Acoustic Sensing

Distributed Acoustic Sensing (DAS) that transforms city-wide fiber-optic cables into a large-scale strain sensing array has shown the potential to revolutionize urban traffic monitoring by providing a fine-grained, scalable, and low-maintenance monitoring solution. However, there are challenges that limit DAS's real-world usage: noise contamination and interference among closely traveling cars. To address the issues, we introduce a self-supervised U-Net model that can suppress background noise and compress car-induced DAS signals into high-resolution pulses through spatial deconvolution. To guide the design of the approach, we investigate the fiber response to vehicles through numerical simulation and field experiments. We show that the localized and narrow outputs from our model lead to accurate and highly resolved car position and speed tracking. We evaluate the effectiveness and robustness of our method through field recordings under different traffic conditions and various driving speeds. Our results show that our method can enhance the spatial-temporal resolution and better resolve closely traveling cars. The spatial deconvolution U-Net model also enables the characterization of large-size vehicles to identify axle numbers and estimate the vehicle length. Monitoring large-size vehicles also benefits imaging deep earth by leveraging the surface waves induced by the dynamic vehicle-road interaction.

Improved depth imaging by constrained full-waveform inversion

We propose a formulation of full-wavefield inversion (FWI) as a constrained optimization problem, and describe a computationally efficient technique for solving constrained full-wavefield inversion (CFWI). The technique is based on using a total-variation regularization method, with the regularization weighted in favor of constraining deeper subsurface model sections. The method helps to promote "edge-preserving" blocky model inversion where fitting the seismic data alone fails to adequately constrain the model. The method is demonstrated on synthetic datasets with added noise, and is shown to enhance the sharpness of the inverted model and correctly reposition mispositioned reflectors by better constraining the velocity model at depth.