Extracting surface wave dispersion curves based on frequency Bessel transform

With the development and renovation of urban infrastructure, geological hazards in cities have become increasingly prominent. Urban background noise imaging has gained widespread attention as a key technology for ensuring urban safety and sustainable development. Distributed Acoustic Sensing (DAS) technology combined with background noise imaging offers a novel approach for high-resolution near-surface structural detection in urban areas. However, the extraction of high-resolution multimode surface wave dispersion curves faces significant challenges due to poor coupling conditions between urban optical cables and geological strata, as well as complex non-ideal scattering field environments. This study conducted field tests of a DAS array using buried underground optical cables at Jilin University's Chaoyang Campus, extracting dispersion curves from 12 hours of background noise data. After preprocessing the acquired data, cross-correlation calculations were performed on the signal segments from virtual source and receiver stations. The noise cross-correlation function (NCF) was obtained using the phase-weighted summation method (PWS), and the multimode dispersion curve was extracted via the frequency-Bessel transform (F-J). The NCF derived through PWS yielded clearer and purer V-shaped surface wave records. Results demonstrate that even under complex background noise fields and weak fiber-soil coupling, the proposed method yields clear, unbiased, and pure NCF. This ensures the reliable extraction of high-resolution multimodal surface wave dispersion curves, advancing the development of urban background noise imaging based on DAS technology.