High Spatial Resolution Optical Frequency-Domain Reflectometry Using Wiener Deconvolution

Wiener deconvolution offers the advantages of effectively suppressing noise, restoring image details, and enhancing signal processing accuracy. It has been successfully applied in image denoising for distributed fiber-optic sensing. However, the effective application of Wiener deconvolution in optical frequency-domain reflectometry (OFDR) systems has not yet been reported. The challenge lies in obtaining an effective point spread function (PSF) or blur function to minimize spatial resolution degradation during image processing. To the best of our knowledge, we first proposed a high spatial resolution OFDR system based on Wiener deconvolution. Crucially, we have theoretically and experimentally validated that the auto-correlation function of local Rayleigh scattering spectrum serves as the PSF for the OFDR system. We have optimized the selection range of the PSF in both spatial and frequency domains to achieve the highest spatial resolution. The effectiveness of this algorithm and its superior denoising performance were validated by comparison with Gaussian filtering. Finally, through multiple experiments, we have demonstrated that the proposed Wiener deconvolution method can restore events at a spatial resolution of 200 $\mu \text{m}$.