MUSIC-Based Super-Resolution CMP Velocity-Depth Analysis for Multilayer Cases

Abstract

This letter proposes a multiple signal classification (MUSIC)-based algorithm to generate a super-resolution velocity-depth spectrum for precise subsurface multilayer analysis, which cannot be achieved by conventional Common MidPoint (CMP) velocity-depth analysis. A self-adaptive peak detection process and a MUSIC algorithm with a modified steering matrix are the key components of the proposed method. Both numerical simulation and experimentation are used to verify the feasibility of this study. Regarding both accuracy and resolution, the results show that the suggested method outperforms the conventional SAR-based method, as the layer information on the spectra is correctly located and strongly focused. Moreover, the proposed approach can distinguish between the layers that cause weak reflections and those that cause strong reflections. Compared to previous studies, the proposed method enables self-adaptive super-resolution imaging in ground penetrating radar (GPR) CMP subsurface layer analysis, and it has a great potential for usage in other GPR subsurface applications.