Analysis of borehole radar logging signal using short-time Fourier transform

As a useful subsurface detection and imaging tool, borehole radar (BHR) has been gradually expanded to petroleum logging field recently. However, due to the serious attenuation of the conductive borehole filled with mud, BHR echoes in variation formations will be different from those collected in a resistive borehole, and bring severe challenges for directly analyzing BHR signals in time domain. Although removing direct wave can enhance the signal of the region of interest, the damage to BHR echo signal is inevitable. With the purpose of studying the propagation characteristic of BHR signals, a horizontal multiple-layered formation hosting mud-filled borehole model is constructed, and two types of data sets are obtained from the model with fixed conductivity or relative permittivity. The classical short time Fourier transform (STFT) is adopted to analyze the original BHR responses to different layers in time-frequency domain. It indicates that STFT is an effective and convenient algorithm to analyze BHR logging data, which can also provide a new idea for reservoir evaluation.