GPR data time varying deconvolution by kurtosis maximization

Stochastic and deterministic deconvolution methods encounter difficulties in increasing the temporal resolution of GPR data. Statistical approaches, such as predictive or spiking deconvolution are not effective when the wavelet is non-minimum phase, which is the case for GPR data. Wavelet deconvolution is not successful due to the non-stationarity of the GPR trace. Here, prior deconvolution, we apply a spectral balancing method in t-f domain which efficiently reduces the non-stationarity. The proposed methodology involves correction for phase residuals using the maximum kurtosis method. The effectiveness of this methodology is demonstrated on synthetic and real GPR data.