Is It Redundant to Use Model-Based Subtraction Together With the Reference Noise Cancellation?

Abstract

Summary Surface NMR is a competitive method in hydro-geophysical investigation due to its directive sensitivity to the water content, permeability and hydraulic conductivity of the aquifer. One of the main obstacles to expand its application in many area of interest is the low signal-to-noise ratio (SNR) of measurement. To improve the SNR by mitigating noise, numerous approaches has been proposed, for example remote reference noise cancellation (RNC) and model-based powerline harmonics subtraction (MB). In this abstract, the question whether MB should be applied before RNC is investigated. First, the theoretical length of Wiener filter (WF) in RNC is analysed by observing harmonics relationship between primary and reference channels. Then the benefits of applying MB before RNC is given within multiple noise sources are present. Last, the noise level of field measurements processed by RNC and MB followed by RNC are compared.