SUMMARY
Adaptive noise cancelling of multichannel magnetic resonance sounding (MRS) signals is investigated. An analysis of the noise sources affecting MRS signals show that the applicability of adaptive noise cancelling is primarily limited to cancel powerline harmonics. The problems of handling spikes in MRS signals are discussed and an efficient algorithm for spike detection is presented. The optimum parameters for multichannel adaptive noise cancelling are identified through simulations with synthetic signals added to noise‐only recordings from an MRS instrument. We discuss the design and the efficiency of different stacking methods. The results from multichannel adaptive noise cancelling are compared to time‐domain multichannel Wiener filtering. Our results show that within the experimental uncertainty the two methods give identical results.
The fidelity of magnetic resonance sounding signals is often severely degraded by noise, primarily electrical interference from powerline harmonics and short electromagnetic discharges. In many circumstances, the noise originates from multiple sources. We show that noise cancelling can be improved if the multiple origins of noise are taken into account. In particular, a method is developed where powerline harmonics are efficiently removed through a modelbased approach. Subsequently, standard multichannel Wiener filtering can be used to provide a further noise reduction. The performance of the method depends on the distribution of noise on the particular site of measurement. Simulations on synthetic signals embedded in real noise recordings show that the combined approach can improve the signal-to-noise ratio with an accompanying improvement in retrieval of model parameters.
We present a comprehensive study of the parameter determination of magnetic resonance sounding (MRS) models in a joint MRS and transient electromagnetic (TEM) data analysis scheme. The parameter determination is assessed by calculating the model parameter uncertainties based on an a posteriori model covariance matrix. An entire MRS data set, dependent on pulse moment and time gate values, together with TEM data, is used for all analyses and realistic noise levels are assigned to the data.
Sensitivity analyses are studied for the determination of water content as a key parameter estimated during inversion of MRS data. We show the results for different suites of (three‐layer) models, in which we investigate the effect of resistivity, water content, relaxation time, loop side length, number of pulse moments and measurement dead time on the determination of water content in a water‐bearing layer. For all suites of models the effect of a top conductive and a top resistive layer are compared. Moreover, we analyse all models for a long (40 ms) and short (10 ms) measurement dead time. The effect of noise level on the parameter determination is also analysed.
We conclude that, in general, the resistivity of the water‐bearing layer (layer of interest, LOI) does not affect the determination of water content in the LOI but the resistivity of the top layer increases depth resolution; the water content of the LOI does not influence its determination considerably in cases where the signal has a relatively long relaxation time in the LOI; determination of the water content in the LOI is improved by increasing the relaxation time of the signal in the LOI; short measurement dead time will improve the parameter determination for signals with a relatively short relaxation time; increasing loop side length and the number of pulse moments do not necessarily improve the parameter determination.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.