Title :
Enhanced denoising scheme for NMR signals
Author_Institution :
Department of Electrical and Computer Engineering, Queen´´s University, Canada
Abstract :
A new scheme for denoising Nuclear Magnetic Resonance (NMR) signals using time-frequency (TF) transforms is presented. This scheme utilizes some inherent characteristics of NMR signals to discriminate them from the noise. More specifically, the envelop of the NMR signal decays exponentially in the TF domain along the time direction while the noise is not. This decay along with the difference in energy level was utilized to differentiate between NMR signals and noise in the TF domain. The evaluation of this scheme was performed using extensive simulation of NMR signals with different noise levels. It was observed that the proposed scheme gives superior results for all noise levels.
Keywords :
Manganese; Noise; Nuclear magnetic resonance; Transforms;
Conference_Titel :
Acoustics, Speech, and Signal Processing (ICASSP), 2002 IEEE International Conference on
Conference_Location :
Orlando, FL, USA
Print_ISBN :
0-7803-7402-9
DOI :
10.1109/ICASSP.2002.5745503