Analyzing NMR spectra with the Morlet wavelet

We study the time-scale representation provided by the Morlet wavelet transform for characterizing NMR signals. From an analytical analysis and simulations, we conclude that the wavelet shows a satisfactory performance even when a baseline, an additive Gaussian noise or a solvent are present in the signals. It can also cope with non-Lorentzian lineshapes which commonly occur because of the inhomogeneous distribution of molecules in a substance. These results mean that the Morlet wavelet transform is a potential tool to quantify in vivo NMR signals.