Maximum Entropy Reconstruction
Maximum entropy (MaxEnt) reconstruction is a technique for computing frequency spectra from time series data, e.g., a free induction decay. It is based on the principle of information entropy introduced by Claude Shannon (1948). The MaxEnt spectrum contains the smallest amount of information consistent with the experimental data. In contrast to the discrete Fourier transform (DFT), MaxEnt references the measured data indirectly, enabling it to handle nonuniformly sampled (NUS, also called sparse sampling) data. Thus an important application of MaxEnt is the computation of multidimensional NMR spectra from NUS data.
- Daniell GJ, Hore PJ (1989) Maximum entropy and NMR – a new approach. J Magn Reson 84:515–536Google Scholar
- Hoch JC, Stern AS (1996) NMR data processing. Wiley-Liss, New YorkGoogle Scholar
- Hoch JC, Stern AS, Donoho DL, Johnstone IM (1990) Maximum entropy reconstruction of complex (phase-sensitive) spectra. J Magn Reson 86:236–246Google Scholar