Apodization: Apodization of the FID
In ManonMartin/PEPSNMR: Pre-process 1H-NMR FID signals

Description Usage Arguments Details Value Author(s) References Examples

The function multiplies the FID by a defined factor to increase the sensibility and/or resolution of the spectra.

Apodization(Fid_data, Fid_info = NULL, DT = NULL, type.apod = c("exp","cos2", 
            "blockexp","blockcos2","gauss","hanning","hamming"), phase = 0,
            rectRatio = 1/2, gaussLB = 1, expLB = 0.3, plotWindow = FALSE, 
            returnFactor = FALSE, verbose=FALSE)

`Fid_data`	Matrix containing the FIDs, one row per signal, as outputted by `ReadFids`.
`Fid_info`	Matrix containing the info about the FIDs, one row per signal, as outputted by `ReadFids`.
`DT`	If given, used instead of `Fid_info` to give the Dwell Time, the time between 2 points of the FID.
`type.apod`	Type of apodization, see details.
`phase`	Phase at which the apodization window is maximum for `cos2`, `hanning` and `hamming` types. For example, if phase is 0.2, the maximum is at 20% of the signal.
`rectRatio`	If there is a rectangular window, ratio between the width of the window and the width of the signal.
`gaussLB`	Line Broadening for the gaussian window, see details.
`expLB`	Line Broadening for the exponential window, see details.
`plotWindow`	If `TRUE`, a plot of the signal applied to the FID is displayed.
`returnFactor`	If `TRUE`, returns a list with the final FIDs and the apodization function.
`verbose`	If`"TRUE"`, will print processing information.

The apodization is usually performed in order to increase the sensitivity, i.e. the Signal-to-Noise Ratio (SNR) of the spectra. This is based on the fact that the signal intensity is decreasing over time unlike the noise that keeps a constant amplitude, leaving a noisy tail at the end of the FID. Multiplying the FID with a decaying signal will then increase the SNR. Since the area under the spectral peak remains unchanged, a faster decay will also result in a reduced peak height in spectra, lowering the spectral resolution. Optimal trade-off parameters for the apodization signal are thus needed to prevent high losses in sensitivity/resolution.

A FID of the form s0 exp(i*2*pi*nu*t) exp(-t/T) has a peak in its spectrum at the frequency nu of width that is inversely proportional to T. This peak is called a spectral line and its width a spectral width.

In the case of the exponential multiplication ("exp"), which is the default apodization, the decaying exponential becomes:

exp(-t(1/T + LB))

The new decay T* which satisfies 1/T* = 1/T + LB is therefore smaller so the spectral line is broader. That is why we call this parameter the Line Broadening.

If LB increases, the SNR increases but at the expense of the spectral resolution. Usual values in proton NMR for “LB” found in the literature are 0.3 for the NOESY presat pulse sequence and -0.01 for the CMPG presat pulse sequence. It should not exceed the value of 1 to avoid information loss.

The different types of apodization are:

exp: The signal is multiplied by a decreasing exponential exp(-t/expLB).
cos2: The signal is multiplied by the value of a cosinus squared from 0 (where its value is 1) until pi/2 (where its value is 0).
blockexp: The first part of the signal (defined by rectRatio) is left unchanged and the second is multiplied by exp(-t/expLB) starting at value 1.
blockcos2: the first part is left unchanged as with blockexp and the second part is multiplied by a cosinus squared where its value starts at 1 at the end of the block and ends at 0 at the end of the signal.
gauss: The signal is multiplied by a gaussian window centered at the beginning of the FID and with sigma=1/gaussLB.
hanning: The signal is multiplied by a hanning window : 0.5 + 0.5 cos.
hamming: The signal is multiplied by a hamming window : 0.54 + 0.46 cos.

If returnFactor is TRUE, will return a list with the following elements: Fid_data and Factor. Otherwise, the function will just return Fid_data.

`Fid_data`	The apodized FIDs.
`Factor`	The apodization signal.

Benoît Legat & Manon Martin

Inspired from the matNMR library.

Martin, M., Legat, B., Leenders, J., Vanwinsberghe, J., Rousseau, R., Boulanger, B., & Govaerts, B. (2018). PepsNMR for 1H NMR metabolomic data pre-processing. Analytica chimica acta, 1019, 1-13.

Rousseau, R. (2011). Statistical contribution to the analysis of metabonomics data in 1H NMR spectroscopy (Doctoral dissertation, PhD thesis. Institut de statistique, biostatistique et sciences actuarielles, Université catholique de Louvain, Belgium).

require(PepsNMRData)
Apod_res <- Apodization(Data_HS_sp$FidData_HS_2, 
                          FidInfo_HS, plotWindow=FALSE)

#or
Apod_res <- Apodization(Data_HS_sp$FidData_HS_2, 
                          FidInfo_HS, plotWindow=FALSE, returnFactor=TRUE)
Apod_fid = Apod_res[["Fid_data"]]
plot(Apod_res[["Factor"]], type="l")