Spectral estimation is a branch of the signal processing, performed on data and based on frequency-domain techniques. Within the LTPDA toolbox many functions of the Matlab Signal Processing Toolbox (which is required) were rewritten to operate on LTPDA Analysis Objects. Univariate and multivariate technique are available, so to estimate for example the linear power spectral density or the cross spectral density of one or more signals. The focus of the tools is on time-series objects, whose spectral content needs to be estimated.
The power spectrum density estimators are based on pwelch from MATLAB, which is an implementation of Welch's averaged, modified periodogram method.
The following pages describe the different Welch-based spectral estimation tools available in the LTPDA toolbox:
As an alternative, the LTPDA toolbox makes available the same set of estimators, based on an implementation of the LPSD algorithm (See "Improved spectrum estimation from digitized time series on a logarithmic frequency axis", M Troebs, G Heinzel, Measurement 39 (2006) 120-129).
The following pages describe the different LPSD-based spectral estimation tools available in the LTPDA toolbox:
More detailed help on spectral estimation can be found in the help associated with the Signal Processing Toolbox (>> doc signal)