On the estimation of the evolutionary power spectral density

Two popular spectral-based approaches for estimating the evolutionary power spectral density (EPSD) function from samples of process are based on short-time Fourier transform (STFT) and continuous wavelet transform. Both rely concept slowly varying modulation or EPSD function, although quantification effect 'slow' variation in estimated is elusive. We propose, present study, to use derivatives quantify smoothness context function. derive equations by using S-transform These as simple that derived STFT. also corresponding assessing residual these transforms, including The provides an approach identifying quantifying, its estimation, numerical results indicate depends both with respect time frequency well adopted