Enhancement of the frequency resolution of the S-transform using the fourier transform

The S-transform is a method of time-frequency analysis of a time series, essentially the Fourier spectrum as a function of time. While attractive in concept, the S-transform suffers from inherently poor frequency resolution, particularly at the high frequencies. We present here a technique (named STF) to enhance the frequency resolution through the use of the global Fourier spectrum. The scaling of the localising Gaussian window in STF is made a function of the global Fourier amplitude spectrum. Tests with several data series show better frequency resolution, with no increase in computation time or resource usage.