An approach for selecting a real-life signal best-performing time-frequency distribution

The difficulty of selecting the optimal time-frequency distribution (TFD) for a given real-life signal has been one of the major limiting factors to a wider use of time-frequency analysis tools in practice. This paper, by extending our earlier works on the objective assessment of the performance of TFDs, presents a novel automatic approach for selecting a real-life signal best-performing time-frequency distribution from a set of considered TFDs. The practical applicability of the approach is illustrated on real-life signals examples.