Illustration of RTFs on a test case
Time-frequency representations in signal processing

Each time-frequency representation has its own particular characteristics for the same signal. To illustrate and compare them, the main RTFs are applied to a synthetic signal made up of several elementary signals whose characteristics are chosen to highlight the qualities or shortcomings of the representations tested.

The 256-point discrete test signal is the sum of white Gaussian noise and the following six signals: a linear frequency modulation, a hyperbolic frequency modulation, a low-frequency pulse signal, a sinusoidal signal and two high-frequency pulse signals.

The various time-frequency representations have been constructed in discrete form using a "time-frequency toolbox"