Low complexity time-frequency analysis methods for efficient implementation

Since the time-frequency distribution can analyze how the instantaneous frequency of a signal varies with time, it is very effective for signal processing. However, due to that the time-frequency distribution has a two-dimensional output, it has a higher complexity and requires more computation time. In this paper, we find that if the window of the short-time Fourier transforms (STFTs) has some specific form, then they can be implemented in very efficient ways. Their performances are similar to that of the conventional STFT but the complexity is reduced from O(TNlog₂N) to O(TF) and have much lower cost for hardware implementation.