Running discrete Fourier transform for time--frequency analysis of biomedical signals.

The present work introduces a running discrete Fourier transform (RDFT) for time-frequency analysis (TFA) of nonstationary short-term signals. The RDFT algorithm adapts a window function, which has adjustable time-frequency characteristics. A generalized weighting filter was proposed, which maps the time-dependent signal into the time-frequency plane. As an example the weighting filters were deduced for exponential, gamma and binomial window sequences. Application of the RDFT algorithm to the analysis of biomedical signals is discussed.

[1]  Bernard Widrow,et al.  Adaptive Signal Processing , 1985 .

[2]  H. Olkkonen,et al.  Running discrete cosine transform. , 1992, Journal of biomedical engineering.

[3]  Alan V. Oppenheim,et al.  Discrete-Time Signal Pro-cessing , 1989 .

[4]  J. Shynk.  Frequency-domain and multirate adaptive filtering , 1992, IEEE Signal Processing Magazine.

[5]  Hannu Olkkonen.  Computation of running discrete Hartley transform coefficients , 1991, Signal Process..

[6]  F. Hlawatsch,et al.  Linear and quadratic time-frequency signal representations , 1992, IEEE Signal Processing Magazine.