Presenting a New Method for Time-Frequency Representation Based on the S Transform

Document Type : Original Article

Authors

Abstract

S transform is a time-frequency transform created based on the concept of short-time Fourier
transform and wavelet transform. The main property of this transform that separates it from the
discrete short-time Fourier transform is the variable-frequency resolution. But in this transform,
the resolution change of high frequencies to low frequencies is very significant. So another
transform as a Modified S Transform(MST) was introduced to adjust a parameter that controls
the rate of change in resolution. In this paper, S and MST has been first discussed.Then, on the
basis of the MST transform,the Optimum S Transform (EST) was introduced with a new
parameter. Next, calculating the length of the used window, the dependence of it to the
proposed parameters was studied and the effect of each of the parameters on the rate of
changing the length of window was illustrated in theory and simulation. Finally, with
simulation, the performance of the proposed method for detection of FMCW signals was
examined , and the effect of each of the parameters on the detection of above signals was shown.
Then the better performance of the proposed method than two last methods was proved.

Keywords


  1. B. Boashash; “Time Frequency signal Analysis and
  2. Processing”; Elevier, 2003.
  3. L. Stankovi´c; M. Dakovi´c; and T. Thayaparan; “Time-
  4. Frequency Signal Analysis with Applications,” Artech
  5. house, 2013.
  6. M. Misiti. Y. Misiti. G. Oppenheim. and J.-M. Poggi.
  7. “Wavelets and their Applications,” In proc.of the Int.
  8. conf.ISTE, 2007, 44-50.
  9. https://en.wikipedia.org/wiki/Window_function.
  10. M. Portnoff; “Time-frequency representation of digital
  11. signals and systems based on short-time Fourier analysis”
  12. IEEE Transactions on Acoustics, Speech, and Signal
  13. Processing ASSP-28 (1), pp. 55–69, 1980.
  14. R. G. Stockwell; L. Mansinha; and R. P. Lowe;
  15. “Localization of the complex spectrum: The S Transform,”
  16. IEEE Trans. Signal Processing, vol. 44, no. 4, pp. 998-1001,
  17. April 1996.
  18. R. G. Stockwell; “A basis for efficient representation of the
  19. S-transform,” Digital Signal Processing, Elsevier, 2006.
  20. V. Nithin ;S. George; “Transform: Time Frequency Analysis
  21. & Filtering,” Msc Thesis, National Institute of Technology,
  22. Rourkela, India, 2009.
  23. R. G. Stockwell; “Why use the S-Transform, AMS
  24. Pseudo–differential operators: partial differential equations
  25. and time–frequency analysis,” vol. 52, pp. 279–309, 2007.
  26. Z. Hao; H. Xu; G. Zheng; and G. Jing; “Study on the Timefrequency
  27. Characteristics of Engine Induction Noise in
  28. Acceleration Based on S Transform,” IEEE CISP, vol. 5, pp.
  29. -246, 2008.
  30. M. Schimmel ; J. Gallart; “The Inverse S-Transform in filters
  31. with Time Frequency Localization” , IEEE Trans. Signal
  32. Processing, vol. 55, no.11, pp. 4417-4422, 2005.
  33. I. Djurovic; E. Sejdic; and J. Jiang; “Frequency-based
  34. window width optimization for S-transform”, International
  35. Journal of Electronics and Communications, Elsevier, pp.
  36. -250, 2008.
  37. S. S. Sahu; G. Panda; and N. V. George; “An Improved
  38. S-Transform for Time-Frequency Analysis”, IEEE
  39. International Advance Computing
  40. Conf. IACC 2009, 20-29.
Volume 4, Issue 2 - Serial Number 2
November 2016
Pages 37-43
  • Receive Date: 15 March 2015
  • Revise Date: 22 January 2024
  • Accept Date: 19 September 2018
  • Publish Date: 22 July 2016