An architecture for the vlsi design of systems for time-frequency analysis and time-varying filtering

A flexible system for time-frequency signal analysis is presented. It is based on the S-method, which has a significant advantage in implementation since it can involve, as a key intermediate step, the Short-time Fourier transform or the Hartley transform, each widely studied and commonly used in practice. Signal invariant and signal dependent system forms are presented. Hardware design, for a fixed-point arithmetic, is well-structured and suitable for vlsi implementation. The same hardware, without additional time requirements, may be shared for the simultaneous realization of the fourth order L-Wigner distribution, as well as for the realization of the cross-terms free fourth order polynomial Wigner-Ville distribution. This possibility makes the designed hardware suitable for wide range of the applications. The proposed hardware is applied to the realization of time-varying filtering, as well. Finally, it has been implemented with fpga chips (Field Programmable Gate Array) in order to verify the results on real devices.RésuméCet article présente un système souple pour l’analyse d’un signal en temps et en fréquence. Ce système est fondé sur la méthode S, ce qui facilite la réalisation grâce àl’emploi de la transformation de Fourier àcourt terme et de la transformation de Hartley, bien connues et largement en usage. L’article considère deux variantes suivant que la fenêtre dépend ou non du signal. La conception du matériel, en arithmétique àvirgule fixe, convient bien àune réalisation par vlsi. Le même matériel peut être utilisé simultanément et sans délai supplémentaire pour la réalisation d’une distribution L-Wigner du quatrième ordre et celle d’une distribution Wigner-Ville polynomiale d’ordre quatre. Le matériel proposé est appliqué àla réalisation de filtrages variables dans le temps. Les résultats ont pu être vérifiés grâce àune réalisation àbase de puces fpga.

[1]  Boualem Boashash,et al.  An efficient real-time implementation of the Wigner-Ville distribution , 1987, IEEE Trans. Acoust. Speech Signal Process..

[2]  Swapna Banerjee,et al.  A VLSI array architecture for realization of DFT, DHT, DCT and DST , 2001, Signal Process..

[3]  Zoran Petrović,et al.  Unified approach to noise analysis in the Wigner distribution and spectrogram , 1996 .

[4]  Stephen A. Dyer,et al.  Digital signal processing , 2018, 8th International Multitopic Conference, 2004. Proceedings of INMIC 2004..

[5]  M. Amin,et al.  Short-time Fourier transforms using cascade filter structures , 1995 .

[6]  R. Bracewell The fast Hartley transform , 1984, Proceedings of the IEEE.

[7]  Ljubisa Stankovic,et al.  A method for improved distribution concentration in the time-frequency analysis of multicomponent signals using the L-Wigner distribution , 1995, IEEE Trans. Signal Process..

[8]  Boualem Boashash,et al.  Polynomial Wigner-Ville distributions and their relationship to time-varying higher order spectra , 1994, IEEE Trans. Signal Process..

[9]  W. Kozek,et al.  Time-frequency signal processing based on the Wigner-Weyl framework , 1992, Signal Process..

[10]  Michael Unser,et al.  RECURSION IN SHORT-TIME SIGNAL ANALYSIS , 1983 .

[11]  K. J. Ray Liu,et al.  Unified parallel lattice structures for time-recursive discrete cosine/sine/Hartley transforms , 1993, IEEE Trans. Signal Process..

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

[13]  Kui Liu,et al.  Novel parallel architectures for short-time Fourier transform , 1993 .

[14]  Boualem Boashash,et al.  Polynomial time-frequency distributions and time-varying higher order spectra: Application to the analysis of multicomponent FM signals and to the treatment of multiplicative noise , 1998, Signal Process..

[15]  Srdjan Stankovic,et al.  An architecture for realization of the cross-terms free polynomial Wigner-Ville distribution , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[16]  K. J. Ray Liu Novel Parallel Architectures for Short-Time Fourier Transform , 2004 .

[17]  LJubisa Stankovic On the time-frequency analysis based filtering , 2000, Ann. des Télécommunications.

[18]  Thomas W. Parks,et al.  Time-varying filtering and signal estimation using Wigner distribution synthesis techniques , 1986, IEEE Trans. Acoust. Speech Signal Process..

[19]  A. Papoulis Signal Analysis , 1977 .

[20]  Srdjan Stankovic,et al.  An architecture for the realization of a system for time-frequency signal analysis , 1997 .

[21]  Louis L. Scharf,et al.  Toeplitz and Hankel kernels for estimating time-varying spectra of discrete-time random processes , 2001, IEEE Trans. Signal Process..

[22]  M. Ackroyd Short‐Time Spectra and Time‐Frequency Energy Distributions , 1971 .

[23]  L. Cohen,et al.  Time-frequency distributions-a review , 1989, Proc. IEEE.

[24]  Gerald Matz,et al.  Generalized evolutionary spectral analysis and the Weyl spectrum of nonstationary random processes , 1997, IEEE Trans. Signal Process..

[25]  Edmond A. Jonckheere,et al.  Split-radix fast Hartley transform in one and two dimensions , 1991, IEEE Trans. Signal Process..

[26]  G. Boudreaux-Bartels,et al.  A comparative study of alias-free time-frequency representations , 1994, Proceedings of IEEE-SP International Symposium on Time- Frequency and Time-Scale Analysis.

[27]  Thomas W. Parks,et al.  The Weyl correspondence and time-frequency analysis , 1994, IEEE Trans. Signal Process..

[28]  Moeness G. Amin Spectral smoothing and recursion based on the nonstationarity of the autocorrelation function , 1991, IEEE Trans. Signal Process..

[29]  Boualem Boashash,et al.  Relationship between the polynomial and the higher order Wigner-Ville distribution , 1995, IEEE Signal Processing Letters.

[30]  Moeness G. Amin,et al.  An application of the LMS algorithm in smoothing pseudo-Wigner distribution , 1993, IEEE Trans. Signal Process..

[31]  Ljubisa Stankovic,et al.  A method for time-frequency analysis , 1994, IEEE Trans. Signal Process..

[32]  Srdjan Stankovic,et al.  About time-variant filtering of speech signals with time-frequency distributions for hands-free telephone systems , 2000, Signal Process..

[33]  Srdjan Stankovic,et al.  Special purpose hardware for time frequency analysis , 1997 .

[34]  Richard G. Baraniuk,et al.  Pseudo affine Wigner distributions: definition and kernel formulation , 1998, IEEE Trans. Signal Process..

[35]  M.G. Amin A new approach to recursive Fourier transform , 1987, Proceedings of the IEEE.

[36]  LJubisa Stankovic,et al.  Finite word-length effects in implementation of distributions for time-frequency signal analysis , 1998, IEEE Trans. Signal Process..