Results on maximally flat fractional-delay systems

The two classes of maximally flat finite-impulse response (FIR) and all-pass infinite-impulse response (IIR) fractional-sample delay systems are thoroughly studied. New expressions for the transfer functions are derived and mathematical properties revealed. Our contributions to the FIR case include a closed-form formula for the Farrow structure, a three-term recurrence relation based on the interpolation algorithm of Neville, a concise operator-based formula using the forward shift operator, and a continued fraction representation. Three types of structures are developed based on these formulas. Our formula for the Farrow structure enhances the existing contributions by Valimaki, and by Vesma and Sarama/spl uml/ki on the subsystems of the structure. For the IIR case, it is rigorously proved, using the theory of Pade approximants, that the continued fraction formulation of Tassart and Depalle yields all-pass fractional delay systems. It is also proved that the maximally flat all-pass fractional-delay systems are closely related to the Lagrange interpolation. It is shown that these IIR systems can be characterized using Thiele's rational interpolation algorithm. A new formula for the transfer function is derived based on the Thiele continued fractions. Finally, a new class of maximally flat FIR fractional-sample delay systems that exhibit an almost all-pass magnitude response is proposed. The systems possess a maximally flat group-delay response at the end frequencies 0 and /spl pi/, and are characterized by a closed-form formula. Their main advantage over the classical FIR Lagrange interpolators is the improved magnitude response characteristics.

[1]  L. M. Milne-Thomson,et al.  The Calculus Of Finite Differences , 1934 .

[2]  J. Thiran.  Recursive digital filters with maximally flat group delay , 1971 .

[3]  Vesa V Alim Aki.  Discrete-Time Modeling of Acoustic Tubes Using Fractional Delay Filters , 1995 .

[4]  Che-Ho Wei,et al.  Programmable fractional sample delay filter with Lagrange interpolation , 1990 .

[5]  P. Wynn,et al.  Sequence Transformations and their Applications. , 1982 .

[6]  S. C. Dutta Roy,et al.  A note on the fir approximation of a fractional sample delay , 1993, Int. J. Circuit Theory Appl..

[7]  Henry C. Thacher,et al.  Applied and Computational Complex Analysis. , 1988 .

[8]  A. Fettweis.  A simple design of maximally flat delay digital filters , 1972 .

[9]  T. Saramaki,et al.  Optimization and efficient implementation of FIR filters with adjustable fractional delay , 1997, Proceedings of 1997 IEEE International Symposium on Circuits and Systems. Circuits and Systems in the Information Age ISCAS '97.

[10]  Vesa Vlimki,et al.  Discrete-Time Modeling of Acoustic Tubes Using Fractional Delay Filters , 1998 .

[11]  Stephan Tassart,et al.  Analytical approximations of fractional delays: Lagrange interpolators and allpass filters , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[12]  C. W. Clenshaw,et al.  The special functions and their approximations , 1972 .

[13]  Soo-Chang Pei,et al.  Closed-form design of maximally flat FIR Hilbert transformers, differentiators, and fractional delayers by power series expansion , 2001 .

[14]  Unto K. Laine,et al.  Splitting the unit delay [FIR/all pass filters design] , 1996, IEEE Signal Process. Mag..

[15]  Herbert E. Salzer.  Misstatements in Milne-Thomson, Calculus of Finite Differences, Macmillan, London, 1933 , 1980 .

[16]  K. Kunz,et al.  Numerical analysis , 1957 .

[17]  H. E. Salzer,et al.  Misstatements in: The calculus of finite differences [Macmillan, London, 1933; Zbl 8, 18] by L. M. Milne-Thomson , 1980 .

[18]  E. Hermanowicz,et al.  Explicit formulas for weighting coefficients of maximally flat tunable FIR delayers , 1992 .

[19]  Akinori Nishihara,et al.  Response of maximally flat lowpass filters to polynomial signals , 2001, ISCAS 2001. The 2001 IEEE International Symposium on Circuits and Systems (Cat. No.01CH37196).

[20]  E. Hansen.  A Table of Series and Products , 1977 .

[21]  Vesa Välimäki,et al.  Fractional Delay Filters—Design and Applications , 2001 .

[22]  G. E. Porter.  On the calculus of finite differences , 1942 .

[23]  H. Baher.  FIR digital filters with simultaneous conditions on amplitude and delay , 1982 .

[24]  Frank Stenger,et al.  Explicit nearly optimal linear rational approximation with preassigned poles , 1986 .

[25]  A. C. Aitken.  On Interpolation by Iteration of Proportional Parts, without the Use of Differences , 1932 .

[26]  Che-Ho Wei,et al.  A new variable fractional sample delay filter with nonlinear interpolation , 1991, 1991., IEEE International Sympoisum on Circuits and Systems.

[27]  Che-Ho Wei,et al.  A new variable fractional sample delay filter with nonlinear interpolation , 1992 .

[28]  M. Ahmadi,et al.  Generation of digital transfer functions of the FIR type approximating z/sup -p/q/ , 1990, IEEE International Symposium on Circuits and Systems.

[29]  C. W. Farrow,et al.  A continuously variable digital delay element , 1988, 1988., IEEE International Symposium on Circuits and Systems.

[30]  Peter J. Kootsookos,et al.  FIR approximation of fractional sample delay systems , 1996 .

[31]  I Iiiii.  PERFORMANCE AND DESIGN CONSIDERATIONS OF THE FARROW FILTER WHEN USED FOR ARBITRARY RESAMPLING OF SAMPLED TIME SERIES , 1998 .

[32]  H. Wall,et al.  Analytic Theory of Continued Fractions , 2000 .

[33]  Vesa Välimäki.  A New Filter Implementation Strategy for Lagrange Interpolation , 1995, ISCAS.

[34]  Akinori Nishihara,et al.  Universal maximally flat lowpass FIR systems , 2000, IEEE Trans. Signal Process..