The design of nonuniform modulated filterbanks

We present an approach to nonuniform filterbank design based on modulated filters and the principle of adjacent channel aliasing cancellation. The approach is a generalization of pseudo QMF designs to nonuniform channel arrangements and the essential idea is to form nonuniform filterbanks from uniform sections joined by "transition" filters. All channel filters are formed by modulating lowpass prototypes. To meet the aliasing cancellation conditions, the transition filters are modulated from complex lowpass prototypes, although the resulting channel filters are still real. The approach provides a computationally simple method for designing filterbanks with large numbers of channels because of the fact that channel filters are modulated from a few lowpass prototypes and because the lowpass prototype designs can be scaled. That is, designs for narrow channel spacing can be produced from designs for wider channel spacing by interpolating the lowpass prototype impulse response. The development is limited to the design of nonuniform filterbanks with integer decimation factors, although we demonstrate that the technique can be used to produce excellent rational decimation factor designs by combining channel outputs using an inverse polyphase transform.

[1]  P. Chu Quadrature mirror filter design for an arbitrary number of equal bandwidth channels , 1985, IEEE Trans. Acoust. Speech Signal Process..

[2]  Mark J. T. Smith,et al.  Nonuniform filter banks: a reconstruction and design theory , 1993, IEEE Trans. Signal Process..

[3]  Richard V. Cox,et al.  The design of uniformly and nonuniformly spaced pseudoquadrature mirror filters , 1986, IEEE Trans. Acoust. Speech Signal Process..

[4]  R. Crochiere,et al.  Quadrature mirror filter design in the time domain , 1984 .

[5]  Mark J. T. Smith,et al.  Time-domain filter bank analysis: a new design theory , 1992, IEEE Trans. Signal Process..

[6]  E. Zwicker,et al.  Subdivision of the audible frequency range into critical bands , 1961 .

[7]  Mark J. T. Smith,et al.  A new filter bank theory for time-frequency representation , 1987, IEEE Trans. Acoust. Speech Signal Process..

[8]  Henrique S. Malvar Modulated QMF filter banks with perfect reconstruction , 1990 .

[9]  Truong Q. Nguyen Near-perfect-reconstruction pseudo-QMF banks , 1994, IEEE Trans. Signal Process..

[10]  P. P. Vaidyanathan,et al.  Cosine-modulated FIR filter banks satisfying perfect reconstruction , 1992, IEEE Trans. Signal Process..

[11]  Martin Vetterli,et al.  Perfect reconstruction FIR filter banks: some properties and factorizations , 1989, IEEE Trans. Acoust. Speech Signal Process..

[12]  Henrique S. Malvar,et al.  Signal processing with lapped transforms , 1992 .

[13]  Joseph Rothweiler,et al.  Polyphase quadrature filters-A new subband coding technique , 1983, ICASSP.

[14]  P. Vaidyanathan Multirate Systems And Filter Banks , 1992 .

[15]  Bernd Edler Codierung von Audiosignalen mit überlappender Transformation und adaptiven Fensterfunktionen , 1989 .

[16]  P. Vaidyanathan,et al.  Non-uniform multirate filter banks: theory and design , 1989, IEEE International Symposium on Circuits and Systems,.

[17]  T. Q. Nguyen,et al.  A simple design method for nonuniform multirate filter banks , 1994, Proceedings of 1994 28th Asilomar Conference on Signals, Systems and Computers.

[18]  Jelena Kovacevic,et al.  Perfect reconstruction filter banks with rational sampling factors , 1993, IEEE Trans. Signal Process..