DocumentCode
748486
Title
Modulated, perfect reconstruction filterbanks with integer coefficients
Author
Mertins, Alfred ; Karp, Tanja
Author_Institution
Sch. of Electr., Comput., & Telecommun. Eng., Wollongong Univ., NSW, Australia
Volume
50
Issue
6
fYear
2002
fDate
6/1/2002 12:00:00 AM
Firstpage
1398
Lastpage
1408
Abstract
We present design methods for perfect reconstruction (PR) integer-modulated filterbanks, including biorthogonal (low-delay) filterbanks. Both the prototype filter and the modulation sequences are composed of integers, thus allowing efficient hardware implementations and fast computation. To derive such filterbanks, we first start with the PR conditions known for cosine modulation and extend them to more general, integer modulation schemes. For the design of biorthogonal PR integer prototypes, a lifting strategy is introduced. To find suitable integer modulation schemes, new algebraic methods are presented. We show solutions where the PR conditions on the prototype filters and the modulation matrices are entirely decoupled and where some simple coupling is introduced. Both even and odd numbers of channels are considered. Design examples are presented for both cases
Keywords
channel bank filters; data compression; delays; digital filters; image coding; image reconstruction; matrix algebra; modulation; network synthesis; algebraic methods; biorthogonal PR integer prototypes; biorthogonal filterbanks; computational complexity; cosine modulation; finite-precision arithmetic; image compression; integer coefficients; integer-modulated filterbanks; lifting strategy; low-delay filterbanks; modulation matrices; perfect reconstruction filterbanks; Arithmetic; Australia; Delay; Design engineering; Design methodology; Filters; Hardware; Prototypes; Signal processing; Telecommunication computing;
fLanguage
English
Journal_Title
Signal Processing, IEEE Transactions on
Publisher
ieee
ISSN
1053-587X
Type
jour
DOI
10.1109/TSP.2002.1003063
Filename
1003063
Link To Document