A high performance, low power VLSI design of discrete wavelet transform for lossless compression in JPEG 2000 standard

Author

Dang, Philip P. ; Chau, Paul M.

Author_Institution

Dept. of Electr. & Comput. Eng., California Univ., San Diego, La Jolla, CA, USA

fYear

2001

fDate

2001

Firstpage

126

Lastpage

127

Abstract

This paper presents an efficient VLSI methodology for realizing a discrete wavelet transform for lossless compression in the JPEG 2000 standard. The proposed method takes advantages of power-of-two format of wavelet coefficients, the linear phase property of FIR filters, the canonic-sign digit representation, and the common subexpression elimination technique to obtain multiplierless wavelet transformation. The proposed approach supports fast transformation and allows achieving very compact design. Besides optimizing the speed and area, this approach significantly reduces power consumption

Keywords

CMOS digital integrated circuits; FIR filters; VLSI; channel bank filters; code standards; data compression; discrete wavelet transforms; image coding; integrated circuit design; transform coding; FIR filters; JPEG 2000 standard; canonic-sign digit representation; common subexpression elimination technique; discrete wavelet transform; high performance low power VLSI design; linear phase property; lossless compression; multiplierless wavelet transformation; power consumption; power-of-two format; wavelet coefficients; Adders; Costs; Discrete wavelet transforms; Filter bank; Finite impulse response filter; Hardware; Image coding; Performance loss; Transform coding; Very large scale integration;

fLanguage

English

Publisher

ieee

Conference_Titel

Consumer Electronics, 2001. ICCE. International Conference on

Conference_Location

Los Angeles, CA

Print_ISBN

0-7803-6622-0

Type

conf

DOI

10.1109/ICCE.2001.935240

Filename

935240