A simple parallel architecture for discrete wavelet transform

Abstract

In this paper, we present a simple parallel architecture for Discrete Wavelet Transform (DWT). Efficient computation of the pyramid algorithm for the computing of the discrete wavelet transform is possible due to the similarity between computation results of each octave. By using similarity, we separated filter into 2 parts, even filter and odd filter. 1 octave and other octave computation are pcrformed in even and odd filter at the same time. Proposed architecture has following features. 1) Critical path is 1 multiply and 1 adder, 2) Number of required register is l+J*( rLd21-1) + 1+J*( [L1/21 -1) + J , where J is the number of octaves ,I-h is length of highpass filter and LI is length of lowpass filter.