🚀 go-pugleaf

From xianggen@solar.usc.edu Thu Nov 19 13:10:50 1992
Return-Path: <xianggen@solar.usc.edu>
Received: from rpi.edu by mail1.its.rpi.edu (4.1/SMHUB40);
	id AA21102; Thu, 19 Nov 92 13:10:49 EST for sassoj
Received: from usc.edu by rpi.edu (4.1/SMHUB39);
	id AA24787; Thu, 19 Nov 92 13:10:44 EST for sassoj@rpi.edu
Received: from solar.usc.edu by usc.edu (5.64+/SMI-3.0DEV3-USC+2.3)
	id AA27270; Thu, 19 Nov 92 10:10:30 PST
Received: by solar.usc.edu (4.1/SMI-3.0DEV3-ucs+1.1) id AA24810; 
                Thu, 19 Nov 92 09:51:12 PST
Date: Thu, 19 Nov 92 9:51:05 PST
From: Xianggen Xia <xianggen@solar.usc.edu>
To: sassoj@rpi.edu, i0c0256@zeus.tamu.edu
Subject: Re: CWT
Message-Id: <CMM.0.90.2.722195465.xianggen@solar.usc.edu>
Status: R

Recently we got some results on the computation of continuous wavelet
transform by using discrete wavelet transform (fast) to sampled
signals. The algorithm is called the
Shensa algorithm which generalizes
the Mallat algorithm by applying DWT to a prefiltered sampled signal
instead of sampled signal directly. We formulated the optimal
prefilter in terms of a signal or a class of signals
such that the error between the computed and the desired ones is
minimized. Howevere, all
results we obtained are only based on dyadic grid, that is, a=2^{-j}
and b=k2^{-j} where k and j are integers. If you are interested, I can
send you our papers.

Xianggen Xia
USC