[Home] [Back Issues] [Search] [Calendar] [Vol 1] [Vol 2] [Vol 3] [Vol 4] [Vol 5] [Vol 6] [Vol 7] [Vol 8]

The Wavelet Digest, Volume 8, Issue 9 (November 17, 99)





Today's Editor: Wim Sweldens
                Lucent Technologies, Bell Laboratories
                http://wim.sweldens.com



Today's Topics:

     1: Book:     Discovering Wavelets
     2: Preprint: Wavelet Thresholding via MDL
     3: Preprint: Generating matrices for fractal interpolation functions
     4: Preprint: Minimum risk threshold for noise reduction
     5: Preprint: (+ Software) Fractional splines 
     6: Preprint: Fast Multiple Antenna Differential Decoding
     7: Preprint: Image noise reduction using wavelets and Bayes
     8: Preprint: Preprints on the CBC algorithm
     9: Preprint: Fast cross-validatory choice of wavelet smoothness
    10: Preprint: Robust Wavelet Denoising
    11: Preprint: Minimax threshold for denoising complex signals
    12: Preprint: Wavelet Shrinkage for Unequally Spaced Data
    13: Preprint: two papers on easily-generated function analysis
    14: Preprint: Eduction of Unsteady Structure in a Turbulent Jet
    15: Preprint: Matched Wavelet approach for EMG signals
    16: Software: Image Fusion Toolbox for Matlab
    17: Web:      Web page on lifting
    18: Web:      Optimal Wavelet Zonal Compression Lecture
    19: Meeting:  IEEE International Conference On Multimedia And Expo
    20: Meeting:  Lifting session at the SPIE meeting in San Diego
    21: Job:      Postdoctoral Positions in The Netherlands
    22: Contents: JoCAAA REPORT 9/99
    23: Contents: J. Approx. Th. - sep99 TOC
    24: Answer:   Wavelets for curves (WD 8.8 #35)
    25: Answer:   Matlab files for M-band wavelets (WD 8.8 #26)
    26: Question: Wavelet Network Software
    27: Question: Signal classification by wavelets ?
    28: Question: 2D wavelet transform to sparse matrix
    29: Question: Wavelets and motion estimation
    30: Question: wavelet transform-invariant representation
    31: Question: Wavelets and seismological data 
    32: Question: Transforming rectangular input
    33: Question: Filter coefficients for orthogonal wavelets
    34: Question: DWT Coefficient derivatives
    35: Varia:    New address for Angela Kunoth (GAMM Adaptive Method Workshop)
     
Current number of members: 14749

E-mail:
   General help:      help@wavelet.org
   Add yourself:      add@wavelet.org
   Remove yourself:   remove@wavelet.org with Subject: $ADDRESS
   Publish in the WD: publish@wavelet.org 
   Publishing policy: policy@wavelet.org

Visit us on the Web at http://www.wavelet.org/

Calendar of events:

 Nov 15-20: Int. Conf. on Wavelet Analysis and its Applications
 Dec 3-4:   Perspectives in Applied Math (iho Gil Strang), MA
 Dec 3-4:   Conference on Mathemat. and Computat. Music, Vienna
 Dec 16-17: Workshop, Mathematics and Information Proc., Signapore

--- 2000
 Jan 21-23: Adaptive Methods and Error Estimators, Kiel, Germany
 May 17-20: Trends in Approximation Theory 2000, Nashville, TN
 Mar 12-16: Australian Society for Exploration Geophysics, Perth
 Jun 5-9:  Wavelets at the SIMAI 2000 conference, Ischia, Italy
 Jul 30 - Aug 2: SPIE Wavelet Applications VIII, San Diego
*Jul 30 - Aug 2: Lifting session at the SPIE meeting,  San Diego
*Jul 30 - Aug 2: IEEE Multimedia And Expo, New York







Date: Mon, 18 Oct 1999 08:41:06 -0400
From: Edward Aboufadel <aboufade@gvsu.edu>
Subject: #1 Book: Discovering Wavelets

"Discovering Wavelets" has just been published by Wiley and Sons.  The
purpose of this book is to make wavelets accessible to anyone with a
background in basic linear algebra (for example, graduate and
undergraduate students), and to serve as an introduction for the
nonspecialist.  The level of the applications and the format of this
book make it an excellent textbook for an introductory course on
wavelets or as a supplement to a first or second course in linear
algebra or numerical analysis.

For more information about the book, including a table of contents:

www.gvsu.edu/mathstat/wavelets/wavebook.htm

To order the book online, use the following URL:

www.gvsu.edu/mathstat/wavelets/acses.htm

The Discovering Wavelets web site is:

www.gvsu.edu/mathstat/wavelets.htm

Edward Aboufadel & Steven Schlicker
Grand Valley State University
e-mail:  aboufade@gvsu.edu



Date: Wed, 29 Sep 1999 10:42:42 -0400
From: "Bin Yu" <binyu>
Subject: #2 Preprint: Wavelet Thresholding via MDL

Title: 
  Wavelet Thresholding via MDL: Simultaneous Denoising and Compression

Authors: 
  Mark Hansen, Bell Laboratories, Murray Hill, NJ, 
  Bin Yu, Bell Laboratories, Murray Hill, NJ and
          University of California at Berkeley, CA.

Abstract:
  In the context of wavelet denoising and compression, we study
  minimum description length (MDL) criteria for model selection
  criteria as flexible forms of thresholding.
  Mixture MDL methods based on a single Laplacian, a two-piece
  Laplacian, and a generalized Gaussian prior are shown to be adaptive
  thresholding rules.
  While achieving mean squared error performance comparable with other
  popular thresholding schemes, the MDL procedures tend to keep far
  fewer coefficients. From this property, we demonstrate that our
  methods represent excellent tools for simultaneous denoising and
  compression. We make this claim precise by analyzing MDL
  thresholding in two optimality frameworks; one in which we measure
  rate and distortion based on quantized coefficients and one in which
  we do not quantize, but instead record rate simply as the number of
  non-zero coefficients.

Status: 
  submitted to Special Issue on Information-Theoretic Imaging
  of IEEE Trans. Information Theory

Download: 
  /stat/binyu/publications.html



Date: Wed, 29 Sep 1999 20:38:00 +0400
From: "Leonid Levkovich" <levkovl@spp.keldysh.ru>
Subject: #3 Preprint: Generating matrices for fractal interpolation functions

I would like to announce a paper.

Title: "Wavelet-based determination of generating matrices for fractal
interpolation functions", Regular and Chaotic Dynamics, v.3, #2,
1998 = (http://web.uni.udm.ru/~rcd/contents/issues.html).

Author: L.I. Levkovich-Maslyuk (The Keldysh Institute of Applied
Mathematics, Russian Academy of Sciences).

Abstract.  Fractal interpolation functions have become popular after
the works of M.Barnsley and co-authors on iterated function
systems. We consider here the following problem: given a set of values
of a fractal interpolation function (FIF), determine the contractive
affine mappings generating this function. The suggested solution is
based on the observation that the fixed points of some of the affine
mappings in question are among the points where the FIF has its
strongest singularity. These points may be detected with the aid of
wavelet-based techniques, such as modulus maxima lines tracing. After
this is done, necessary matrices are computed from a system of linear
equations. The method was tested numerically on FIFs with local
H\"older exponent as low as 0.3, and allowed to recover the generating
matrices almost precisely. When applied to segments of financial time
series, this approach gave FIFs reproducing some of the apparently
chaotic patterns in the series. This suggests the potential usefulness
of this techniques for detection of hidden rescaling parameters in the
observed data.



Date: Mon, 4 Oct 1999 16:21:13 +0200 (MET DST)
From: Maarten Jansen <Maarten.Jansen@cs.kuleuven.ac.be>
Subject: #4 Preprint: Minimum risk threshold for noise reduction

Title: Asymptotic behavior of the minimum mean squared error threshold for
       noisy wavelet coefficients of piecewise smooth signals

Authors: Maarten Jansen and Adhemar Bultheel, K.U.Leuven, Belgium

Abstract:

This paper investigates the minimum risk threshold for wavelet
coefficients with additive, homoscedastic, Gaussian noise, and for a
soft-thresholding scheme.  We start from $N$ samples from a signal on
a continuous time axis.  For piecewise smooth signals, and for $N \to
\infty$, this threshold behaves as $C \sqrt{2\log N} \sigma$, where
$\sigma$ is the noise standard deviation.  The paper contains an
original proof for this asymptotic behavior as well as an intuitive
explanation.  This behavior is necessary to prove the asymptotic
optimality of a generalized cross validation procedure in estimating
the minimum risk threshold.

Status: technical report TW294, October 1999, submitted for publication

Postscript file can be found at:
http://www.cs.kuleuven.ac.be/~maarten/publications/TW294.ps.gz

Maarten Jansen
http://www.cs.kuleuven.ac.be/~maarten/



Date: Thu, 21 Oct 1999 18:46:24 +0200
From: Michael Unser <michael.unser@epfl.ch>
Subject: #5 Preprint: (+ Software) Fractional splines 

New Wavelets With A Continuously-Varying Order Parameter

The fractional splines are an extension of the polynomial splines for
all fractional degrees alpha > -1. Their basic constituents are
piecewise power functions of degree alpha. One constructs the
corresponding B-splines through a localization process similar to the
classical one, replacing finite differences by fractional
differences. The fractional B-splines share virtually all the
properties of the classical B-splines, including the two-scale
relation, and can therefore be used to define new wavelet bases with a
continuously-varying order parameter. They only lack positivity and
compact support. The fractional splines have the following remarkable
properties:

- Generalization: for alpha integer, they are equivalent to the classical
polynomial splines.
- Adjustable holder regularity: alpha.
- Decay: inverse polynomial; compactly supported for alpha integer.
- Fractional order of approximation: alpha + 1
- Vanishing moments: the fractional spline wavelets have ceil(alpha)+1
vanishing moments, while the fractional B-splines reproduce the polynomials
of degree ceil(alpha).
- Fractional spline wavelets behave like fractional derivative operators of
order alpha + 1.

More information and papers on these wavelets can be found at
http://bigwww.epfl.ch/demo/fractsplines/index.html

The web site also includes demos, animations, and fractional spline
art.  You can visualizes fractional B-splines and fractional wavelet
bases and apply a wavelet transform of order pi (or some other number)
to one of your favorite images.

Downloadable matlab code that implements these new wavelet transforms is
available at:
http://bigwww.epfl.ch/blu/fractsplinewavelets/index.html

REFERENCES
M. Unser and T. Blu, "Fractional splines and wavelets," SIAM Review,
Accepted for publication, June 1999.
M. Unser and T. Blu, "Construction of fractional spline wavelet bases,"
Proc. SPIE vol 3813, Wavelet Applications in Signal and Image Processing
VII, Denver, CO, July 18-23, 1999, in press.

Michael Unser, Professor
Biomedical Imaging Group
Swiss Federal Institute of Technology
EPFL, DMT/IOA
BM 4.127
CH-1015 Lausanne
Switzerland

E-Mail: michael.unser@epfl.ch
URL    : http://bigwww.epfl.ch/

Tel.: +41(21)693.51.75
Secr.: +41(21)693.51.85
Fax : +41(21)693.37.01



Date: Fri, 22 Oct 1999 13:22:20 -0400
From: Wim Sweldens <wim@bell-labs.com>
Subject: #6 Preprint: Fast Multiple Antenna Differential Decoding

Title: Fast Multiple Antenna Differential Decoding

Authors: Kenneth L. Clarkson, Wim Sweldens, Alice Zheng 
         Bell Laboratories, Lucent Technologies

Abstract: We present an algorithm based on lattice reduction for the
fast decoding of diagonal differential modulation across multiple
antenna. While the complexity of the maximum likelihood algorithm is
exponential both in the number of antenna and the rate, the complexity
of our approximate lattice algorithm is polynomial in the number of
antennas and the rate. We show that the error performance of our
lattice algorithm is very close to the maximum likelihood algorithm.

Status: Technical Memorandum, Bell Laboratories, Lucent Technologies,
October 1999.

ps: This paper give an algorithm on how to find the closest vector
in a frame used for wireless data communication.

On the web at:

   http:cm.bell-labs.com/who/wim/papers/fast/

Wim Sweldens
wim.sweldens.com



Date: Mon, 25 Oct 1999 18:28:13 +0200 (MET DST)
From: Maarten Jansen <Maarten.Jansen@cs.kuleuven.ac.be>
Subject: #7 Preprint: Image noise reduction using wavelets and Bayes

Title: Empirical Bayes approach to improve wavelet thresholding for image noise
       reduction

Authors: Maarten Jansen and Adhemar Bultheel, K.U.Leuven, Belgium

Abstract:

Wavelet threshold algorithms replace wavelet coefficients with small
magnitude by zero and keep or shrink the other coefficients. This is
basically a local procedure, since wavelet coefficients characterize
the local regularity of a function.  Although a wavelet transform has
decorrelating properties, structures in images, like edges, are never
decorrelated completely, and these structures appear in the wavelet
coefficients. We therefore introduce a geometrical prior model for
configurations of large wavelet coefficients and combine this with the
local characterization of a classical threshold procedure into a
Bayesian framework. The threshold procedure selects the large
coefficients in the actual image. This observed configuration enters
the prior model, which, by itself, only describes configurations, not
coefficient values. In this way, we can compute for each coefficient
the probability of being ``sufficiently clean''.  The parameters of
the prior model are estimated on an empirical basis.

Status: technical report TW296, October 1999, submitted for publication

Postscript file can be found at:
http://www.cs.kuleuven.ac.be/~maarten/publications/TW296.ps.gz

Maarten Jansen
http://www.cs.kuleuven.ac.be/~maarten/



Date: Tue, 26 Oct 1999 02:15:17 -0400 (EDT)
From: Bin Han <bhan@math.Princeton.EDU>
Subject: #8 Preprint: Preprints on the CBC algorithm

Hello, Dear Wavelet Friends:

        I'd like to bring to your attention of some joint work in
several papers done jointly with Professor Di-Rong Chen, Professor
Rong-Qing Jia, Professor Qing-Tang Jiang and Professor Sherman
D. Riemenschneider on construction of fundamental refinable functions,
biorthogonal wavelets, Hermite refinable functions and biorthogonal
multiwavelets.  In these papers, we discussed an algorithm called CBC
algorithm where CBC stands for coset by coset. In other words, dual
masks or interpolatory masks with vanishing moments are constructed
coset by coset, separately.

        Purposes of the CBC algorithm:

        1. To construct optimal fundamental refinable functions and
interpolatory subdivision schemes with the dilation matrix 2I_s.
        2. To construct families of quincunx fundamental refinable
functions.
        3. To Construct families of Hermite refinable function vectors
and Hermite interpolatory vector subdivision schemes in the univariate
setting.
        4. To construct biorthogonal wavelets and quincunx biorthogonal
wavelets with arbitrary order of vanishing moments for any primal mask.
        5. To construct biorthogonal multiwavelets with arbitrary order of
vanishing moments for any primal matrix mask with any dilation matrix.

        Features of CBC algorithm:
        1. Simple and easy to implement (can be implemented in C/C++).  
        2. No need to solve any equations in some cases.(This feature is
useful if you are interested in constructing biorthogonal wavelets in
R^3 or higher dimensions since it will take MAPLE a long way to find a
solution to a linear system.) 
        3. Achieve optimal vanishing moments with respect to the support.
        4. Completeness. That is, for any given primal mask and any
positive integer k, any dual mask with sum rules (also called accuracy
order or vanishing moments) of order k must come from this algorithm in
some sense.
        5. It keeps symmetry. That is, if your primal mask has certain
symmetry, the resulting dual mask also possesses such symmetry (in most
cases).
        6. No big differences in designing biorthogonal scalar wavelets
and biorthogonal multiwavelets(at least the idea is essentially the same),
even though it is a little bit more complicated for multiwavelets.

Result: By using the idea from CBC algorithm, we can prove in a
constructive way that for any matrix mask with any dilation matrix, if
there exists a dual mask of the given primal mask, then for any
positive integer k, by using the CBC algorithm, we can construct a
dual mask of the given primal mask with sum rules of order k. That is,
if you have one dual mask at hand, then you can obtain dual masks with
arbitrary order of sum rules easily from the CBC algorithm.

Algorithms: I have several general programs in MAPLE format for
constructing interpoaltory masks, quincunx interpolatory masks,
biorthogonal scalar wavelets in dimensions 1, 2, and 3 by CBC,
quincunx biorthogonal wavelets by CBC, univariate multiwavelets with
any dilation factor and any multiplicity. Such maple programs are
available from me upon request, but they are not well-documented
though the inputs are simple -- your primal mask, an initial dual mask
and an integer k to control the order of sum rules. Put such input in
a file, then the program gives you a dual mask or all possible dual
masks with sum rules of order k or 2k depending on the symmetry of
your primal mask. If someone is interested to develop C/C++ programs
for such CBC algorithms, I will be very glad to collaborate with you
to work it out.

The following is a list of papers related to the CBC algorithm. I hope
such CBC algorithms will be helpful in your research.

1. Bin Han and Rong-Qing Jia,
Optimal interpolatory subdivision schemes in
multidimensional spaces, SIAM Journal on Numerical Analysis , Vol. 36, No.
1 (1998) pp. 105-124.

-->This paper deals with analysis and construction of optimal fundamental
refinable functions with the dilation matrix 2I_s.

2. Bin Han,
Analysis and Construction of Optimal Multivariate
Biorthogonal Wavelets With Compact Support, accepted in Feb. 1998. To
appear in SIAM Journal on Mathematical Analysis (In press).

-->This paper deals with analysis and construction of optimal multivariate
wavelets. The CBC algorithm is proposed in this paper to construct dual
masks with arbitrary order of sum rules for any interpolatory mask.
A family of optimal biorthogonal wavelets is given by using a modified CBC
algorithm.

3. Di-Rong Chen, Bin Han and Sherman D. Riemenschneider,
Construction of Multivariate Biorthogonal Wavelets With Arbitrary
Vanishing Moments, submitted to Adv. Comput. Math., revised in October
1999.

-->This paper generalized the CBC algorithm in paper 2. We proposed
several versions of the CBC algorithms to construct biorthogonal
wavelets with dilation matrix 2I_s, quincunx biorthogonal wavelets and
biorthogonal wavelets on the hexagonal lattice. The construction can
deal with any primal mask and any dilation matrix. Advantages of CBC
algorithm are demonstrated in this paper. A lot of examples are given.

4. Bin Han and Rong-Qing Jia,
Quincunx Fundamental Refinable Functions and Quincunx Biorthogonal
Wavelets. submitted.

-->This paper deals with the construction of quincunx fundamental
refinable functions and quincunx of biorthogonal wavelets. Families of
quincunx fundamental refinable functions and quincunx biorthogonal
wavelets are given. A remarkable thing is that all the involved masks
have nonnegative symbols, which means the Holder regularity can be
computed exactly. the dual masks by CBC in this paper were compared
with dual masks constructed by Cohen and Daubechies in "Non-separable
bidimensional wavelet bases", Revista Matematica Iberoamericana, 9
(1993), 51--137.

5. Bin Han,
Construction of Multivariate Biorthogonal Wavelets by CBC
Algorithm (1999), Wavelet Analysis and Multiresolution Methods,
T. X. He (ed.), Marcel Dekker's Lecture Notes in Pure and Applied
Mathematics. To appear. 

-->This paper is a survey paper on the CBC algorithm. In particular, we
demonstrated that if a bivariate primal mask is symmetric about the
origin, then a family of dual masks with sum rules of order 2k can be
constructed.

6. Bin Han,
Approximation Properties and Construction of Hermite Interpolants and
Biorthogonal Multiwavelets. submitted.

-->This paper deals with construction of biorthogonal multiwavelets
with any dilation matrix in any dimension. For any primal matrix mask
with any dilation matrix, a dual mask with sum rules of order $k$ can
be constructed. The general existence result of dual masks with
arbitrary order of sum rules is proved in this paper. As an examples,
C1 dual refinable function vector with support [-4,4] for the cubic
Hermite splines is given. In this paper, we also discussed Hermite
interpolatory masks and Hermite refinable functions (which is a
generalization of a result in paper 1 jointly with Jia). Families of
Hermite interpolatory masks with any multiplicity are given. In
particular, biorthogonal multiwavelets are constructed from them by
using CBC algorithm.  We also constructed a dual refinable function
vector for a non-interpolatory B-spline function with multiplicity
given by Plonka and Strela "Construction of multi-scaling functions
with approximation and symmetry", SIAM J. Math. Anal., 29, (1998),
481-510

7. Bin Han,
Hermite Interpolants and Biorthogonal Multiwavelets with
Arbitrary Order of Vanishing Moments (1999), submitted to SPIE
Proceeding (3813) for conference at Denver, July 18-21, 1999.

-->This paper surveys the results in paper 6 in the univariate case.
More examples of biorthogonal multiwavelets with multiplicity other than 2
are given. It is much easier to read than paper 6.

8. Bin Han and Qing-Tang Jiang,
Multiwavelets on the interval, submitted.

-->In this paper, We constructed C^2 orthogonal and biorthogonal
multiwavelets with multiplicity 4 and vanishing moments 4. Multiwavelets
on the interval [0,1] are constructed. CBC algorithm is employed in this
paper to construct the dual mask.

The above papers can be downloaded (it will be available before November
10, 1999) from my web site:

Current web site at Princeton University:
               http://www.math.princeton.edu/~bhan
Or by my email address at Princeton University:
                bhan@math.princeton.edu
Permanent web site at University of Alberta: 
               http://xihu.math.ualberta.ca/~bhan

(Thanks Professor Jia for allowing me to keep my web site and account
in his computer at University of Alberta).

Paper 3 is also available from Professor Riemenschneider's web page:
        http://www.math.wvu.edu/~sherm

I warmly welcome your criticism, comments, suggestions, improvement on
the CBC algorithm. We definitely do not claim (or intend to do so)
that the CBC algorithm is the best among so many excellent ways in
constructing wavelets in the literature. We know that there are so
many powerful methods on construction of wavelets in the literature
which are better than CBC algorithm in many aspects. CBC algorithm
provides us a convenient way to construct biorthogonal
(multi)wavelets. Since the CBC algorithm works well for all the
examples we have considered so far, we'd like all of you who are
interested in the construction of wavelets to test such algorithm and
to continue the effort in constructing all kinds of wavelets by a
general fast algorithm.

With best wishes!
Bin Han

Program in Applied and Computational Math.  Office Phone: 609-258-6488
Princeton University                        Fax Number:   609-258-1735
Fine Hall, Washington Road                  Web: www.math.princeton.edu/~bhan
Princeton, NJ 08544-1000, USA                    xihu.math.ualberta.ca/~bhan 



Date: Thu, 28 Oct 1999 18:21:21 +0100
From: Guy Nason <G.P.Nason@bristol.ac.uk>
Subject: #9 Preprint: Fast cross-validatory choice of wavelet smoothness

Title:Fast cross-validatory choice of wavelet smoothness, primary
resolution and threshold in wavelet shrinkage using the
Kovac-Silverman algorithm

Author: Guy P Nason

Abstract: This article introduces a fast cross-validation algorithm
that performs wavelet shrinkage on data sets of arbitrary size and
design and also simultaneously selects good values of the primary
resolution and number of vanishing moments.  We demonstrate the
utility of our method by suggesting alternative estimates of the
conditional mean of the well-known Ethanol data set. Our alternative
estimates outperform the Kovac-Silverman method with a global variance
estimate by 25\% because of the careful selection of number of
vanishing moments and primary resolution. Our alternative estimates
are simpler than, and competitive with, results based on the
Kovac-Silverman algorithm equipped with a local variance estimate.  We
include a detailed simulation study that illustrates how our
cross-validation method successfully picks good values of the primary
resolution and number of vanishing moments for unknown functions based
on Walsh functions (to test the response to changing primary
resolution) and piecewise polynomials with zero or one derivative (to
test the response to function smoothness).

On the web at http://www.stats.bris.ac.uk:81/~guy/Research/RepByDat.html

Guy Nason is Lecturer in Statistics, Department of Mathematics,
University of Bristol, Bristol, UK.
http://www.stats.bris.ac.uk/~guy



Date: Tue, 02 Nov 1999 10:07:15 +0100
From: Sylvain Sardy <sardy@epfl.ch>
Subject: #10 Preprint: Robust Wavelet Denoising

Title: Robust Wavelet Denoising

Authors: Sardy, S., Tseng, P. and Bruce, A.

Abstract: For the problem of signal extraction from noisy data,
Waveshrink and Basis Pursuit have proven to be powerful tools both
from an empirical and asymptotic point of view.  The two procedures
are especially efficient at estimating spatially inhomogeneous
signals.  They are however based on the assumption of Gaussian noise,
and in practice the performance of Waveshrink and Basis Pursuit is
severely altered when the noise has a long tail distribution, for
instance when outliers are present.  In this paper, we investigate a
robust version of the two wavelet based estimators by using a robust
loss function instead of the $l_2$ loss function involved in
Waveshrink and Basis Pursuit.  We show how the new optimization
problem can be solved by an interior point algorithm and by a block
coordinate relaxation algorithm, and we compare the efficiency of the
two algorithms.  We illustrate the advantage and limitation of the
robust wavelet denoising procedure on simulated and real data.

Keywords: block coordinate relaxation; Basis Pursuit; interior point;
robust function estimation; wavelet; Waveshrink.

On the web at: http://statwww.epfl.ch/people/sardy/PS/article-robust.ps

Keywords: block coordinate relaxation; Basis Pursuit; interior point;
robust function estimation; wavelet; Waveshrink.

Sylvain Sardy, Swiss Federal Institute of Technology, Switzerland.



Date: Tue, 02 Nov 1999 10:18:47 +0100
From: Sylvain Sardy <sardy@epfl.ch>
Subject: #11 Preprint: Minimax threshold for denoising complex signals

Title: Minimax threshold for denoising complex signals with Waveshrink

Author: Sardy, S.

Abstract: For the problem of signal extraction from noisy data,
Waveshrink has proven to be a powerful tool both from an empirical and
asymptotic point of view. Waveshrink is especially efficient at
estimating spatially inhomogeneous signals. A key step of the
procedure is the selection of the threshold parameter. Donoho and
Johnstone propose a selection of the threshold based on a minimax
principle. Their derivation is specifically for real signals and real
wavelet transforms.  In this article we propose to extend the use of
Waveshrink to denoising complex signals with complex wavelet
transforms. We illustrate the problem of denoising complex signals
with an Electronic Surveillance application.

On the web at: http://statwww.epfl.ch/people/sardy/PS/article-minimax.ps

Key word: Nonparametric denoising; Waveshrink; Minimax; Complex signals;
Complex Wavelet Transform; Electronic Surveillance.

Sylvain Sardy, Swiss Federal Institute of Technology, Switzerland.



Date: Tue, 02 Nov 1999 10:24:42 +0100
From: Sylvain Sardy <sardy@epfl.ch>
Subject: #12 Preprint: Wavelet Shrinkage for Unequally Spaced Data

Title: Wavelet Shrinkage for Unequally Spaced Data

Authors: S. Sardy, D. B. Percival, A. G. Bruce, H.-Y. Gao and W.
Stuetzle

Abstract: Wavelet shrinkage (WaveShrink) is a relatively new technique
for nonparametric function estimation that has been shown to have
asymptotic near-optimality properties over a wide class of functions.
As originally formulated by Donoho and Johnstone, WaveShrink assumes
equally spaced data. Because so many statistical applications (e.g.,
scatterplot smoothing) naturally involve unequally spaced data, we
investigate in this paper how WaveShrink can be adapted to handle such
data. Focusing on the Haar wavelet, we propose four approaches that
extend the Haar wavelet transform to the unequally spaced case.  Each
approach is formulated in terms of continuous wavelet basis functions
applied to a piecewise constant interpolation of the observed data,
and each approach leads to wavelet coefficients that can be computed
via a matrix transform of the original data. For each approach, we
propose a practical way of adapting WaveShrink. We compare the four
approaches in a Monte Carlo study and find them to be quite comparable
in performance. The computationally simplest approach (isometric
wavelets) has an appealing justification in terms of a weighted mean
square error criterion and readily generalizes to wavelets of higher
order than the Haar.

On the web at: http://ftp.statsci.com/pub/WAVELETS/papers/unequal.ps.gz

Sylvain Sardy, Swiss Federal Institute of Technology, Switzerland.



Date: Tue, 9 Nov 1999 11:10:14 +0800 (CST)
From: yuchuanwei@21cn.com
Subject: #13 Preprint: two papers on easily-generated function analysis

1.Title: Bisic Vibrations and Anharmonic Analysis of a Vibration
          
   Abstract The simple harmonic vibration is the most important of all
   vibrations.  There are two reasons for this. One is that the simple
   harmonic vibration is simple in mathematics and mechanics. The more
   important reason is that simple harmonic vibration is basic,
   i.e. any vibration can be considered as a superposition of many
   simple harmonic vibrations with different frequencies.  This is the
   well-known Fourier analysis.  In this paper, we prove that, besides
   the simple harmonic vibration, some anharmonic vibrations are basic
   as well, because an general vibration can be considered as a
   superposition of such vibrations with different frequencies. The
   results in this paper is a generalization of Fourier analysis and a
   new method of vibration analysis.  Status: Published in
   Inter. J. of Non-linear Mechanics, Vol.34, No.6, 1061-1069,
   Nov. 1999.

2. Title: New methods for Fourier analysis and Fourier Synthesis

   A basic and important relation between square wave and sine-cosin
   function is presented, that is \sin (x) = Y(x)-{1 \over 3} Y(3x)
   -{1\over5} Y(5x)+...  +{u(2n+1) \over (2n+1)}Y((2n+1)x)+..., where
   $Y(x)={\pi \over 4}sgn(\sin x) $ is square wave, and $u(2n+1)$ is
   the Moebius function in number theory.  This relation leads to
   completely new techniques for Fourier analysis and Fourier
   synthesis.  Status: to be published in IEEE signal Proc. Lett.,
   Dec. issue, 1999.

See Back issues: WD 7.8 #7, #14, WD 7.11 #6,  WD8.6 #2.

A list of papers on easily-generated function analysis:
1. Square wave  analysis,
   J. Math. Phys., 39(1999), Augest Issue, 4226-4245,
2. Triangular wave analysis,
   Computers and Mathematics with Applications, 37, 1999, No. 6, 37-56.
3. Frequency analysis based on general periodic functions
   J. Math. Phys. 40 (1999), 3654-3684. July issue, 
4. Basic vibrations and anharmonic analysis of a vibration,
   Inter. J. of Non-linear Mechanics, 34(1999),No.6,1061-1069
5. Frequency analysis based on Easily-generated functions,
   Letter-to-editor, Applied and Computat. Harm. Analysis. (accepted)
6. Dirichlet Multiplication and Easily-generated Function Analysis,
   (accepted)
7. Square wave series and its applications, submitted.
8. New methods for Fourier analysis and Fourier synthesis,
   IEEE signal processing letters,  Dec. issue, 1999.
9. Transmission of information by biorthogonal functions,
   IEEE communications letters, (accepted)

Dr. Yuchuan Wei
Department of Electrical Engineering, Box 202,
Beijing University of Aeronautics and Astronautics,
Beijing 100083,China.
and
Dept. Physics, Beijing Univ. of Sci. and Tech.,
Beijing 100083, China
Phone: (8610)82317238(office), (8610)62327579 (home)
Fax:   (8610)82328254 
Email: yuchuanwei@21cn.com



From: "Hui LI" <li@mech.kagoshima-u.ac.jp>
Subject: #14 Preprint: Eduction of Unsteady Structure in a Turbulent Jet

Dear sir,

I would like to publish the following two abstracts of my paper in the
Wavelet Digest.

(1)Titel: "Eduction of Unsteady Structure in a Turbulent Jet by using of
Continuous and Discrete Wavelet Transforms"

Authors: Hui LI, Masahiro, TAKEI, Mitsuaki OCHI, Yoshifuru SAITO and
Kiyoshi HORII

Abstract: To evaluate unsteady eddy structures of a plane turbulent
jet in the dimension of time and scale, the velocity signals were
analyzed using the continuous and discrete wavelet transforms in this
paper. By analyzing the distribution of coefficients of continuous
wavelet transform, localized nearly periodic eddy motions with a=64ms,
100ms and 180ms were observed in the shear layer of x/d=8.5 in the
time range of t=0~220ms, 280~600ms and 680~1000ms, respectively. From
multiresolution analysis or discrete wavelet transform, the peak that
appears in the component of fluctuating velocity represented the
passing of eddy through the shear layer and concentration of the
energy of the flow at instant. The intermittent eddy phenomenon or
zero components of fluctuating velocity can be observed at higher
levels or smaller scales.

Keywords: Continuous Wavelet Transform, Discrete Wavelet Transform, Eddy,
Jet, Multiresolution Analysis, Turbulence, Unsteady Structure

Status:to be published in Transactions of the Japan Society for Aeronautical
and Space Sciences.

(2)Title:Application of Two-dimensional Orthogonal Wavelets to
Multiresolution Image Analysis of a Turbulent Jet
     Authors: Hui LI, Masahiro, TAKEI, Mitsuaki OCHI, Yoshifuru SAITO and
Kiyoshi HORII

Abstract: This paper develops an application of two-dimensional
orthogonal wavelets to turbulence for the identification of the
multiresolution turbulent structures and coherent structures. The
digital imaging slice photographs of a turbulent jet in a far field
with Re 4.5x103 and 9.0x103 are respectively decomposed into six image
components with different broader scales by the wavelet
multiresolution analysis. These image components provide visualized
information on the multi-scale structures in the turbulent jet. It was
found that the edges of the vortices at different resolutions or
scales and the coherent structure might be easily extracted. In this
problem, three types of image structure, namely the large-scale
structure near the center of jet, coherent structure in the shear
layer, and the small-scale vortices in the whole flow field, which
dominate the turbulent structure, were easily extracted. With
increasing Reynolds number, the large-scale structure becomes weaker,
and the scale of active vortices and the scale of coherent structure
decrease.

Keywords: Coherent Structure, Digital Image Processing, Jet, Multiresolution
Analysis, Orthogonal Wavelet Transform, Turbulence, Vortex

Status: Transactions of the Japan Society for Aeronautical and Space
Sciences, Vol.42 (1999), No.137, pp.120-127

(3) Title: Wavelet Analysis of the Near-Field Structure in a Bounded Jet
     Authors: Hui  LI, Tsutomu NOZAKI, Takahide TABATA and   Shinichiro
OSHIGE

Abstract: In order to reveal the flow structure of a turbulent bounded
jet in both Fourier and physical spaces, the wavelet cross spectrum
function is firstly proposed based on the wavelet transform in this
paper. Then the wavelet transform, local wavelet Reynolds stress and
wavelet Reynolds stress are used to analyze the large eddy motions and
the secondary flow structure over a measured frequency-physical
plane. The experimental fluctuating velocities at various positions in
the near field are used for the analyses. From wavelet Reynolds stress
analysis, the frequency and the location of the large-scale motion in
the shear layer and the secondary flow near the wall can be
determined. The branching structure of multi-scale eddies in the wall
shear layer becomes more complex than that in the free shear
layer. Due to the effect of the secondary flow the scales of eddy and
entrainment process become small.

Key Words: Branching Structure, Eddy, Bounded Jet, Free Shear Flow, Wall
Shear Flow, Turbulent Flow, Wavelet Transform, Local Wavelet Reynolds
Stress, Wavelet Reynolds Stress

Status: Transactions of the Japan Society for Aeronautical and Space
Sciences, Vol.42 (1999), No.135, pp.27-33

Thank you.

Hui LI,  Ph.D.
Assistant Professor
Department of Mechanical Engineering
Kagoshima University
1-21-40, Korimoto, Kagoshima City 890-0065
JAPAN
Tel: +81-99-285-8252  Fax: +81-99-250-3181
E-mail: li@mech.kagoshima-u.ac.jp
Home Page: http://www.mech.kagoshima-u.ac.jp/~li/



Date: Wed, 17 Nov 1999 15:06:59 +0100
From: Laterza Francesco <f.laterza@crf.it>
Subject: #15 Preprint: Matched Wavelet approach for EMG signals

Title: Matched Wavelet approach in stretching analysis of 
        electrically evoked surface EMG signal

Authors: G. Olmo, F. Laterza, L. Lo Presti
             Department of Electronics, Politecnico di Torino (Italy)

Abstract: During a sustained muscular contraction, variations in the
myoelectric signal parameters can be noticed, before a force decrease
is observed. This phenomenon is known as localized muscular fatigue,
and always precedes the true physical tiredness.  In particular, for
the electrically elicited contraction, the signal variations are
essentially due to two different phenomena: - a time stretching, that
can be interpreted as the occurrence of a scaling factor, due to
either an enlargement of the depolarized area or a decrease of the
conduction velocity; - a shape variation, which takes into account all
those phenomena that are not due to a time stretching, and could be
associated to a change in the recruited muscular fibers activity (for
example, a different distribution of the conduction velocity).  Both
of these effects are of potential clinical interest, as they are
related to different physical phenomena.  In this paper, we analyze
electromyography signals with the objective of quantifying the
muscular fatigue by means of proper parameters. We look for indicators
which are simple, correct, reliable, easy to interpret, able to
separately quantify the time stretching and the morphological
asymmetry of the signal during contraction.

Conclusions and further developments: We have presented a method to
extract reliable information of clinical interest from the MWT
(Matched Wavelet Transform) of electrically evoked surface EMG
signals.  The employed method is able to put into evidence specific
characteristics of the EMG signal. In particular the asymmetric
variation index is positive in all cases under study, it underlines
that the second semiwave is stretched more than the first one.  This
phenomenon could be due to either an enlargement of the repolarization
zone of the muscle fibers with respect to the depolarization zone.
This phenomenon could be due to the slower repolarization process of
the muscle fiber membrane, typical in periodic paralysis pathologies.
Moreover also the normalized amplitude variation index shows the same
behavior over each subject and underlines the energy decrease of the
signal during contraction, that is a decrease in the muscle activity.
The clinical interpretation of these results is beyond the scope of
this paper. Moreover, these preliminary results on the fatigue indices
should be integrated with more data in order to compare normal and
pathological (or altered by diet or drugs) subjects. These tasks are
subject of further research.  Finally, it is worth noticing that,
implementing the MWT as a bank of filters on a DSP, it will be
possible to improve on-line monitoring or holter systems for the
evaluation of muscle performance in neuromuscular disease and in FES
(Functional Electrical Stimulation) therapy.

Acknowledgements: The authors wish to thank Prof. R. Merletti,
ing. D. Farina and Dr. U. Dimanico for many useful conversations and
for their important suggestions.  Moreover, they wish to thank the two
anonymous rewievers for their useful comments. This work was made
possible by grants from the Istituto Superiore di Sanita` in Rome
(Italy).

The paper will be published in volume 80 issue 4 of 
Signal Processing, Elsevier Science Press 



Date: Sun, 3 Oct 1999 17:08:29 +0200
From: "Oliver Rockinger" <rockinger@gmx.de>
Subject: #16 Software: Image Fusion Toolbox for Matlab

I would like to announce the first release of the image fusion toolbox
for matlab 5.x: The image fusion toolbox (ver 1.0) for matlab 5.x
comprises a set of =20 m-file functions for the pixel-level image
fusion of spatially = registered =20 grayscale images. The toolbox
implements the following fusion methods:

linear methods
  linear superposition
  pca weighted superposition
nonlinear methods
  select minum value
  select maximum value
multiresolution approaches
  laplacian pyramid
  fsd pyramid
  ratio pyramid
  contrast pyramid
  gradient pyramid
  dwt using DBSS(2,2) wavelet
  sidwt using haar wavelet
  morphological pyramid

The functions are accompanied by an graphical user interface which =
allows
image i/o and the interactive selection of all relevant parameters.
The image fusion toolbox is free of charge for academic and non-profit
users. It can be downloaded from http://www.rockinger.purespace.de



Date: Tue, 19 Oct 1999 14:33:28 -0400
From: Wim Sweldens <wim@bell-labs.com>
Subject: #17 Web: Web page on lifting

I am building a Web page on lifting and related methods. If you
have done work, written papers, or developed software related
to lifting, please send me your URL and I will include it
in the page.

Thanks!

Wim.

--
Wim Sweldens
wim@bell-labs.com
http://wim.sweldens.com



Date: Mon, 25 Oct 1999 16:28:23 -0500
From: "Arthur Petrosian" <neuaap@ttuhsc.edu>
Subject: #18 Web: Optimal Wavelet Zonal Compression Lecture

My wavelet compression lecture (narrated) is on TechOnline University
= site.  You may access it through:

http://www.ttuhsc.edu/pages/neuro/petro/TOLU_files/tol(1).htm

Greetings,
Art Petrosian,
Asst Professor
Texas Tech University
(806)743-2495
neuaap@ttuhsc.edu



Date: Wed, 29 Sep 1999 21:04:23 -0400
From: sankar <sankar@watson.ibm.com>
Subject: #19 Meeting: IEEE International Conference On Multimedia And Expo

                          CALL FOR PAPERS 

        IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO

               JULY 30 -- AUGUST 2, 2000, NEW YORK CITY

                http://://www.ieee-multimedia.poly.edu

International Conference on Multimedia & Expo 2000 is the first IEEE
multimedia conference that will take place in joint collaboration with
four IEEE societies. The same four IEEE societies that have recently
launched the IEEE Transactions on Multimedia, namely the Circuits and
Systems Society, the Computer Society, the Signal Processing Society
and the Communications Society have again come together in making this
one-of-a-kind conference a reality.

The goal is to unify all IEEE as well as non-IEEE multimedia related
activities under the common umbrella of ICME 2000 -- the millennium
multimedia meeting. It is expected that this will become a yearly
event as the flagship Multimedia conference of the IEEE.  New York
City (the Big Apple) is uniquely suited for this first-time event in
view of its large concentration of high tech Industries, first-rate
academic institutions and numerous small multimedia companies. The
intent is to blend the traditional IEEE conference framework into an
Expo by providing the industry an opportunity to showcase products.

Authors should submit a four-page manuscript in double-column format
including authors' names, affiliations and a short abstract. Only
electronic submission (in postscript format) will be accepted.
Electronic submission procedure will be made available at the
conference web site:

        http://www.ieee-multimedia.poly.edu

Topics covered include but are not limited to the following:
        * Signal processing for media integration
        * Components and technologies for multimedia systems
        * Human-machine interface and interaction
        * Multimedia databases and file systems
        * Multimedia communication and networking
        * System integration
        * Applications
        * STDS standards and related issues

Once accepted, the authors will be asked to prepare a four-page camera ready
paper to appear in the conference proceedings and CD-ROM proceedings.

Proposals for special sessions and half-day or full-day tutorial courses
may be submitted to the respective chairs before the respective
deadlines. Visit the conference web site for further information.

                        DEADLINES

        December 1, 1999        Submission of regular papers
        February 1, 2000        Proposals for special sessions,
        panels, tutorials, demos 
        March 1, 2000           Notification of acceptance for papers
        March 15, 2000          Notification of acceptance for proposals
        April 1, 2000           Paper camera-ready copy due 

	        General Chair: 
                Sankar Basu, IBM
                sankar@watson.ibm.com



Date: Mon, 4 Oct 1999 15:09:41 -0400
From: Wim Sweldens <wim@bell-labs.com>
Subject: #20 Meeting: Lifting session at the SPIE meeting in San Diego

I am organizing a session on lifting at the SPIE meeting 
"Wavelet Applications in Signal and Image Processing VIII"
(San Diego, Jul 30 - Aug 4, 2000). Besides several
invited presentations, I would like to encourage you to
submit lifting related papers to this conference.

Conference or submission details cn be found at:

http://www.wavelet.org/wavelet/digest_08/digest_08.08.html#21
and
http://www.spie.org/web/meetings/calls/am00/confs/AM210.html

Conference chairman: Akram Aldroubi, Andrew Laine, Michael Unser.

Abstract due date: January 3, 2000 !

ps: Mention the word lifting as a keyword so your talk can
    be scheduled in the lifting section.



Date: Fri, 29 Oct 1999 17:47:09 +0200
From: C. R. Traas <traas@math.utwente.nl>
Subject: #21 Job: Postdoctoral Positions in The Netherlands

Project: Wavelets And Their Applications

This 4-year project, funded by the Dutch Science Foundation (NWO),
consists of three research projects dealing with different aspects of
wavelet research, and the establishment of a national seminar on
wavelets, including an international visitor's programme.

The projects are:

(i) MORPHOLOGICAL WAVELETS WITH APPLICATIONS IN BINARY IMAGE
PROCESSING (project leader: dr. H.J.A.M. Heijmans, Centre for
Mathematics and Computer Science, Amsterdam).

(ii) WAVELET-BASED SIGNAL DETECTION IN FUNCTIONAL NEUROIMAGING
(project leader: dr. J.B.T.M. Roerdink, Institute for Mathematics and
Computing Science, University of Groningen).

(iii) WAVELET METHODS FOR NUMERICAL ANALYSIS (project leaders:
prof. dr. C.R. Traas, University of Twente, and dr. H.G. ter Morsche,
Technical University of Eindhoven).

Below you find a more detailed description of projects (ii) and (iii),
for which we have two open positions (one PhD, one postdoc).
Applications should be addressed to the respective project leaders, as
mentioned below under the different sub-projects.

POST-DOC POSITION (30 MONTHS) AT THE UNIVERSITY OF TWENTE.

TITLE: Wavelet Methods for Numerical Analysis

SUMMARY: Wavelets and multiscale type methods are becoming
increasingly important with respect to numerical simulation and
scientific computing, in particular where large and complex problems
have to be solved. Specifically in the following tasks a significant
contribution of these methods is envisaged: Preconditioning of large
systems which may arise in the discretization of elliptic PDE's or
integral equations.  Sparsening of dense matrices.  Adaptive
approximation.  The fundamental property of wavelets to be capable to
zoom in onto local properties of a function, is a key to the
expectations mentioned above.

The research project concerns the investigation of the impact of
wavelets onto the solution methodology in and around the Boundary
Element Method, in particular the possibility of sparsening the
systems' matrix. This impact is certainly dependent on the class of
kernels considered.

RESEARCH GROUP: The project will be carried out in the research group
"Numerical Analysis" of the Mathematics Department of the University
of Twente.  The research interest of this group is in Splines and
Wavelets and their applications in PDE's, integral equations and
filtering.  There will be close cooperation with the Numerical
Analysis group at the university of Eindhoven, where related research
is done within the same covering 4-year project.

REQUIREMENTS:
1: PhD in Applied Mathematics
2: Affinity with wavelets and PDE's/Integral equations.

HOW TO APPLY: Applications for the position, including a Curriculum
Vitae (name, address, degree, date, etc.) and name and email address
of two professional references, should be addressed (by surface mail)
to
   
Prof.dr.C.R.Traas
University of Twente
Faculty of Applied Mathematics
P.O.Box 217
7500 AE Enschede
The Netherlands
Tel:   +31 53 4893408
Email: traas@math.utwente.nl   

PhD POSITION (4 YRS) AT THE UNIVERSITY OF GRONINGEN

Institute for Mathematics and Computing Science

TITLE: Wavelet-Based Signal Detection in Functional Neuroimaging

SUMMARY: Many techniques currently exist to produce images of the
human brain, such as CT (Computerized Tomography), MRI (Magnetic
Resonance Imaging) or PET (Positron Emission Tomography). These
imaging techniques are not only useful to obtain images of brain
structure, but also of brain function. The detection of significant
changes in neurological activity is not an easy task, since in general
these changes are small and distributed over the whole brain, although
not in the same amount. Statistical analysis of image differences
corresponding to different experimental conditions can be subdivided
in two classes: techniques which first partition the brain into
regions of interest, based upon anatomical structure, and techniques
in which one creates images of a statistical parameter: Statistical
Parametric Mapping, by applying statistical hypothesis testing to each
element of the difference between the images with and without
activation, based upon modelling the difference image as a realization
of a Gaussian random field.  The aim of the research is a study of
wavelet decompositions of brain images to improve signal detection,
and to establish a connection between the current detection methods in
neuroimaging.

This project is connected to GNIP: Groningen NeuroImaging Project, which
involves many research groups from the Academic Hospital (AZG) and the local
research school BCN (Behavioural, Cognitive and Neurosciences).

RESEARCH GROUP: 
The project will be carried out in the research group `Scientific Computing
and Imaging' of the Department.  The research interests of the group are
particularly concentrated in high performance computing for visual
perception and neural-net modelling, morphological image processing,
computerized tomography, scientific visualization and geometric algorithms.

REQUIREMENTS:
1. MsC in Mathematics with strong affinity to Computer Science, or vice versa.  
2. Experience in image processing and/or spatial statistics.
3. Demonstrably good computing skills.

HOW TO APPLY: 
Applications for the position, including a Curriculum Vitae (name, address,
degree, date, etc.) and name and email address of two professional
references, should be addressed (by surface mail) to:

Dr. Jos Roerdink   
University of Groningen
Department of Mathematics and Computing Science
P.O. Box 800
9700 AV Groningen
The Netherlands
Tel:    +31-50-3633931    
Email:  roe@cs.rug.nl
URL:    http://www.cs.rug.nl/~roe



Date:          Wed, 29 Sep 1999 14:19:47 -600
From: "George Anastassiou" <ANASTASG@msci.memphis.edu>
Subject: #22 Contents: JoCAAA REPORT 9/99

 Journal Of Computational Analysis And Applications
  
  Table of Contents JoCAAA Vol.1,No.2,April 1999    

1)A Learning Algorithm for the Control of Continuous Action Set-Point 
Regulator Systems,A.O.Esogbue and W.E.Hearnes,II,121

2)A General Sampling Theorem Associated with Differential Operators,
A.G.Garcia and M.A.Hernandez-Medina,147

3)Fundamental Solutions of Linear Differential Equations in the Sense
of Mnemofunctions,N.Y.Radyno and V.G.Papanicolaou,163

4)Univariate Geometric Stable Laws,T.J.Kozubowski and S.T.Rachev,177

5)Fast Iterative Methods for Solving of Boundary Nonlinear Integral
Equations with Singularity,D.Rostami Varnos Fadrani and 
K.Maleknejad,219.

 Table of Contents JoCAAA Vol.1,No.3,July 1999

1)Dedication to Professor P.L.Butzer,235

2)Introduction,G.A.Anastassiou,237

3)Simultaneous Polynomial Approximation in the Weighted Chebyshev 
Norm,T.Kilgore and M.Tasche,239

4)Interpolation on Sparse Grids and Tensor Products of Nikol'skij-
Besov Spaces,W.Sickel and F.Sprengel,263

5)General Theory of Global Smoothness Preservation by Singular 
Integrals,Univariate Case,G.A.Anastassiou and S.G.Gal,289.

 Table of Contents Journal of Computational Analysis and Applications 
 Vol.1,No.4,October 1999    

1) Sturm-Liouville Problems with Coupled Boundary Conditions and
Lagrange Interpolation Series                     319
W.N.Everitt and G.Nasri-Roudsari

2) Multivariate Geometric Stable Laws             349
T.J.Kozubowski and S.T.Rachev
 


Date: Thu, 7 Oct 1999 13:42:09 -0400 (EDT)
From: Thomas Hogan <hogan@math.ohio-state.edu>
Subject: #23 Contents: J. Approx. Th. - sep99 TOC

Table of Contents: J. Approx. Theory, Volume 100, Number 1, September 1999

Gilbert Helmberg
A corner point Gibbs phenomenon for Fourier series in two dimensions
1--43

M. Alfaro, A. Mart\'inez-Finkelshtein, and M. L. Rezola
Asymptotic properties of balanced extremal Sobolev
   polynomials: coherent case
44--59

Tam\'as Erd\'elyi
Notes on inequalities with doubling weights
60--72

Les{\l}aw Skrzypek
The uniqueness of norm-one projection in James-type spaces
73--93

Walter Roth
Korovkin approximation for weighted set-valued functions
94--112

D. Leviatan and I. A. Shevchuk
Some positive results and counterexamples in comonotone
   approximation, II
113--143

L. De Michele and D. Roux
The Gibbs phenomenon for $L_{\rm loc}^1$ kernels
144--156

I. Bruj and G. Schmieder
Best approximation and saturation on domains bounded by curves of
   bounded rotation
157--182

Borislav Bojanov
Markov interlacing property for perfect splines
183--201

Table of Contents: J. Approx. Theory, Volume 100, Number 2, October 1999

Ying Guang Shi
On some problems of P. Tur\'an concerning $L_m$ extremal polynomials
   and quadrature formulas
203--220

Yukitaka Abe and Paolo Zappa
Universal functions on complex general linear groups
221--232

Colston Chandler and Archie G. Gibson
Uniform approximation of functions with discrete approximation
   functionals
233--250

Joan Cerd\`a, Joaquim Mart/'in, and Natan Ya. Krugljak
Commutators for approximation spaces and Marcinkiewicz-type multipliers
251--265

David H. Armitage and Stephen J. Gardiner
Best harmonic and superharmonic $L^1$-approximants in strips
266--283

Georgey S. Smirnov and Roman G. Smirnov
Best uniform approximation of complex-valued functions by generalized
   polynomials having restricted ranges
284--303

Antonio J. Duran
Ratio asymptotics for orthogonal matrix polynomials
304--344

Ana Foulqui'e Moreno, Francisco Marcell\'an, and K. Pan
Asymptotic behavior of Sobolev-type orthogonal polynomials on the unit
   circle
345--363

D. Barrios Rolan\'ia, G. L\'opez Lagomasino, and H. Pijeira Cabrera
The moment problem for a Sobolev inner product
364--380

A. Aptekarev, E. Berriochoa, and A. Cachafeiro
Strong asymptotics for the continuous Sobolev orthogonal polynomials on
   the unit circle
381--391

Book reviews
392--401

Author index for Volume 100
402



Date: Wed, 29 Sep 1999 12:12:31 -0300
From: Roberto Marcondes Cesar Junior <cesar@ime.usp.br>
Subject: #24 Answer: Wavelets for curves (WD 8.8 #35)

> Question: Using wavelets on 2D curves in GIS ?
> 
> Could you please point me to some papers about using wavelets/MRA
> on 2D curves ? Especially to process/compress/store 2D vector data
> in GIS (geographical information systems) ?
> 
> Thank you very much for your help
> Radek Doulik (rodo@pha.eridan.cz)
> 

Hi.

Please take a look at:

http://www.wavelet.org/wavelet/digest_08/digest_08.08.html#5
http://www.wavelet.org/wavelet/digest_06/digest_06.05.html#22
http://www.wavelet.org/wavelet/digest_06/digest_06.06.html#13
http://www.ime.usp.br/~cpq/publ/MAC-cesar.bib
and at the references therein.

Geetings,
Roberto

    Roberto Marcondes Cesar Junior 
    Department of Computer Science 
    Institute of Mathematics and Statistics - IME 
    University of Sao Paulo - USP 
    e-mail: cesar@ime.usp.br 
    http://www.ime.usp.br/~cesar 



Date: Tue, 09 Nov 1999 09:23:50 +1100
From: David Lee <david.lee@eng.monash.edu.au>
Subject: #25 Answer: Matlab files for M-band wavelets (WD 8.8 #26)

The reply from Dr. Kautsky is as follows :-

We have worked in this area for a while. Besides the theoretical
papers we've also developed a matlab based package for using m-band
wavelets.  Unfortunately, the package is not in a customer-ready state
although there is some documentation. An outdated brief statement is
enclosed.

You can find  our papers starting from the web page below
or trying directly the ftp:

ftp://ftp.ist.flinders.edu.au/pub/wavelets/papers/

regards,
 
David Lee



Date: Wed, 29 Sep 1999 16:55:59 +0100
From: Brian Fleming <phybf1@phy.hw.ac.uk>
Subject: #26 Question: Wavelet Network Software

        Could anyone tell me if and where I can get hold of some relatively recent wavelet network software? Free (which would be nice) or commercial.

        Thank you.

        Brian Fleming (B.Fleming@hw.ac.uk)



Date: Tue, 12 Oct 1999 22:02:51 +0800
From: "nanson" <nshuang@tehlu.es.ncku.edu.tw>
Subject: #27 Question: Signal classification by wavelets ?

Hello, daer editor and all readers,

  Now I got a question as follows : A compound signal is made of two =
random noises that their statical property is so closed but not a
whole = bunch.=20 The work I must do is design a system to classify
them in real time, = ie. to detect what signal is represented
now. Hence, I came up with an = idea, using wavelet coefficient to
distinguish them. The problem I face is how to extract = their
features and what is the features of it(one kine signal) via = wavelet
transform. Could=20 you/readers give me some references or recommends
? I will be = appreciated with any comment.  Thanks in advances,

 Nan-Sheng Huang
nshuang@tehlu.es.ncku.edu.tw

[Answer WD 8.10 #17]
[Answer WD 8.10 #18]
[Answer WD 9.1 #27]



Date: Tue, 12 Oct 1999 18:14:05 +0200
From: German Larrazabal <glarra@ac.upc.es>
Subject: #28 Question: 2D wavelet transform to sparse matrix

Hello!!

Is there a C or Fortran program of public domain to calculate 2D
wavelet transform for a Sparse matrix?

Any help will be appreciated. bye.

German Larrazabal S.                          Tel: 93 401 71 87
Departamento de Arquitectura de Computadores  Fax: 93 401 70 55
Universidad Politecnica de Catalunya (UPC)
Campus Nord --- Modulo D-6 Despacho 116
C/Jordi Girona 1-3
08034 BARCELONA, ESPA#A                E-mail: glarra@ac.upc.es
http://www.ac.upc.es/homes/glarra/



Date: Tue, 19 Oct 1999 14:12:10 -0600
From: Robertson Wayne <robertsonw@ll.mit.edu>
Subject: #29 Question: Wavelets and motion estimation

Question: Could someone indicate whether or not wavelet transforms
would be applicable to the following problem.  I have several images
separated by 1-2 hrs and need to determine if there has been any
motion within the images.  If so, some direction for further study and
algorithm development would be greatly appreciated.

Wayne Robertson
MIT - Lincoln Labs
robertsonw@ll.mit.edu

[Answer WD 8.10 #15]



Date: Fri, 22 Oct 1999 15:50:18 -0500
From: "Lei Sun" <slei@eng.iastate.edu>
Subject: #30 Question: wavelet transform-invariant representation

Hi:

I am working on representing CAD models surfaces by wavelet. I
digitized surfaces in the models created by some mechanical cad
packages, such as Pro/E.  Then wavelet tranforms were run on the
digitized surfaces and extract feature vectors. Unfortunately,
orthonormal wavelets, such as Harr and Daubechies, are transform
variant, and can not be used to represent surfaces under translation,
rotation and dilation. Does anybody can let me konw referances or
ongoing researches?

Thanks.

Lei Sun
mechanical engineering dept.
Iowa state university
slei@eng.iastate.edu
515-572-4807



Date: Tue, 26 Oct 1999 09:53:32 -0400
From: Wim Sweldens <wim@bell-labs.com>
Subject: #31 Question: Wavelets and seismological data 

We are 16 year old students, and are undertaking a project concerned
with wavelets.  Using wavelets we hope to analyse seismological data,
and require assistance to do so.  We are entering a Young Scientists
Competition in January and need urgent help.

We are looking for anyone who has access to software for wavelet analysis 
which is written in Turbo Pascal, ForTran, or 'C'.

We would be extrememly grateful to anyone who helps us!

Our e-mail addresses are:
jamespmckenna@yahoo.com
and
clare_oconnell@yahoo.com

Please reply to either of the above addresses as soon possible, as we have 
limited time to complete our research.

James and Clare.



Date: Wed, 3 Nov 1999 14:38:28 +0000
From: dbritch@co.umist.ac.uk (Dave Britch)
Subject: #32 Question: Transforming rectangular input

Does anyone have any knowledge/references on how to adapt a wavelet
transform to work on images which are not a power of 2?

Thanks,

Dave

Dave Britch
MultiMedia Signal Processing Group  
Computation Department
MSS/G3, UMIST, Manchester, UK
Tel  : 0161 236 3311 ext 2081
Email: dbritch@co.umist.ac.uk



Date:   Fri, 5 Nov 1999 14:51:59 -0500 (EST)
From:   Lawrence Kolasa <lkolasa@scs.Ryerson.CA>
Subject: #33 Question: Filter coefficients for orthogonal wavelets

To whom it may concern,

Besides Daubechies book "Ten Lectures on Wavelets", where can one find
tabulated filter coefficients for compactly supported orthogonal wavelets.

Thank you,

Sebastian Ferrando  ferrando@acs.ryerson.ca
Larry Kolasa        lkolasa@acs.ryerson.ca

Ryerson Polythecnic University
Toronto Ontario

[Answer WD 8.10 #16]



Date: Fri, 12 Nov 1999 10:24:36 +1100
From: Julian Magarey <jfam@research.canon.com.au>
Subject: #34 Question: DWT Coefficient derivatives

Hello,

I am trying to locate previous work on derivative DWTs - that is,
filter bank structures that produce wavelet coefficients and their
time or space derivatives simultaneously.  I would appreciate any
pointers to published material.

Many thanks.
Julian Magarey.



Date: Wed, 03 Nov 1999 14:35:37 +0000
From: Angela Kunoth <kunoth@igpm.rwth-aachen.de>
Subject: #35 Varia: New address for Angela Kunoth (GAMM Adaptive Method Workshop)

    Please note that the following coordinates have changed !

    Local organization for the GAMM Workshop on Adaptive Methods and
    Error Estimators: 

             Angela Kunoth,
             Tel.: ++49-228-73-3144
             Fax : ++49-228-73-7527
             kunoth@iam.uni-bonn.de
    
    http://www.iam.uni-bonn.de/~kunoth/workshop.html


<webmaster@wavelet.org>