[Home]
[Back Issues]
[Search]
[Calendar]
[Vol 1]
[Vol 2]
[Vol 3]
[Vol 4]
[Vol 5]
[Vol 6]
[Vol 7]
The Wavelet Digest, Volume 7, Issue 12 (December 19, 1998)
Today's Editor: Wim Sweldens
Lucent Technologies, Bell Laboratories
wim@bell-labs.com
Today's Topics:
1: Book: Mathematics in Signal Processing
2: Preprint: Three preprints on image/video coding and transmission
3: Preprint: Report on Wavelet Shrinkage Denoising
4: Preprint: Multivariate Subdivision Schemes and Divided Differences
5: Preprint: The Comparison of Four Filters on Image Compression
6: Preprint: Wavelet analysis of non-stationary processes
7: Preprint: Error Resilient Image Coding Using Wavelets.
8: Preprint: Frames in Hilbert C* modules and algebras
9: Preprint: An Approximate Wavelet MLE...
10: Thesis: Subdivision Schemes, Wavelets and Image Compression
11: Meeting: CISS '99
12: CFP: IEEE Trans Inform Th on Information-Theoretic Imaging
13: Course: Introduction to Mathematical and Physical Wavelets
14: Job: Wavelet Research Position at MathSoft, Seattle
15: Job: Doctoral/Post-doctoral at University of Karlsruhe, Germany
16: Job: PhD position in Switzerland, Wavelets in medicine
17: Job: Postdoctoral Position at Univ. Cath. Louvain in Belgium
18: Contents: Advances in Computational Mathematics 9-3,4
19: Contents: Numerical Algorithms 18-2
20: Answer: VLSI architectures for DWT (WD 7.11 #22)
21: Answer: Vibration and fault diagnosis (WD 7.9 #19, 7.9 #21, 7.8 #26)
22: Question: Wavelets in Vibration Transients
23: Question: Decomposing frequency bands
24: Question: Looking for C,C++ code for wavelet analysis
25: Question: Continuous wavelet transform
26: Question: Cont. wavelet transform coefficients to % variance
27: Question: Non parametric model specification
28: Question: Wavelet basics
29: Question: Wavelets in ecology
30: Question: Wavelets in hydrology
31: Question: Wavelets for control of chemical engineering systems
Current number of members: 12999
E-mail:
General help: help@wavelet.org
Add yourself as a member: add@wavelet.org
Remove yourself as a member: remove@wavelet.org
Publish in the next WD: publish@wavelet.org
Receive publishing policy: policy@wavelet.org
Visit us on the Web at http://www.wavelet.org/
Calendar of events:
Jan 13-16: Introduction to Wavelets minicourse, San Antonio TX
Feb 24-25: Royal Society London, UK
Feb 24-25: Wavelets, Royal Society Meeting, London
Feb 24-26: IEEE Information Theory Workshop, Santa Fe, NM
*Mar 8-11: Course: Mathematical and Physical Wavelets, Boston, MA
*Mar 17-19: CISS '99 (Information Sciences and Systems), Baltimore
Mar 22-26: II Symposium on Artificial Intelligence, Havana, Cuba
Apr 6-8 : SPIE Wavelet Applications VI, Orlando
Jun 14-19: Mathtools 99, St-Petersburg, Russia
Jun 14-16: Fractals in Engineering, Delft, The Netherlands
Jul 1-7: Curves and Surfaces, St-Malo, France
Jul 18-23: SPIE Wavelet Applications VII, Denver, CO
Date: Thu, 17 Dec 1998 16:48:43 -0000
From: "PEACOCK, Tom" <peacockt@oup.co.uk>
Subject: #1 Book: Mathematics in Signal Processing
NEW BOOK
Mathematics in Signal Processing IV
Edited by J. G. McWhirter, Head of Signal Processing Group, Defence
Evaluation and Research Agency, Malvern, and
I. K. Proudler, Defence Evaluation and Research Agency, Malvern
This volume provides an overview of the wide range of mathematical
topics in signal processing. The three main contributions explore the
theme of alternative algebras for signal processing, covering the
already significant impact of multilinear algebra, and the potential
applications of geometric algebra and Gröbner bases. Other
chapters cover a wide range of subjects from array processing and
digital communications to wavelets and nonlinear signal
processing. These, in turn, encompass a rich variety of mathematical
techniques including Padé approximation, convex optimization
and generalized eigenvalue decomposition. This intriguing blend of
theory and application should appeal to a wide range of engineers and
mathematicians.
Contents: From matrix to tensor: multilinear algebra and signal
processing; The analysis of linear functions using geometric algebra
as an alternative to matrices and tensors; Rational approximation
using Grobner bases: some numerical results; Gröbner basis
design of incomplete Chebyshev polynomials; Fractals, linear channels,
and delay methods; Blind binary signal separation using joint
diagonalization techniques; Adaptive algorithms for single user blind
channel equalization; Assessing stability for adaptive filtering
algorithms in signal processing; Constrained Kalman filter for
least-squares estimation of time-varying beamforming weights; Using
algorithmic engineering to derive a fast parallel weight extraction
algorithm based on QRD recursive least squares; The simultaneous
strict positive real problem: similarities with the simultaneous
stability problem and applications in adaptive recursive algorithms;
Asymptotic performance analysis of Yang's subspace tracking algorithm;
Generalized prediction error approach to spectral analysis and
filtering of irregularly-sampled data; Decaying exponentials and
signal decomposition; Robust envelope-constrained filter design; On
the use of positivity in reconstructing images from noisy data;
Treatment of noise in positive reconstructions in linear inverse
problems; Ill conditioning and gradient based optimization of
multi-layer perceptrons; Online modelling of time series with resource
allocating neural networks; Bayesian model selection for linear and
non-linear time series using the Gibbs sampler; Bayesian image
estimation from sparse raw MRI data; Polynomial models for 3D motion
parameters and depth estimation in monocular computer vision; Symmetry
and wavelet transforms for image data compression; Vanishing moments
and biorthogonal wavelet systems; On sliding window schemes for
discrete least-squares approximation by trigonometric polynomials;
Jump detection by Pade approximation; Discrete simulation
models for multidimensional systems based on functional
transformations; Analytic performance evaluation of third-order
statistics based estimators for Allpass identification; Detection of
sleep apnoea using time-frequency analysis of heart rate variability;
Robust estimation of oceanic background spectrum.
Institute of Mathematics and its Applications Conference Series No.67
0-19-850202-8, 352 pages, 9 halftones, 86 line figures
November 1998 Hardback GBP95.00
Special discount of 25% available for members of the IMA. Please
contact Tom Peacock (peacockt@oup.co.uk) for further details.
Date: Thu, 19 Nov 1998 11:07:10 -1000
From: zx@lena.eng.hawaii.edu (Zixiang Xiong)
Subject: #2 Preprint: Three preprints on image/video coding and transmission
The following preprints on wavelet image coding, 3D SPIHT video coding,
and joint source channel coding are available from:
http://www-ee.eng.hawaii.edu/~zx/articles.html
1. Wavelet image coding using trellis coded space-frequency quantization
2. Very Low Bit-Rate Embedded Video Coding with 3D Set Partitioning
in Hierarchical Trees (3D SPIHT)
3. Joint Source-Channel Image Coding for a Power Constrained Noisy Channel
Zixiang Xiong, Dept. of Electrical Engineering, University of Hawaii
Date: Sat, 21 Nov 1998 14:41:19 -0800
From: Carl Taswell <taswell@toolsmiths.com>
Subject: #3 Preprint: Report on Wavelet Shrinkage Denoising
Title: The What, How, and Why of Wavelet Shrinkage Denoising
Author: Carl Taswell
Document: Technical Report CT-1998-09
Website: <www.toolsmiths.com>
Abstract: Principles of wavelet shrinkage denoising are reviewed.
Both 1-D and 2-D examples are demonstrated. The performance of various
ideal and practical Fourier and wavelet based denoising procedures are
evaluated and compared in a new Monte Carlo simulation
experiment. Finally, recommendations for the practitioner are
discussed.
Date: Mon, 23 Nov 1998 00:49:20 -0500 (EST)
From: "Igor V. Guskov" <ivguskov@math.princeton.edu>
Subject: #4 Preprint: Multivariate Subdivision Schemes and Divided Differences
Title: Multivariate Subdivision Schemes and Divided Differences
Author: Igor Guskov, Princeton University,
ivguskov@math.princeton.edu
Abstract:
A new generalization of divided differences to the multivariate case
is introduced to investigate the regularity of functions produced by
subdivision. A general commutation formula is presented for a class of
interpolating subdivision schemes. The technique developed is then
used to construct subdivision schemes on irregular multilevel meshes.
Status: submitted for publication
Available on the Web at:
http://www.math.princeton.edu/~ivguskov/papers.html
Date: Tue, 24 Nov 1998 14:39:10 +0800
From: lxj <lxj@ns.glc.cn.net>
Subject: #5 Preprint: The Comparison of Four Filters on Image Compression
Title:Comparison of Four Filters on Image Compression
Author:Xuejun Li
Wavelet transform (WT) is the cutting edge technology in image
compression. The key to image compression is wavelet decomposition,
there are many of wavelet bases to do it, such as Mallat Daubechies,
Haar and Spline basis. We should consider the three aspects of filters
in image decomposition : (1) Compact Support (2) Regularity (3)
Symmetry. The results shows that the Spline filter is the best in PSNR
and the Haar filter is the fastest in running time.
Date: Tue, 01 Dec 1998 18:49:20 +0100
From: Charles-Antoine Guerin <caguerin@math.chalmers.se>
Subject: #6 Preprint: Wavelet analysis of non-stationary processes
Title: Wavelet analysis and covariance structure of some classes of
non-stationary processes
Author: Charles-Antoine Guerin
Abstract:
We study four classes of non-stationary processes: processes with
stationary n-increments, processes with stationary fractional
increments, locally stationary processes and processes with locally
stationary n-increments. We establish a simple characterisation of
each class of processes by means of the continuous wavelet
transform. Then we give two applications of these results. First, we
derive the explicit covariance structure of processes with stationary
n-increments. Second, we prove that the fractional Brownian motion
with Hurst parameter H has stationary fractional increments of order a
for a>H.
Report 1998:53, Department of Mathematics, University of Chalmers,
Gothenburg, Sweden.
The paper can be downloaded at
http://www.math.chalmers.se/~caguerin/preprint/preprint.html
Date: Tue, 1 Dec 1998 18:01:34 -0600 (CST)
From: Sergio Servetto <servetto@ifp.uiuc.edu>
Subject: #7 Preprint: Error Resilient Image Coding Using Wavelets.
Title:
Multiple Description Wavelet Based Image Coding.
Authors:
Sergio D. Servetto, Kannan Ramchandran, Vinay A. Vaishampayan,
and Klara Nahrstedt.
Abstract:
We consider the problem of coding images for transmission over
error-prone channels. The impairments we target are transient channel
shutdowns, as would occur in a packet network when a packet is lost, or
in a wireless system during a deep fade: when data is delivered, it is
assumed to be error-free, but some of the data may never reach the
receiver. The proposed algorithm is based on a combination of multiple
description scalar quantizers with techniques successfully applied to
the construction of some of the most efficient wavelet-based image
coders. A given image is encoded into two *independent* packets of
equal length: either one of these packets guarantees a minimum image
quality, but both combined guarantee better quality than either single
one alone. We formulate and solve an optimization problem, whose
solution gives parameters of the proposed encoder yielding optimal
performance (in an operational sense). When compared with previously
reported results on the performance of robust image coders based on
multiple descriptions, on standard test images, our coder attains a
bitrate reduction of about 50-60% at equal PSNRs.
This paper was submitted last November to the IEEE Transactions on Image
Processing, and is currently under review. A preprint can be obtained from:
ftp://ftp.ifp.uiuc.edu/pub/servetto/papers/mdImages:tIP.ps.gz
Also, more information about this work (source code, matlab files with
plots, etc.) is available from our website, at
http://www.ifp.uiuc.edu/~servetto/research/
Sergio Servetto
PhD Candidate, Department of Computer Science
Graduate Research Assistant, Coordinated Science Lab. and Beckman Institute
University of Illinois at Urbana-Champaign
http://www.ifp.uiuc.edu/~servetto/
Date: Mon, 7 Dec 1998 09:58:24 -0600 (CST)
From: Michael Frank <frank@math.uh.edu>
Subject: #8 Preprint: Frames in Hilbert C* modules and algebras
Dear colleagues,
I would like to announce the existence of two preprints that are related
to questions in wavelet-theory. The authors would be glad if the
information could be included into the wavelet newsletter.
Thank you very much for your support,
greatings and best wishes
Michael Frank.
TITLE:
Frames in Hilbert C*-modules and C*-algebras
AUTHORS:
Michael Frank
Dept. Mathematics, Univ. of Houston, Houston, TX 77204-3476, U.S.A.
frank@math.uh.edu, frank@mathematik.uni-leipzig.de
David R. Larson
Dept. Mathematics, Texas A&M Univ., College Station, TX 77843, U.S.A.
larson@math.tamu.edu
HOWPUBLISHED:
preprint, LaTeX2e, 33 pages, submitted. Will be presented at the Joint
Mathematics Meeting in San Antonio, Texas, U.S.A., on January 13th, 1998.
(Section: The Functional and Harmonic Analysis of Wavelets.)
ABSTRACT:
We present a general approach to a module frame theory in C*-algebras and
Hilbert C*-modules. The investigations rely on the idea of geometric dilation
to standard Hilbert C*-modules over unital C*-algebras that possess orthonormal
Hilbert bases, and of reconstruction of the frames by projections and other
bounded module operators with suitable ranges. We obtain frame representation
and decomposition theorems, as well as similarity and equivalence results.
The Hilbert space situation appears as a special case. Using a canonical
categorical equivalence of Hilbert C*-modules over commutative C*-algebras
and (F)Hilbert bundles the results find a reinterpretation for frames in
vector and (F)Hilbert bundles.
The results are of interest in wavelet theory (esp. MRA wavelets), Hilbert
C*-module and C*-algebra theory, vector bundle theory. Particular results
on Hilbert-Schmidt operators are even new in the Hilbert space situation.
We show that any algebraically generating set of an algebraically finitely
generated Hilbert C*-module is a module frame.
TITLE:
Symmetric approximation of frames and bases in Hilbert spaces
AUTHORS:
Michael Frank
Dept. Mathematics, Univ. of Houston, Houston, TX 77204-3476, U.S.A.
frank@math.uh.edu, frank@mathematik.uni-leipzig.de
Vern I. Paulsen
Dept. Mathematics, Univ. of Houston, Houston, TX 77204-3476, U.S.A.
vern@math.uh.edu
Terry R. Tiballi
Dept. Mathematics, SUNY at Oswego, Oswego, NY 13126, U.S.A.
tiballi@oswego.edu
HOWPUBLISHED:
preprint, LaTeX2e, 16 pages, submitted.
ABSTRACT:
We consider existence and uniqueness of symmetric approximation of frames
by normalized tight frames and of symmetric orthogonalization of bases by
orthonormal bases in Hilbert spaces H . More precisely, we determine whether
a given frame or basis possesses a normalized tight frame or orthonormal
basis that is quadratically closest to it, if there exists such frames
or bases at all. A crucial role is played by the Hilbert-Schmidt property
of the operator (P-|F|) , where F is the adjoint operator of the frame
transform F*: H --> l_2 of the initial frame or basis and (1-P) is the
projection onto the kernel of F .
Date: Fri, 11 Dec 1998 10:07:56 -0600
From: jensen@haar.econ.missouri.edu
Subject: #9 Preprint: An Approximate Wavelet MLE...
Title: An Approximate Wavelet MLE of Short and Long-Memory Parameters
Author: Mark J. Jensen (University of Missouri - Columbia)
Abstract :
By design a wavelet's strength rests in its ability to simultaneously
localize a process in time-scale space. The wavelet's ability to
localize a time series in time-scale space directly leads to the
computational efficiency of the wavelet representation of a $N \times
N$ matrix operator by allowing the $N$ largest elements of the wavelet
represented operator to adequately represent the matrix operator
[Devore, et al. (1992a) and (1992b)]. This property allows many dense
matrices to have sparse representation when transformed by wavelets.
In this paper we generalize the long-memory parameter estimator of
McCoy and Walden (1996) to simultaneously estimate short and
long-memory parameters. Using the sparse wavelet representation of a
matrix operator, we are able to adequately approximate an ARFIMA
models likelihood function with the series wavelet coefficients and
their variances. Maximization of this approximate likelihood function
over the short and long-memory parameter space results in the
approximate wavelet maximum likelihood estimates of the ARFIMA model.
By simultaneously maximizing the likelihood function over both the
short and long-memory parameters, and using only the wavelet
coefficient's variances, the approximate wavelet MLE provides a fast
alternative to the frequency-domain MLE. Furthermore, the simulation
studies found herein reveal the approximate wavelet MLE to be robust
over the invertible parameter region of the ARFIMA model's moving
average parameter, whereas the frequency-domain MLE dramatically
deteriorates as the moving average parameter approaches the boundaries
of invertibility.
Available at http://econwpa.wustl.edu/eprints/em/papers/9802/9802003.abs
Mark J. Jensen Voice: (573) 882-9925
Department of Economics, Fax: (573) 882-2697
University of Missouri
118 Professional Building
Columbia, Missouri 65211 jensen@haar.econ.missouri.edu
Date: Wed, 2 Dec 1998 23:10:19 -0600 (CST)
From: Bin Han <binhan@mail.math.okstate.edu>
Subject: #10 Thesis: Subdivision Schemes, Wavelets and Image Compression
Dear wavelet friends and colleagues:
In July of 1998 I completed my Ph.D. thesis at the Department of
Mathematical Sciences in the University of Alberta. :)
PhD Thesis: Subdivision Schemes, Biorthogonal Wavelets and Image Compression
Supervisor: Professor Rong-Qing Jia
Completed: Fall 1998
Location: University of Alberta
Short Abstracts of This Thesis:
Chapter 1: A characterization of L_p (p between 1 and infinity) convergence
of any subdivision scheme associated with any general dilation
matrix in arbitrary dimensions is presented.
Chapter 2: A sharp error estimate in the L_p norm sense is given for any
multivariate subdivision scheme when there is a perturbation
in its mask.
Chapter 3: We study several optimal properties (approximation order and
smoothness order) of fundamental refinable functions with
respect to their support. A family of bivariate optimal
interpolatory subdivision schemes is presented.
Chapter 4: We study several optimal properties of multivariate
biorthogonal wavelets. A family of optimal bivariate
biorthogonal wavelets is constructed. Moreover, a general
construction by coset (CBC) algorithm is given to construct
biorthogonal wavelets with arbitrary vanishing moments. That is,
given any primal interpolatory mask, this CBC algorithm can be
used to generate all the dual masks with any preassigned order
of vanishing moments. This CBC algorithm can be easily
implemented. I have a general MAPLE program for this algorithm.
Chapter 5: We shortly discuss some experimental results by using bivariate
wavelets on image compression.
A more detailed abstract, each chapter of this thesis, and the whole
thesis are available in Postscript format or in HTML at:
http://www.math.okstate.edu/~binhan/research.html
Some preprints on wavelets are also available at
http://www.math.okstate.edu/~binhan/
Thank you for your time and have a merry Christmas! :)
Bin Han
Department of Mathematics
444 Mathematical Sciences
Oklahoma State University
Stillater, Oklahoma 74078-1058
USA
E-mail: binhan@math.okstate.edu
Permanent Email: bhan@approx.math.ualberta.ca
Web : http://www.math.okstate.edu/~binhan/
Permanent Web: http://approx.math.ualberta.ca/~bhan/
Date: Tue, 24 Nov 1998 15:49:05 -0500 (EST)
From: "Trac D. Tran" <trac@thanglong.ece.jhu.edu>
Subject: #11 Meeting: CISS '99
Submitted by: Trac D. Tran, The Johns Hopkins University
CISS '99
CALL FOR PAPERS
The 33rd Annual Conference on Information Sciences and Systems (CISS
'99) will be held on March 17-19, 1999, in Baltimore, Maryland. The
conference is sponsored by the Department of Electrical and Computer
Engineering, The Johns Hopkins University.
Authors are invited to submit previously unpublished papers describing
new advances, applications, and ideas in the fields of signal, image,
and video processing, systems and control, communications, pattern
recognition, statistical inference, and computer engineering.
Two kinds of contributed papers are solicited. (1) Regular papers
requiring approximately thirty minutes for presentation; these will be
reproduced in full (up to six pages) in the conference proceedings.
(2) Short papers suitable for presentation in 15 minutes or less; one
page summaries of these papers will be published in the proceedings.
Three copies of a summary together with a regular or short designation
are to be submitted by January 15, 1999. Summaries should be of
sufficient detail and length to permit careful reviewing. Authors
will be notified of acceptance by February 17, 1999. Instructions for
the preparation of accepted papers for the proceedings will be sent to
each author. All manuscripts are to be submitted to:
1999 CISS
105 Barton Hall
Department of Electrical and Computer Engineering
The Johns Hopkins University
Baltimore, MD 21218
Conference Office:
1999 CISS
105 Barton Hall
Department of Electrical and Computer Engineering
The Johns Hopkins University
Baltimore, MD 21218
email: ciss99@ece.jhu.edu
URL: http://www.ece.jhu.edu/ciss99/
Program Directors:
Jerry L. Prince
Trac D. Tran
Important Dates:
Submission deadline: January 15, 1999
Notification of acceptance: February 17, 1999
Advance Registration: Before March 10, 1999
Conference Dates: March 17-19, 1999
Date: Mon, 7 Dec 1998 13:56:55 -0600
From: D. L. Snyder <dls@essrl.wustl.edu>
Subject: #12 CFP: IEEE Trans Inform Th on Information-Theoretic Imaging
Submitted by: D. L. Snyder (dls@ee.wustl.edu)
Call for papers for a special issue of the IEEE Transactions
on Information Theory on Information-Theoretic Imaging
The Transactions on Information Theory is soliciting original papers
for a special issue to be published in 2000 on information-theoretic
imaging. The last decade has witnessed impressive advances in
statistical imaging and in statistical, model-based image processing
in general. Recent research has addressed two categories of problems:
1, novel applications of fundamental statistical and information
theoretic principles to imaging; and 2, exploring optimal and
sub-optimal methods for extracting information from intrinsically high
dimensional image data. In the first category are topics such as
hierarchical image modeling and representation,image compression and
coding, minimax analysis, robustness analysis, learning theory,
statistical pattern recognition and pattern matching, inference from
compressed image-data, and fundamental performance bounds on
estimation, detection, classification and compression. In the second
category are topics such as approximation theory for multidimensional
objects, multi-resolution data analysis, content based indexing of
image databases, projection pursuit methods, and image reconstruction
from incomplete and noisy data; for example, from magnitude-only
Fourier data or tomographic projections. The first goal of this
special issue is to publish original papers addressing fundamental
theoretical and computational aspects of such problems and at the same
time raising the awareness for such research within the
information-theory community. The special issue will complement the
upcoming Information Theory Workshop on Detection, Estimation,
Classification and Imaging, to be held in Santa Fe, New Mexico, in
February, 1999. A long-term goal is to build a broad, high-quality
forum for addressing imaging problems of fundamental
information-theoretic significance and to help bridge the current,
significant gap that exists between emerging, advanced theoretical
concepts and image processing practice.
Schedule:
First call for papers: November, 1998
Submission deadline: September 1, 1999
Notification of decisions: February, 2000
Publication: August, 2000
Manuscripts should be submitted by August 1, 1999
to the Guest Editor-in-Chief:
Donald L. Snyder
Department of Electrical Engineering, Box 1127
Washington University
One Brookings Drive
St. Louis, Missouri 63130-4899
The Guest Associate-Editors for the special issue will be:
Alfred O. Hero III
Dept. of Electrical Engineering and Computer Science
The University of Michigan
1301 Beal Avenue
Ann Arbor, MI 48109-2122
Pierre Moulin
University of Illinois at Urbana-Champaign
2265 Beckman Institute
405 North Mathews
Urbana, IL 61801
Jose M. F. Moura
Department of Electrical Engineering
Carnegie Mellon University
5000 Forbes Avenue
Pittsburgh, Pennsylvania 15213-3890
Joseph A. O'Sullivan
Department of Electrical Engineering, Box 1127
Washington University
One Brookings Drive
St. Louis, Missouri 63130-4899
Date: Thu, 10 Dec 1998 06:57:22 EST
From: atiinfo@aol.com
Subject: #13 Course: Introduction to Mathematical and Physical Wavelets
COURSE ANNOUNCEMENT:
A Detailed Introduction to Mathematical & Physical Wavelets: March 8-11,1999
March 8-11, 1999 at the Days Inn, 1234 Soldiers Field Rd,
Boston, MA 02135, Phone (617) 254-1234.
http://catalog.com/hitekweb/sked.htm
SUMMARY:
Wavelet analysis has undergone an explosive growth in the past few
years, with many successes in the efficient analysis, processing, and
compression of signals and images. It is based on the idea of
time-scale decomposition of signals and images, which is fundamentally
different from the traditional frequency and time-frequency
decompositions. Unfortunately, many potential users of this new
technique are hindered by the inaccessibility of its mathematical
framework, partly due to its novelty. The objectives of this course
are as follows:
a) To give a practical introduction to space-time-scale analysis in
one and two dimensions, and show that it may be regarded as a wideband
generalization of the usual time-frequency analysis.
b) To develop space-time-scale analyses specifically dedicated to
acoustic and electromagnetic waves, and apply them to sonar and radar.
Each participant will receive a copy of Dr. Kaiser's textbook "A
Friendly Guide to Wavelets," as well as a newly updated 235-page
packet of typeset lecture notes with many graphical examples. For a
description and reviews of the textbook, please visit the author's
website at www.wavelets.com.
INSTRUCTOR:
Dr. Gerald Kaiser heads the Virginia Center for Signals and Waves. He
has a Ph.D. in physics from the University of Wisconsin-Madison, and a
Ph.D. in mathematics from the University of Toronto. His research has
concentrated on developing a theory of acoustic and electromagnetic
wavelets with applications to sonar and radar. He has published many
papers on mathematics, physics and signal processing, as well as two
books: Quantum Physics, Relativity, and Complex Spacetime and A
Friendly Guide to Wavelets, (Birkhauser, 1994).
REGISTRATION AND INFORMATION:
For registration or more information call (410) 531-6034, email
"atiinfo@aol.com" or visit the ATI homepage http://catalog.com/hitekweb
For Technical Information contact:
Dr. Gerald Kaiser, Director
email: kaiser@wavelets.com
Date: Tue, 1 Dec 1998 09:26:34 -0800 (PST)
From: Jill Goldschneider <jrgold@statsci.com>
Subject: #14 Job: Wavelet Research Position at MathSoft, Seattle
WAVELET RESEARCH POSITION AT MATHSOFT, SEATTLE, WA
The Research Group of MathSoft, Inc. in Seattle, WA. is seeking
engineers and scientists to fill a Research Scientist position. The
ideal candidate will have expertise in wavelets (from theory to
practical implementations) as well as wavelet processing and analysis
techniques. The candidate will further strengthen the research program
at MathSoft by defining new research directions and developing new
product ideas. He/she will not only guide the implementation of
state-of-the-art wavelet algorithms, but will also be responsible for
the software engineering aspects of the product.
The candidate should have a Ph.D. in Statistics, Mathematics, Applied
Mathematics, Electrical Engineering, or related field, and have strong
communication skills including the ability to write proposals, make
technical presentations, and write technical reports. The candidate
will also be an experienced programmer with fluency in the
C-language. He/she will be a team player and will have good
interpersonal skills.
The Research Group at MathSoft is a dynamic group with expertise in
image processing and analysis, wavelets, signal processing, and
statistics. A variety of projects are underway, ranging from
commercial and custom software development to research in wavelets,
partial differential equation-based image processing, image
segmentation, query-by-image content, image compression, and target
detection for applications including medical imaging, document
imaging, tactical imaging, and remote sensing.
Mail, fax, or e-mail your resume to: Human Resources, Data Analysis
Product Division, MathSoft, Inc., 1700 Westlake Ave. N., Suite 500,
Seattle, WA 98109 FAX: (206) 283-6310. e-mail: hr@statsci.com
Date: Wed, 2 Dec 1998 13:20:37 +0100 (MEZ)
From: "Kai Schneider [Chemie]" <kschneid@rhrk.uni-kl.de>
Subject: #15 Job: Doctoral/Post-doctoral at University of Karlsruhe, Germany
DOCTORAL / POST-DOCTORAL POSITION
AT UNIVERSITY OF KARLSRUHE, GERMANY
TMR NETWORK
Wavelets and Multiscale Methods in Numerical Analysis and Simulation.
The ÔInstitut f\"ur Chemische TechnikÕ offers doctoral and
post-doctoral grants to conduct research within the TMR project
"Wavelets and Multiscale Methods in Numerical Analysis and
Simulation". The grants are available in the period January 1, 1999 -
March 31, 2001, with a minimal duration of 3 months. The salary will
be fixed according to the situation of the applicant.
Young scientists whose research interests are in one of the areas:
applied harmonic analysis, partial differential equations, numerical
analysis, chemical engineering, turbulence and reactive flows and
scientific computing are encouraged to apply.
The following topics: modelling of reactive and non-reactive turbulent
flows, adaptive methods for linear and nonlinear PDEs, multiscale
methods for PDEs, spectral methods, turbulence modelling, interaction
of turbulence and chemical reactions, numerical simulation of
turbulence, scientific computing and related areas of studies are
among the interests of our group.
Candidates have to be nationals of a European Community Member State
other than Germany or a state associated with the TMR program and
she/he must not be older than 35 years. People being interested are
kindly invited to contact us by e-mail. They should submit their
application by sending: Curriculum Vitae, copies of certificates, list
of publications, research interests and one letter of recommendation
to the following address:
Prof. Dr. Henning Bockhorn
Institut f\"ur Chemische Technik
Universit\"at Karlsruhe (TH)
Kaiserstrasse 12
76128 Karlsruhe
Germany
Tel: + 49 721 608 2120
Fax: + 49 721 608 4820
email: bockhorn@ict.uni-karlsruhe.de
WWW : http://www.ict.uni-karlsruhe.de/
Date: Fri, 04 Dec 1998 14:50:21 +0100
From: Michael Unser <michael.unser@epfl.ch>
Subject: #16 Job: PhD position in Switzerland, Wavelets in medicine
An assistant position for a Ph.D. candidate is available within the
Biomedical Imaging Group, Swiss Federal Institute of Technology
(EPFL), Lausanne, Switzerland. The research involves the development
of new wavelet-based algorithms for the processing of biomedical
images; in particular, the detection of anatomical
structures. Application areas include magnetic resonance imaging and
ultrasound.
The study requirements are a masters degree in electrical engineering,
computer science, or applied mathematics. Previous experience in
image/signal processing and wavelets is highly desirable. Applicants
should have a strong theoretical background and a desire to use
mathematical tools in their research. A good knowledge of C, IDL,
Matlab, and Mathematica is also required.
The initial appointment is for one year, renewable in case of admission
to the Ph.D. program. Applications (with detailed CV, summary of grades,
and three reference names) should be sent to:
Prof. Michael Unser
Swiss Federal Institute of Technology (Lausanne)
EPFL/DMT/IOA/Biomedical Imaging Group
BM 4.127
CH-1015 Lausanne
Switzerland
+41(21)693.51.85 (phone)
+41(21)693.37.01 (fax)
URL : http://bigwww.epfl.ch/
Date: Fri, 11 Dec 1998 14:19:36 +0200
From: antoine@fyma.ucl.ac.be (Jean-Pierre Antoine)
Subject: #17 Job: Postdoctoral Position at Univ. Cath. Louvain in Belgium
Job opening: POSDOCTORAL POSITION AVAILABLE
Univ. Cath. Louvain (Belgium)
In the framework of a European Community TMR program, entitled
"Advanced Signal Processing for Medical Resonance Imaging and
Spectroscopy", a postdoctoral positions is available at Universit\'e
Catholique de Louvain, Louvain-la-Neuve (Belgium). Other institutions
involved are in Delft, Leuven, Lyon, Leipzig, Xanthi and Athens.
The general theme is MRS/MRI signal processing, in particular rapid
analysis and reconstruction of signals/images, starting from raw,
undersampled, data in k-space. The particular program to be realized
in Louvain-la-Neuve consists in the use of wavelet methods for the
analysis of data taken in non-Cartesian geometry (gridding), for image
reconstruction, image segmentation, contrast enhancement in images,
and also peak suppression and quantitation problems in NMR spectra.
Programming is done in C, C++ and Matlab. The work will be performed
in a research team based in the Institute of Theoretical Physics, in
close collaboration with the Laboratory for Chemical Physics and
Crystallography (Prof. J-M. Dereppe),where data are taken.
Applicants must be citizens from one of the 15 EU countries (except
Belgium), or Iceland, Israel, Liechtenstein or Norway, no more than 35
years old. They should hold a recent Ph.D. in physics, mathematics or
electrical engineering and be fluent in programming. They should have
an interest in MRS/MRI signal processing and a basic knowledge in
wavelet techniques. Travel to other partner institutions is expected.
Period: one year, renewable, starting anytime after January 1, 1999,
but the sooner the better. Program runs from 1.1.98 to 31.12.2000
Please send a CV with publication list, two references and a summary
of research interests and experience to:
Prof. J-P. Antoine
Inst. Theor. Physics, UCL
2, chemin du Cyclotron
B-1348 Louvain-la-Neuve (Belgium)
Tel. + 32 10 47 32 83; fax + 32 10 47 24 14; E-mail: antoine@fyma.ucl.ac.be
From: Thomas Hogan <math.ohio-state.edu!hogan>
Subject: Contents: J. Approx. Th. - nov98 TOC
To: math.ohio-state.edu!enyem4
Date: Thu, 19 Nov 1998 14:55:29 -0500 (EST)
Content-Type: text/plain; charset=us-ascii
Table of Contents: J. Approx. Theory, Volume 95, Number 2, November 1998
In Memoriam
T. S. Chihara and M. E. H. Ismail
In Memoriam Waleed Al-Salam (July 15, 1926--April 13, 1996)
153--160
Regular Articles
Semyon Rafalson
Some sharp inequalities for algebraic polynomials
161--177
G. Loh\"ofer
Inequalities for the associated Legendre functions
178--193
Rong-Qing Jia
Stability of the shifts of a finite number of functions
194--202
Philippe Guillaume, Alain Huard, and Vincent Robin
Generalized multivariate Pad\'e approximants
203--214
Y. Makovoz
Uniform approximation by neural networks
215--228
Leonid Golinskii
Akhiezer's orthogonal polynomials and Bernstein-Szeg\H{o} method for a
circular arc
229--263
Zuhua Luo and Jeremy Levesley
Error estimates and convergence rates for variational Hermite
interpolation
264--279
Tom Lyche, Larry L. Schumaker, and Sonya Stanley
Quasi-interpolants based on trigonometric splines
280--309
J. V. Stokman and T. H. Koornwinder
On some limit cases of Askey-Wilson polynomials
310--330
Erratum
Volume {\bf 90}, Number 1 (1997), in Article No. AT973065, ``Umbrellas
and polytopal approximation of the Euclidean ball'' by Y. Gordon, S.
Reisner, and C. Sch\"utt, pages 9--22
331
Date: Wed, 25 Nov 1998 23:48:24 +0100
From: Baltzer Science <mailer@ns.baltzer.nl>
Subject: #18 Contents: Advances in Computational Mathematics 9-3,4
Advances in Computational Mathematics 9 (1998) 3,4
Ming-Jun Lai and Larry L. Schumaker
On the approximation power of bivariate splines 251-279
M.J. Hagger
Automatic domain decomposition on unstructured grids (DOUG) 281-310
Gero Nießen
An explicit norm representation for the analysis of multilevel methods
311-335
Gabriele Steidl
A note on fast Fourier transforms for nonequispaced grids 337-352
Michael S. Floater and J.M. Peña
Tensor-product monotonicity preservation 353-362
Hideaki Kaneko, Richard D. Noren and Peter A. Padilla
Singularity preserving Galerkin method for Hammerstein equations with
logarithmic kernel 363-376
Hartmut Prautzsch
Smoothness of subdivision surfaces at extraordinary points 377-389
S.M. Kirkup
Fortran codes for computing the discrete Helmholtz integral operators
391-40
Date: Fri, 11 Dec 1998 01:50:47 +0100
From: Baltzer Science <mailer@ns.baltzer.nl>
Subject: #19 Contents: Numerical Algorithms 18-2
Numerical Algorithms 18 (1998) 2
Avram Sidi and Yair Shapira
Upper bounds for convergence rates of acceleration methods with initial
iterations 113-132
Prabir Daripa and Daoud Mashat
Singular integral transforms and fast numerical algorithms 133-157
Prabir Daripa and Daoud Mashat
An efficient and novel numerical method for quasiconformal mappings of
doubly connected domains 159-175
Ubaldo M. García-Palomares and Francisco J. González-Castaño
Incomplete projection algorithms for solving the convex feasibility problem
177-193
Ilan Bar-On and Marcin Paprzycki
High performance solution of the complex symmetric eigenproblem 195-2
Date: Mon, 16 Nov 1998 10:32:41 -0800
From: Robert Lang <rlang@kakadu.engr.sgi.com>
Subject: #20 Answer: VLSI architectures for DWT (WD 7.11 #22)
> Date: Tue, 10 Nov 1998 15:23:03 +0000 (IST)
> From: Jitendra Ajmera <ajmra5ue@ccs.iitb.ernet.in>
> Subject: #22 Question: VLSI architectures for DWT implementation
>
> Dear Wavelet Digest,
>
> I am a final year student of Electrical Engg. at Indian
> Institute of Technology, Bombay. Currently I am doing a project on
> `VLSI implementation of a DSP chip' and this DSP chip computes Wavelet
> transform. Although I have got quite a lot of references regarding
> wavelets, I haven't able been able to find any on the implementation
> part. Could any of you please help me out?
> My e-mail address: ajmra5ue@ccs.iitb.ernet.in
> Thanx,
> Jitendra Ajmera.
Jitendra,
It's a little old now since I've not been working in the wavelet arena
since completing my PhD in 1996, but my thesis was on VLSI
implementations of wavelet transforms. Although the primary thesis
may not be of use to you (as it focused on interconnection networks
for large-scale, fine-grained SIMD implementations), there is a
literature survey in chapter 2 which discusses the implementations I
was aware of up to 1996 and references the relevant papers.
Note also that my work focuses on the discrete wavelet transform.
My thesis can be found online at:
http://reality.sgi.com/rlang/papers.html
Feel free to contact me for more information.
Regards,
Robert.
Robert Lang, Ph.D. Tel: 650 933 6607 //rlang@sgi.com
Visual Systems Group Fax: 650 932 6607 http://reality.sgi.com/rlang
Silicon Graphics Pgr: 650 317 0796 //rlang_p@pager.sgi.com
Date: Wed, 9 Dec 1998 10:38:05 +0800
From: Liu Bao <matliub@leonis.nus.edu.sg>
Subject: #21 Answer: Vibration and fault diagnosis (WD 7.9 #19, WD 7.9 #21, WD 7.8 #26)
In recent years, we have done some work in the area of machinery
vibration analysis and fault diagnosis using wavelets. Hope our work
is helpful to answer the related questions: WD 7.9 #19, WD 7.9 #21 and
WD 7.8 #26.
For vibration analysis, we applied wavelet techniques (DWT, wavelet
packets, matching pursuit, CWT, etc.) to enhance the nonstationary
features such as spikes and transients of vibration signals generated
in rolling-element bearings and gears. It is shown that in many cases
wavelet analysis is more effective than traditional techniques in the
fault detection of these mechanical systems.
Our research focused more on the feature extraction and classification
of nonstationary machinery signals for automated fault diagnosis and
for dealing with some complicated fault diagnosis problems. Several
wavelet-based methods have been developed by taking advantage of the
principle of information theory, the structure of neural networks, and
the search strategy of genetic algorithms. The basic idea behind these
methods is to select from a redundant wavelet dictionary or the domain
of CWT a set of wavelets that are sensitive to machinery condition
variation to serve as a feature extractor. These methods have been
successfully applied to the detection of some malfunctions of a
reciprocating engine, such as wear of cylinder liner and piston ring,
injector clogging, injection timing advance, etc., using vibration
signals. It was found that the proposed methods perform much better
than principal component analysis, linear discriminant analysis and
some other traditional feature extraction techniques in dealing with
nonstationary machinery signals. A point worthy of noting is that
these methods are also applicable to other areas which have a similar
nature to machinery fault diagnosis.
We have some papers and preprints. Interested please contact
Bao Liu
lb@wavelets.math.nus.edu.sg
Wavelets Strategic Research Programme
Department of Mathematics
National University of Singapore
Date: Sun, 15 Nov 1998 10:18:29 -0500
From: kmatta@ebmail.gdeb.com
Subject: #22 Question: Wavelets in Vibration Transients
I am looking for information on using wavelets to analyze vibration
transients. I would be grateful if someone could supply me with references
or sources of information.
Regards,
Kamal Matta
Electric Boat Corp.
General Dynamics
Groton, Connecticut, USA
kmatta@ebmail.gdeb.com
Date: Sun, 15 Nov 1998 15:04:25 -0600
From: "Arthur A. Petrosian" <neuaap@ttuhsc.edu>
Subject: #23 Question: Decomposing frequency bands
I have the following simple question regarding separation of frequency
bands in wavelet decomposition levels.
Suppose you have a segment of a signal with frequency band 0-128Hz
sampled at 256Hz. You decompose this segment into wavelet subbands
using some Daubechies filter. Then on the first level you will have
separation of your signal's frequency band into 0-64Hz in low-pass
subband and 64-128Hz in high-pass subband not counting the frequency
overlap which depends on the length of the wavelet filter. Then on the
second level you will have 0-32Hz in low-pass and 32-64Hz in high-pass
subbands, etc.
Is this technically correct? Any related references?
Thanks in advance,
Art Petrosian
neuaap@ttuhsc.edu
Date: Wed, 18 Nov 1998 17:12:22 +0300
From: Kutsev Sergey <ksv@chs.ru>
Subject: #24 Question: Looking for C,C++ code for wavelet analysis
I've plundered all wavelet sites looking for C,C++ code for wavelet
analysis. But it was in vain. Some links are
broken(http://www-sim.int-evry.fr/~bourges/WaveLib.html), others have
only executable. Could anybody provide me with information, where I
can get it?
Kutsev Sergey
Date: Thu, 19 Nov 1998 14:54:55 -0700
From: cemami@xn-tech.com
Subject: #25 Question: Continuous wavelet transform
Dear inhabitants of the Wavelet world, I was wondering if anyone has
done "real life" DSP applications using continuous wavelet transform
(preferably using Matlab!)! What is the frequency range of each SCALE
level of the continuous wavelet decomposition?
Thank you very much in advance,
Cyrus Emami
Date: Fri, 20 Nov 1998 10:17:39 -0500
From: mdmoore@ucsd.edumdmoore
Subject: #26 Question: Cont. wavelet transform coefficients to % variance
Hi,
I posed this question to the Matlab people, and they referred me to
you folks--
>>Is there a simple relationship between the units of the continuous wavelet
>>transform coefficients and the variance of the input signal? In other
>>words, I need to be able to convert the coefs to % variance of the original
>>signal-- such that the sum of the coefs across a range of wavelet scales
>>would be equal to the variance passed by a FIR filter with a passband of
>>equivalent width. -Michael Moore
>
If you could point me in the right direction-- the algorithm or some
Matlab5 code that does this, I 'd really appreciate it. The time
series I work with are series of geochemical measurements, and our
confidence in the wavelet results depends heavily on knowing that the
variance represented is large and well above our measurement error.
The Matlab toolkits don't take this need into account.
Thank you and best wishes,
- - -Michael Moore
Dr. Michael D. Moore
Postgraduate Research Geologist
Geosciences Research Division
Scripps Institution of Oceanography, La Jolla CA 92093-0220 USA
Work: 619-534-5911 Home: 619-453-9725 Fax: 619-534-9784
Date: Mon, 23 Nov 1998 10:40:06 -0500 (EST)
From: Yiguo Sun <yiguosun@chass.utoronto.ca>
Subject: #27 Question: Non parametric model specification
Hi,
I am doing non-parametric model specification test using the wavelet
estimates and trying to construct Daubechies' wavelet orthogonal
series by myself, but it took me a long time to run the program, so
finally i gave up. Could you please tell me where i can get the related
software? It will save me a great deal of time.
Thank you very very much.
Sincerely,
Yiguo Sun
Department of Economics TEL: (O) (416)978-4964
University of Toronto
150 St. George Street office : SB05B
Toronto, ON, M5S 1A1 Fax: (416)978-6713
Canada
Date: Tue, 24 Nov 1998 17:06:29 +0530 (IST)
From: Kiran T <kgowda@sasi.com>
Subject: #28 Question: Wavelet basics
Hi,
I am an electrical engineering student with specialization in DSP.
Off late I have developed a keen interest in the area of Wavelets. I
would be glad if anybody can help me in gaining knowledge of wavelet
basics.It would be very helpful if anybody can send a hardcopy of any
material on wavelets to the below address.Any sort of help is welcome.
Thanx in advance.
E-mail:kirant@poboxes.com
Kiran.T
#169,7th 'D' Main
4th Stage,3rd Block
Basaveshwarnagar
Bangalore
pin code-560 079
Karnataka
India
Date: Wed, 25 Nov 1998 13:35:56 +0100
From: =?ISO-8859-1?Q?Pep_Pi=F1ol?= <jpinol@cc.uab.es>
Subject: #29 Question: Wavelets in ecology
Hi!
I am a newcomer in the wavelets field. I would really appreciate if any of
you can give me any reference regarding the application of wavelets to
ecology. Thanks.
J. Piñol
Josep Piñol
CREAF-Unitat d'Ecologia
Facultat de Ciencies
Universitat Autonoma de Barcelona
08193 Bellaterra
Spain
Tel: 34 93 5812383 -- 34 93 581312
Fax: 34 3 5811312
e-mail: jpinol@uab.es
Date: Fri, 27 Nov 1998 12:18:09 +0100
From: LABAT David <labat@imft.fr>
Subject: #30 Question: Wavelets in hydrology
Hi, I am an engineer PhD student at the Institute of Fluid Mechanics
of Toulouse (France) working on Hydrology and Hydrogeology. My
special field of interest is the hydraulic modelisation of very
heterogeneous watersheeds (karstic systems) from a systemic point of
view. So, we study relations between two stochastic processes
(rainfall rate - input and runoff - output). Models are based on
convolution and we try to identify kernels of increasing orders
introducing Volterra expansions. Since those processes are
multi-scales, we would like to use Wavelet transforms. So, I wonder
whether any of you had experiences on kernel identifications thanks
wavelets or know others applications of wavelets in hydrological
modelisation.
e mail : labat@imft.fr
Thank you very much in advance.
David Labat
David LABAT, Institut de Mécanique des Fluides, UMR CNRS/INP-UPS 5502
Institut National Polytechnique / ENSEEIHT Tel: (33) 05 61 28 58 73
Allee du Professeur Camille SOULA Fax: (33) 05 61 28 58 99
31400 TOULOUSE, FRANCE e-mail: labat@imft.fr
Date: Sun, 29 Nov 1998 17:05:53 -0700
From: "Arun K. Tangirala" <arun.tangi@ualberta.ca>
Subject: #31 Question: Wavelets for control of chemical engineering systems
Hi,
I have been trying hard to understand how wavelets can be used for
multirate chemical engineering systems. Is it possible that if I have
a model based on input/output data at a certain rate Ts, then can I
use wavelets to estimate the intersample outputs given a faster rate
input ?. (multirate system). Or in other words, can I use wavelets for
estimation of intersample outputs given data at a certain rate ?. If
so, how ?.
Thank you very much for your guidance.
regards
Arun,
CPC Group, U of A, Edmonton.
http://www.ualberta.ca/~akt1
<webmaster@wavelet.org>