[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 7 (August 3, 99)




Today's Editor: Wim Sweldens
                Lucent Technologies, Bell Laboratories
                wim@bell-labs.com



Today's Topics:

     1: Book:     Wavelets in Physics
     2: Preprint: Compressing as well as the best tiling of an image
     3: Preprint: Analysis of Bivariate Nonstationary Time Series
     4: Preprint: Differential Unitary Space-Time Modulation
     5: Preprint: Smooth Wavelet Tight Frames with Limited Redundancy
     6: Software: Certified WSQ freeware
     7: Software: WAILI and WAILI.xl -- Wavelets with Integer Lifting
     8: Meeting:  Workshop on Adaptive Methods and Error Estimators
     9: Tutorial: Wavelet, lifting & ezw tutorial
    10: Job:      PhD Position Available in UK
    11: Job:      Post-Doc in Neuroimaging Statistics at Univ. Pennsylvania
    12: Job:      PhD post at King's College London
    13: Contents: J. Approx. Th. - June/July
    14: Answer:   Binary Wavelets (WD 8.6 #15)
    15: Answer:   Binary Wavelets (WD 8.6 #15)
    16: Answer:   Wavelets and fault detection (WD 8.6 #11)
    17: Question: Continuous Wavelets in C
    18: Question: Analysing vibration data using wavelets
    19: Question: Scale space and wavelets
    20: Question: Algorithm for the Klauder Wavelet
    21: Question: Using wavelets to extract data from radar returns
    22: Question: Nonlinear wavelet transform
    23: Varia:    A new Machine Learning mailing list
     
Current number of members: 14354

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:

 Aug 18-20: Wavelets and Filter Banks Course, Minn.-St.Paul
 Sep 10-12: Image Proc, Multires, Statistics, GaTech, Atlanta
 Oct 7-10:  60th Birthday of John J. Benedetto, College Park, MD
 Oct 8-12:  Course: IASTED SIP'99 Bahamas Tutorial on Wavelets
 Nov 15-20: Int. Conf. on Wavelet Analysis and its Applications
 Dec 3-4:   Perspectives in Applied Math (iho Gil Strang), MA

--- 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




Date: Tue, 03 Aug 1999 21:45:42 +0200
From: Hans van den Berg <hansos@xs4all.nl>
Subject: #1 Book: Wavelets in Physics

Dear editor,

I herewith submit the announcement of the book "Wavelets in Physics"
of which I am the editor, and which will appear with Cambridge
University Press in two weeks. I hope this can still be entered in the
next issue of the Digest.

Yours sincerely, 

Hans van den Berg (ed.)

--

WAVELETS IN PHYSICS

Edited by J.C. van den Berg
Wageningen Agricultural University

   This book surveys the application of the recently developed
technique of the wavelet transform to a wide range of physical fields,
including astrophysics, turbulence, meteorology, plasma physics,
atomic and solid state physics, multifractals occurring in physics,
biophysics (in medicine and physiology) and mathematical physics.  The
wavelet transform can analyze scale-dependent characteristics of a
signal (or image) locally, unlike the Fourier transform, and more
flexibly than the windowed Fourier transform developed by Gabor 50
years ago.  The continuous wavelet transform is used mostly for
analysis, but the discrete wavelet transform allows very fast
compression and transmission of data and speeds up numerical
calculation, and is applied, for example, in the solution of partial
differential equations in physics.  This book will be of interest to
graduate students and researchers in many fields of physics, and to
applied mathematicians and engineers interested in physical
application.

CONTENTS:
   Preface
0. A guided tour 
   J. C. van den Berg; 
1. Wavelet analysis, a new tool in physics 
   J.-P.    Antoine; 
2. The 2-D wavelet transform, physical applications
   J.-P. Antoine; 
3. Wavelets and astrophysical applications 
   A. Bijaoui; 
4. Turbulence analysis, modelling and computing using wavelets 
   M. Farge, N.K.-R. Kevlahan, V. Perrier and K. Schneider; 
5. Wavelets and detection of coherent structures in  fluid turbulence 
   L. Hudgins and J. H. Kaspersen; 
6. Wavelets, non-linearity and turbulence in fusion       plasmas 
   B. Ph. van Milligen; 
7. Transfers and fluxes of wind kinetic energy between orthogonal
   wavelet components during atmospheric blocking 
   A. Fournier; 
8. Wavelets in atomic physics and in solid state    physics 
   J.-P. Antoine ,Ph. Antoine and B.Piraux; 
9. The thermodynamics of fractals revisited with    wavelets 
   A.Arneodo, and E. Bacry J. F. Muzy; 
10. Wavelets in medicine and physiology
    P. Ch. Ivanov, A. L. Goldberger, S. Havlin, 
    C.-K. Peng, M. G. Rosenblum and H. E. Stanley; 
11. Wavelet dimension and time evolution 
    Ch.-A. Guérin and M. Holschneider.

CONTRIBUTORS: J. C. van den Berg, J.-P. Antoine, A. Bijaoui, M. Farge,
N. K.-R. Kevlahan, V. Perrier, K. Schneider, L. Hudgins,
J. H. Kaspersen, B. Ph. van Milligen, A. Fournier, Ph. Antoine,
B. Piraux, A. Arneodo, E. Bacry, J. F. Muzy, P.Ch. Ivanov,
A. L. Goldberger, S. Havlin, C. -K.  Peng, M. G. Rosenblum,
H.E. Stanley, Ch.-A. Guérin, M. Holschneider.

1999   247 x 174 mm   476pp      
76 line diagrams   42 half-tones   
30 colour plates   3 tables
0 521 59311 5   Hardback                £ 60.00/ $95.00
Publication 19 August 1999

All Cambridge books are available from good booksellers. You can
also order direct from Cambridge University Press - simply phone
using a credit card on (UK) +44 (0)1223 326050 or
fax +44 (0)1223 326111. Or if you prefer, you can order via the
Cambridge University Press Web site: http://www.cup.cam.ac.uk
or via email:  directcustserve@cup.cam.ac.uk. For those based in
North America please call 1-800-872-7423 or visit http://www.cup.org
for details on how to order.

J.C. van den Berg         e-mail:   hansos@xs4all.nl 
Oude Schans 67C
1011 KW Amsterdam
The Netherlands           tel:   + 31 20 6232945



Date: Mon, 5 Jul 1999 15:32:42 +0800 (GMT-8)
From: Lee Wee Sun <leews@comp.nus.edu.sg>
Subject: #2 Preprint: Compressing as well as the best tiling of an image

The following two papers are available from 

http://www.comp.nus.edu.sg/~leews/

Title: Compressing as well as the best tiling of an image
Author: Lee, Wee Sun.

Abstract: We investigate the task of compressing an image using
different probability models for different regions of the image. In
this task, using a larger number of regions would result in better
compression of the coefficients of the image but would also require
more bits for describing the regions and probability models in the
regions. We discuss using quadtrees and windowing for performing the
compression and introduce the class of tilings of an image with a
small number of arbitrarily sized rectangular tiles of probability
models. For an image of size n times n, we give a sequential
probability assignment algorithm with a computational complexity of
O(Nn^3) and a redundancy of O(k log Nn) relative to the class k
rectangular tiling of an image using N probability models. For the
simpler class of tilings using rectangles with width no more than W,
we give an algorithm with redundancy O(k\log Nn) and a computational
complexity of O(WNn^2).  Another interesting class is the class of
tilings using rectangles with widths which are of a power of two. This
class is far more flexible than quadtrees and yet has a competitive
probability assignment algorithm with a redundancy O(k\log Nn) and a
computational complexity of O(Nn^2\log n) which is similar to the
computational complexity of probability assignment using quadtrees. We
also consider progressive transmission of the coefficients of the
image for all the classes.

Title: Windowing versus Best Tiling for Wavelet Image Compression
Author: Lee, Wee Sun.

Abstract: We compare two methods for entropy coding uniformly
quantized wavelet coefficients: windowing, which uses local statistics
to adapt the probability assignment and best tiling, which aims to
compress as well as the best tiling of a subband with rectangular
tiles of probability models.  We find that on some synthetic images,
the best tiling method works considerably better than windowing, but
on a set of natural images, the improvement provided by the best
tiling method over windowing is very small.



Date: Wed, 7 Jul 1999 13:25:50 -0400 (EDT)
From: Hernando Ombao <hombao@umich.edu>
Subject: #3 Preprint: Analysis of Bivariate Nonstationary Time Series

Title:  Automatic Statistical Analysis of Bivariate Nonstatationary 
        Time Series

Authors: Hernando Ombao, Jonathan Raz, Rainer von Sachs and Beth Malow.

Gzipped postscript file can be downloaded from the site below. Look
under the heading "Submitted for Publication" :
http://www.sph.umich.edu/~jonraz/papers.html

Abstract: We propose a new method for analyzing bivariate
non-stationary time series. The proposed method is a statistical
procedure that automatically (i.) segments the time series into
approximately stationary blocks and (ii.) gives the span to be used in
obtaining the smoothed estimates of the time-varying spectra and
coherence. It is based on the SLEX ({\bf S}mooth {\bf L}ocalized
Complex {\bf EX}ponential) transform which forms a library of
orthogonal complex-valued transforms that are simultaneously localized
in time and frequency. We show that the smoothed SLEX periodograms are
consistent estimators, report simulation results and apply the method
to an electroencephalogram (EEG) data set recorded during an epileptic
seizure.
 
Keywords: Non-Stationary time series; Time-varying spectrum and
coherence; Time-frequency analysis; SLEX transform; SLEX periodogram;
Kernel smoothing.

H. Ombao is a student in the Department of Biostatistics at the
University of Michigan. J. Raz is Associate Professor of Biostatistics
at the University of Pennsylvania. R. von Sachs is Associate Professor
at the Catholique Universite de Louvain. B. Malow is a neurologist at
the University of Michigan Medical School.



Date: Tue, 20 Jul 1999 13:51:50 -0400
From: Wim Sweldens <wim@bell-labs.com>
Subject: #4 Preprint: Differential Unitary Space-Time Modulation

Title: Differential Unitary Space-Time Modulation

Authors: Bert Hochwald and Wim Sweldens

Abstract: We present a framework for differential modulation with
multiple antennas across a continuously fading channel, where neither
the transmitter nor the receiver knows the fading coefficients. The
framework can be seen as a natural extension of standard differential
phase shift keying (DPSK) commonly used in single-antenna
unknown-channel systems. We show how our differential framework links
the unknown-channel system with a known-channel system. As a special
case, we introduce a class of diagonal signals where only one antenna
is active at any time, and show how these signals may be used to
achieve full transmitter diversity and low probability of error.

On the web at: /who/wim/papers/dust

Note: This is a follow-up from an earlier paper on 
"Systematic Design of Unitary Space-Time Constellations" where
it is shown how multiple antenna constellations can be built
from tight frames. That paper can be found at 

 /who/wim/papers/constell



Date: Mon, 2 Aug 1999 18:16:06 -0400 (EDT)
From: Ivan Selesnick <selesi@taco.poly.edu>
Subject: #5 Preprint: Smooth Wavelet Tight Frames with Limited Redundancy

Smooth Wavelet Tight Frames with Limited Redundancy

This paper considers the design of wavelet tight frames based on
iterated oversampled filter banks.  The greater design freedom
available makes possible the construction of wavelets with a high
degree of smoothness, in comparison with orthonormal wavelet bases.
Groebner bases are used to obtain the solutions to the nonlinear
design equations.  Following the dual-tree DWT of Kingsbury, one goal
is to achieve near shift-invariance while keeping the
redundancy-factor bounded by 2, instead of allowing it to grow as it
does for the undecimated DWT (which is exactly shift-invariant).  Like
the dual-tree, the overcomplete DWT described in this paper is less
shift-sensitive than an orthonormal wavelet basis.

Ivan Selesnick 
Electrical Engineering
Polytechnic University
Brooklyn, NY

http://taco.poly.edu/selesi/

selesi@taco.poly.edu



Date: Sun, 4 Jul 99 12:07:28 -0500
From: Mladen Victor Wickerhauser <victor@math.wustl.edu>
Subject: #6 Software: Certified WSQ freeware

SOFTWARE: certified executables to perform the FBI's WSQ fingerprint
image compression and decompression algorithm on 6 flavors of UNIX,
Windows 98 and Windows 95 may be downloaded from my home page,

        http://www.math.wustl.edu/~victor/

Supported versions for 16-bit DOS and Windows, derived from the same
sources, can be licensed from Pegasus Imaging Corporation:

        http://www.jpg.com/technology/wsq_fingerpri.html

------
Prof. M. Victor Wickerhauser            Dept. of Math., Box 1146
Tel: +1(314)935-6771                    Washington University
Fax: +1(314)935-6839                    One Brookings Drive
<victor@math.wustl.edu>                 St Louis, MO 63130 USA



Date: Fri, 23 Jul 1999 14:49:07 +0200 (CEST)
From: Geert Uytterhoeven <Geert.Uytterhoeven@cs.kuleuven.ac.be>
Subject: #7 Software: WAILI and WAILI.xl -- Wavelets with Integer Lifting

We would like to announce some changes in the status of the software library
WAILI:

  - An updated version of WAILI became free software and was rereleased under
    the GPL
  - WAILI.xl, a version of WAILI with Extensions for Very Large Images is now
    available

Please visit

    http://www.cs.kuleuven.ac.be/~wavelets/

for more information.

                    WAILI -- Wavelets with Integer Lifting
     _________________________________________________________________

                                 Introduction

   WAILI is a wavelet transform library. It includes some basic image
   processing operations based on the use of wavelets and forms the
   backbone of more complex image processing operations.

                                   Features

   WAILI
     * Uses integer wavelet transforms based on the Lifting Scheme
     * Provides various wavelet transforms of the
       Cohen-Daubechies-Feauveau family of biorthogonal wavelets
     * Provides crop and merge operations on wavelet-transformed images
     * Provides noise reduction based on wavelet thresholding using
       Generalized Cross Validation
     * Provides scaling of images
     * Provides edge enhancement of images
     * Provides also some simple image operations (addition and
       subtraction of images)
     * Allows different image representations (RGB, YUV, Lab, ...)

   WAILI.xl is a version of WAILI with Extensions for Very Large Images.
   Very large images are divided in blocks, which do not necessarily have
   to be present in the computer system's main memory. WAILI.xl is based
   on WAILI and provides most of the operations implemented in WAILI.

                                Implementation

   The software library is written in C++ and extensively uses features
   of the ISO C++ 97 Standard to allow for a cleaner design. The
   development was done under Linux/ia32 2.0.x and 2.2.x, using GNU C++
   2.7.2.x, and later egcs. It was tested under Solaris/SPARC 2.5 and
   2.6, and on various other Linux platforms (m68k, PPC, AXP).

                                 Availability

   WAILI is free software; you can redistribute it and/or modify it under
   the terms of the GNU General Public License as published by the Free
   Software Foundation. It can be downloaded free of charge.

   WAILI.xl is available in source form for research purposes only and
   may not be further distributed. Enquiries about the license conditions
   should be directed to wavelets@cs.kuleuven.ac.be. Alternatively you
   can use the online form.

                                    Authors

     * Geert Uytterhoeven
     * Filip Van Wulpen
     * Maarten Jansen

                               Acknowledgements

   This library was developed within a project on Wavelet Based
   Interactive Video Communication and Image Database Consulting, funded
   by the Flemish Information Technology Action Program (Vlaams
   Actieprogramma Informatietechnologie), project number ITA/950244.

     _________________________________________________________________

   And remember, it's written WAILI, but pronounced WILY, except in
   Dutch, where it's pronounced PIEFPAK.

Greetings,

                                                Geert
--
Geert Uytterhoeven                     Geert.Uytterhoeven@cs.kuleuven.ac.be
Wavelets, Linux/{m68k~Amiga,PPC~CHRP}  http://www.cs.kuleuven.ac.be/~geert/
Department of Computer Science -- Katholieke Universiteit Leuven -- Belgium



Date: Mon, 02 Aug 1999 14:41:18 +0200
From: Angela Kunoth <kunoth@igpm.rwth-aachen.de>
Subject: #8 Meeting: Workshop on Adaptive Methods and Error Estimators

GAMM Workshop Kiel on

          Adaptive Methods - Error Estimators      
              
                 January 21-23, 2000     
 
    Organizers:
      Carsten Carstensen (Kiel)
      Wolfgang Hackbusch (Kiel & Leipzig)
      Angela Kunoth (Aachen)   

    The aim of the workshop is to present and discuss the state
    of the art of error estimators for partial differential
    and integral operators to be used within adaptive strategies
    and related topics.
    While error estimators for finite element methods are established 
    in the literature, adaptively generated wavelet spaces have been 
    introduced very recently.
    Possible topics of the workshop include reliability, efficiency, 
    (Besov) regularity, saturation assumptions, and approximation 
    properties of exact and discrete solutions to elliptic partial 
    differential equations.
    We particulary intend to discuss parallel developments in both 
    the finite element and the wavelet areas.
 
    Topics of interest include, but are not restricted to, the
following:
 
          o  A-posteriori error estimators for partial differential
             operators based on
                o finite elements
                o wavelets
  
          o  A-posteriori error estimators for integral operators based
on
                o finite elements
                o wavelets
  
          o Adaptive boundary element methods
          o Adaptive multigrid methods
   
    Invited Speakers (Confirmed):
      Ronald A. DeVore (South Carolina)
      Willy Doerfler (Freiburg)
      Rolf Rannacher (Heidelberg)

    Everybody interested is warmly invited to attend.
    A second announcement with further information will be distributed 
    in October 1999.

    Location: Mathematisches Seminar und Institut fuer Informatik und 
    Praktische Mathematik, Universitaet Kiel, Germany

    Conference fee: DM 70.- (to be paid after arrival)

    Call for papers: Participants wanting to give a talk (30 min)
    should submit an abstract (10-20 lines) before November 30, 1999.
    Notification of acceptance will be given by December 15, 1999.

    Local organization:  Angela Kunoth, Tel.: ++49-241-80-6578
                                        Fax: ++49-241-8888-317
                         kunoth@igpm.rwth-aachen.de
 
    All correspondence in connection with the workshop, including
    registration and submission of abstracts, is to be made via
    e-mail on address: kunoth@igpm.rwth-aachen.de
  
    Visit the homepage on
http://www.igpm.rwth-aachen.de/kunoth/workshop.html
    for further upcoming information.



Date: Thu, 22 Jul 1999 03:41:31 -0400 (EDT)
From: Clemens Valens <cvalens@yahoo.com>
Subject: #9 Tutorial: Wavelet, lifting & ezw tutorial

To whom it may concern:

Please check out my tutorial on wavelets, lifting and embedded zerotree
wavelet encoding at:

http://perso.wanadoo.fr/polyvalens/clemens/clemens.html

The main features are:
- A clearly written tutorial on basic wavelet theory aimed at
non-mathmaticians and starters;
- A very detailed explanation of the fast lifting wavelet transform and
integer lifting, including a detailed example;
- A tutorial on embedded zerotree wavelet (EZW) encoding, including a
detailed example and sample source code.

I would really appreciate to hear if it is useful to somebody. Please
send all comments to:

c.valens@mindless.com

Best regards,
Clemens VALENS



Date: Mon, 5 Jul 1999 14:15:10 +0100
From: p.addison@napier.ac.uk
Subject: #10 Job: PhD Position Available in UK

     PhD Position Available
     
     A PhD studentship is available concerning the practical
     application of wavelet transform analysis. The focus of the work
     will very much depend upon the interest of the applicant. Current
     research projects within the department include the application
     of wavelet transform analysis to Non-Destructive Testing,
     Turbulent Fluid Flows and Biomedical Signal Processing. It is
     envisaged that the applicant would pursue one of these areas,
     however, a research programme within the applicant's own area of
     expertise would also be considered.
     
     Applicants should hold or expect to gain a good degree in
     Engineering, Science or Applied Mathematics. Experience in
     computer programming including either MATLAB or MATHCAD would be
     very advantageous, though not essential.
     
     The studentship includes a grant and the payment of tuition fees
     for students resident within the European Union. Those outwith
     the EU will be paid a full grant but will have to contribute
     towards the tuition fee.
     
     For more information contact Dr Paul S Addison directly at 
     p.addison@napier.ac.uk
     
     See also:
     
     http://sbe.napier.ac.uk/staff/paddison/wavelet.htm



Date: Wed, 7 Jul 1999 10:43:42 -0400 (EDT)
From: Jonathan Raz <jonraz@umich.edu>
Subject: #11 Job: Post-Doc in Neuroimaging Statistics at Univ. Pennsylvania

Post-Doc in Neuroimaging Statistics

The University of Pennsylvania is seeking new PhDs to fill one to
three post-doctoral position(s). Candidates should have completed or
be nearing completion of a Doctorate in Biostatistics, Statistics,
Electrical Engineering, Bioengineering, Computer Science, Mathematics,
or a related field. The successful candidate will develop new
statistical wavelet methods for analysis of functional brain images
and time series; write methodological and theoretical papers; and
collaborate with top neuroscientists. This is a great opportunity to
gain experience in one of the most exciting and rapidly growing areas
of wavelet applications and statistics. Previous experience with
wavelets, time series, imaging, and/or neuroscience will be helpful.
The post-doctoral appointment can begin any time between September 1,
1999 and September 1, 2000. Send CV and 3 letters of recommendation
to:

Jonathan Raz, Ph.D.
Division of Biostatistics -- CCEB
University of Pennsylvania Medical Center
612 Blockley Hall, 423 Guardian Drive
Philadelphia, PA 19104-6021

or send electronic mail to: 

jraz@cceb.upenn.edu

AA/EOE.



Date: Thu, 08 Jul 1999 11:32:17 +0100
From: Mark Sandler <mark.sandler@kcl.ac.uk>
Subject: #12 Job: PhD post at King's College London

KING'S COLLEGE LONDON
DEPARTMENT OF ELECTRONIC ENGINEERING
POST-GRADUATE RESEARCH IN 
"MUSICAL ANALYSIS AND METADATA BY TIME-FREQUENCY TECHNIQUES"

A 3 year studentship to undertake exciting research in the extraction
of content-related information (metadata) from musical audio signals
has been funded by JISC and is due to start in October 1999.

The studentship is one part of a larger project called OMRAS (Online
Musical Retrieval and Search), worth ~$1M, involving the Departments
of Electronic Engineering, Computer Science and Music at King's
College (KCL) and the Institute for Intelligent Systems of the
University of Massechusetts (UMass). OMRAS is jointly supported by
JISC of the UK and NSF of USA under their overall programme of
research in Digital Libraries.

Working closely alongside computer scientists and musicologists, the
student will examine various time-frequency and time-scale signal
processing techniques (especially Wavelets) for transcribing
unstructured audio (ie digital recordings of instruments) into
structured audio which captures the underlying musical description of
the notes played. Other members of OMRAS will work on pattern matching
techniques, searching large databases for items that closely match the
captured, structured audio. In due course, a post- doctoral researcher
will be appointed to work on other aspects of musical audio
metadata. The student will couple Wavelet theory with Hidden Markov
and Neural Network techniques in order to refine and abstract the
signal representation, making it more suitable for recognition
purposes.

The Department of Electronic Engineering has significant expertise in
Wavelet technology for Audio Compression, especially over Internet,
which will be of benefit to this project.Both areas of research have
great relevance to standards development under MPEG-4 and MPEG-7,
which include structured representations and the extraction of
metadata of musical audio signals. Through extensive links across
Europe and with UMass directly, the student can expect to be offered
the possibility of working in other leading laboratories for periods
of a few months.

To be eligable for this post, you should have or expect shortly to be
awarded a First or Upper Second Class degree (BEng, MEng or non-UK
equivalent) in Electronic Engineering, Computer Science, Physics, or
Music Technology. Interests in Digital Signal Processing and Music are
essential, as is the ability to work well within a team. Interests in
multimedia, pattern recognition, Hidden Markov and Neural Network
techniques and a good background in software development will be
beneficial.

The studentship includes maintenance at EPSRC levels and fees at
home/EU rates.

Applicants should apply in writing to Professor Mark
Sandler,Department of Electronic Engineering, Kings College London,
Strand, LONDON WC2R 2LS, UK including a current CV, the names of two
referees and a covering letter of up to 2 pages explainingyour reasons
for being interested in the post and what you can contribute.  Please
also include a completed KCL Postgraduate Application form, which can
be downloaded via http://www.kcl.ac.uk/depsta/pse/diveng/eleceng/

Closing date for applications is 20 August 1999, though applicants 
are strongly encouraged to respond by 31 July.Envelopes should be 
clearly marked "PhD OMRAS".Interviews are planned for early 
September with a view to starting work soon afterwards.

For an overview of research in Music and Audio Technology 
in the Dept of EE, 
visit http://www.kcl.ac.uk/depsta/pse/diveng/eleceng/audio/

Mark Sandler
Professor of Signal Processing,
Department of Electronic Engineering
King's College London, Strand,
London WC2R 2LS, UK
Tel: +44 (207) 848 2365
Fax: +44 (207) 848 2932

Editor-in-Chief,
Applied Signal Processing
The International Journal of Analog and Digital Signal Processing
http://www.springer.co.uk/comp/journals/0635.html

Also
http://www.kcl.ac.uk/eleceng    for the Dept of Electronic Eng
http://www.kcl.ac.uk/eleceng/audio      for research in Audio & Music Technology



Date: Thu, 15 Jul 1999 17:46:48 -0400 (EDT)
From: Thomas Hogan <hogan@math.ohio-state.edu>
Subject: #13 Contents: J. Approx. Th. - June/July

Table of Contents: J. Approx. Theory, Volume 98, Number 2, June 1999

G. Min
On the denseness of rational systems
197--202

Alexander M. Stokolos and Walter Trebels
On the rate of almost everywhere convergence of certain classical
   integral means
203--222

Chrisina Draganova
Smoothest interpolation in the mean
223--247

Thomas A. Hogan
Stability and independence for multivariate refinable distributions
248--270

Fran\c cois Dubeau and Jean Savoie
On optimal error bounds for derivatives of interpolating splines on a
   uniform partition
271--302

Nikolai Nikolski
Remarks concerning completeness of translates in function spaces
303--315

Marie-Laurence Mazure and Pierre-Jean Laurent
Piecewise smooth spaces in duality: application to blossoming
316--353

B. L. Chalmers, D. Leviatan, and M. P. Prophet
Optimal interpolating spaces preserving shape
354--373

Author index for Volume 98
374

Table of Contents: J. Approx. Theory, Volume 99, Number 1, July 1999

F. Zeilfelder
Strong unicity of best uniform approximations from periodic spline 
   spaces
1--29

G. L\'opez Lagomasino and H. Pijeira Cabrera
Zero location and $n$th root asymptotics of Sobolev orthogonal polynomials
30--43

A. R. Alimov and H. Berens
Examples of Chebyshev sets in matrix spaces
44--53

S. Noschese and L. Pasquini
On the nonnegative solution of a Freud three-term recurrence
54--67

V. E. Maiorov
On best approximation by ridge functions
68--94

Vitaly Maiorov, Ron Meir, and Joel Ratsaby
On the approximation of functional classes equipped with a uniform
   measure using ridge functions
95--111

Luoqing Li
The Peetre K-functional and the Riesz summability operator for the
   Fourier-Legendre expansions
112--121

Yuan Xu
Asymptotics of the Christoffel functions on a simplex in $\BR^d$
122--133

Andr\'as Kro\'o and Szil\'ard R\'ev\'esz
On Bernstein and Markov-type inequalities for multivariate polynomials
   on convex bodies
134--152

V. S. Buyarov, J. S. Dehesa, A. Mart\'\i nez-Finkelshtein,
   and E. B. Saff
Asymptotics of the information entropy for Jacobi and Laguerre
   polynomials with varying weights
153--166

A. B. J. Kuijlaars and W. Van Assche
The asymptotic zero distribution of orthogonal polynomials with varying
   recurrence coefficients
167--197

Qiyu Sun and Zeyin Zhang
A characterization of compactly supported both $m$ and $n$ refinable
   distributions
198--216



Date: Thu, 8 Jul 1999 02:13:23 +0000 (   )
From: Eric Sakk <esakk@ANISE.EE.CORNELL.EDU>
Subject: #14 Answer: Binary Wavelets (WD 8.6 #15)

>From: Louise Thornton <louise@spd.eee.strath.ac.uk>
>Subject: #15 Question: Wavelet transform of binary data
>
>I want to apply the DWT to a binary stream and was wondering if anyone
>had any ideas on how to do so. How can one obtain a frequency-time
>representation of a bitstream?  Any suggestions, explanations and
>sources of information would be greatly appreciated.
>
>Regards,
>Louise Thornton.

Hi,
 Two good starting points are:
1) Swanson, M.D., and Tewfik, A.H., "A Binary Wavelet Decomposition of
Binary Images,", IEEE Transactions on Image Processing, V5, pp1637-1650,
Dec 1996.

2) Caire, G., Grossman, R.L., Poor. H.V., "Wavelet Transforms Associated
with Finite Cyclic Groups," IEEE Transactions on Information Theory, V39,
pp1157-1166, July 1993.

Ref #1 (Swanson and Tewfik) gives an interpretation of the 'frequency'
of a binary stream (handed down from the Hadamard transform) known as
'sequency'. The sequency of a binary stream is related to (depnding on
whose definition you're using) the number of 0-->1 transitions.
'Perfect reconstruction' filters are designed that categorize the
response as either high sequency or low sequency (in analogy to PR
filters with coefficients over a real field).

Ref #2 (Caire et al) provides a clear presentation of wavelet bases
over finite fields. Interestingly enough, an implicit assumption about
the periodicity of wavelet coefficients when analyzing a periodic
signal is not explicitly proved. A similar 'cyclic' assumption about
the signal in question is made, for example, when applying the DFT.
Depending on your level of expertise, a helpful supplement might be
Burrus, C.S., Gopinath, R.A., and Guo, H., "Introduction to Wavelets
and Wavelet Transforms", pp190-194, Prentice-Hall, 1998.

Those pages deal directly with the periodicity issue.

If you want more you can e-mail me directly. I've been up to my ears
lately with binary wavelets, so I might be able to help out depending
on your application.
                        Good Luck
                        Eric Sakk
                        esakk@ee.cornell.edu



Date: Tue, 20 Jul 1999 13:51:50 -0400
From: Wim Sweldens <wim@bell-labs.com>
Subject: #15 Answer: Binary Wavelets (WD 8.6 #15)

Carolyn Johnston has written a paper on the use of the lifting
scheme for binary wavelet transforms. The reference is:

"The lifting scheme and finite precision error free filter banks"
C. Johnston, SPIE Wavelet Applications in Signal and Image
Processing IV, SPIE 2825, pp. 307-314



Date: Thu, 08 Jul 1999 19:17:58 +0400
From: "Gustavo A. Ruiz Robredo" <grr@dyvci.unican.es>
Subject: #16 Answer: Wavelets and fault detection (WD 8.6 #11)

>> Date: Tue, 08 Jun 1999 11:59:40 -0700
>> Subject: #11 Question: Wavelets and fault detection

>>    I wonder if somebody knows where I can find some papers or
>> information related to fault detection using wavelets. I hope you
>> can help me!!!

>> Vicente Alarcon, Imperial College, London, UK

A few years ago, I worked in fault detection and diagnosis in MOS
digital circuits using Haar and Walsh (it's not a wavelet)
transforms. Now, I have not new references about fault detection using
wavelets.  My old publications are:

G. A. Ruiz, A.M. Burón, J.A. Michell.  NMOS & CMOS Physical Faults
Diagnosis by Spectral Techniques: Simulation and Associated Tools.
Proceedings of the IEEE Workshop on Languages for Automation. Vienna,
Austria.  Computer Society Press of the IEEE, pp. 191-194, August
1987.

G. A. Ruiz, J.A. Michell, A.M. Burón Fault Detection and Diagnosis for
MOS Circuits from Haar & Walsh Spectrum Analysis: On the Fault
Coverage of Haar Reduced Analysis. Proceedings of the
3rd. International Workshop On Spectral Techniques, C. Moraga
(Editor), University of Dortmund, pp. 97-106. Dortmund (F.R.G.),
October 1988.

G. A. Ruiz, J.A. Michell, A.M. Burón.  Diagnosis of Stuck-Open Faults
in CMOS by Spectral Techniques with Pseudorandom Excitation.
Proceedings of the Seventh IASTED International Symposium on Applied
Informatics. Grindelwald, Switzerland. Editor: M.H. Hamza, publication
of the IASTED, ACTA PRESS (Anaheim-USA; Calgary-Canada;
Zurich-Switzerland), pp. 89-92 , February 1989.

G. A. Ruiz, J.A. Michell, A.M. Burón.  An Approach to the Spectral
Testing of MOS Sequential Circuits.  Proceedings of the 39th ISMM
International Conference: Mini and Microcomputers and their
Applications. Zurich, Switzerland. Editor: M.H. Hamza, publication of
the International Society for Mini and Microcomputers (ISMM), ACTA
PRESS (Anaheim-USA; Calgary-Canada; Zurich-Switzerland), pp. 109-112,
June 1989.

G. A. Ruiz, J.A. Michell, A.M. Burón.  Automatic Test Vector Set
Generation in the Spectral Testing of Sequential Circuits.
Proceedings of the 40th ISMM International Symposium: Mini and
Microcomputers and their Applications. Lugano, Switzerland. Editor:
M.H. Hamza, publication of the International Society for Mini and
Microcomputers (ISMM), ACTA PRESS , pp.  55-58, June 1990.

G. A. Ruiz, J.A. Michell, A.M. Burón.  Switch-level Fault Detection
and Diagnosis Environment for MOS Digital Circuits using Spectral
Techniques.  IEE-E Computers and Digital Techniques. Vol 139, No 4,
Pag 293-307, July 1992



Date: Fri, 02 Jul 1999 14:17:17 -0700
From: cemami@xn-tech.com
Subject: #17 Question: Continuous Wavelets in C

Dear wavelets enthusiasts,

I know that Matlab has the capability of performing the continuous
wavelet transform. Has anyone seen continuous wavelet transform
routine written in C or C++?  It would be greatly appreciated.

With regards,
Cyrus Emami



Date: Tue, 06 Jul 1999 15:53:48 -0500
From: Bappaditya Banerjee <banerb@CAT.com>
Subject: #18 Question: Analysing vibration data using wavelets

Hi!

I was referred by Mathworks to this group.  I use the matlab Wavelet
Toolbox.

I am new to Wavelets.  I would like to be able to study nonstationary
impact vibration data using wavelets and communicate the analysis
using a "time-frequency" representation rather than the "time-scale"
representation.  I would appreciate any references/your thoughts and
ideas on how to convert "scales" to "frequencies" for communicating
the results of the wavelet analysis of vibration data.

Best wishes,

bappa.

Peoria Proving Grounds
Caterpillar Machine research
banerb@cat.com



Date: Fri, 9 Jul 1999 14:35:30 +0100 (BST)
From: "R. Periasamy" <perias@compsci.bristol.ac.uk>
Subject: #19 Question: Scale space and wavelets

Could you please help me to understand the relationship between scale
space theory and wavelet theory. If there is any introductory article
about the connection between them, it will be a great help.

Regards

RK
perias@cs.bris.ac.uk

[Answer WD 8.8#24]



Date: Mon, 05 Jul 1999 12:35:13 +0800
From: rgullco@ikoda.com.au (Ikoda - Rgullco)
Subject: #20 Question: Algorithm for the Klauder Wavelet

Could it be possible to e-mail me the ALGORITHM OF THE KLAUDER WAVELET
IN THE TIME DOMAIN, or put me in touch with any person who can send it
to me? I feel a bit embarrassed in asking such a simple question, but
the fact is that I have to prepare some synthetic seismograms and this
algorithm cannot be found in any of the standard texts on exploration
geophysics. Thank you very much for your help.

Sincerely yours 
Robert Gullco
Senior Development Geologist

Ikoda Pty Ltd
Level 3 41-43 Ord Street 
West Perth, Western Australia 6005



Date: Mon, 19 Jul 1999 16:11:51 BST
From: "Henry Knowles" <henry_knowles@hotmail.com>
Subject: #21 Question: Using wavelets to extract data from radar returns

Dear All,

For my research project I am using wavelets to extract features from
radar returns, where there are multiple objects in the path.  data
that is to be determined includes are the velocities of the individual
objects, as well as the tumble rate and position.  What is the best
way to start?

One of the problems I have encountered so far is that at high tumble
rates, there is an overlap of the different objects due to the doppler
shift.  Is there any way to resolve the individual objects?

Also, when I take the FFT of the signal, there are no -ve frequency
components present. Does this matter?

Any help will be gratefully received.  Please reply to:

        hk6228@bris.ac.uk
   or:  henry_knowles@hotmail.com

Many thanks
Henry Knowles



Date: Thu, 22 Jul 1999 18:31:42 +0300
From: "Mr. Sword" <sword@nord.vostok.net>
Subject: #22 Question: Nonlinear wavelet transform

Dear All!  Who can recommend me any papers on topic "Nonlinear wavelet
transform for image coding"? I wait for your answer.

Many thanks.

Mike Sabelkin
Kharkov Aerospace University
Ukraine
e-mail: sword@nord.vostok.net



Date: Mon, 5 Jul 1999 23:32:14 -0500
From: Tjen-Sien Lim <limt@istat10.stat.wisc.edu>
Subject: #23 Varia: A new Machine Learning mailing list

The Machine Learning mailing list is an unmoderated mailing list
intended for people in Computer Sciences, Statistics, Mathematics, and
other areas or disciplines with interests in Machine Learning.
Researchers, practitioners, and users of Machine Learning in academia,
industry, and government are encouraged to join the list and
participate.

You're invited to discuss and exchange ideas regarding any aspect of
Machine Learning, e.g., various learning algorithms, data
pre-processing, variable selection mechanism, instance selection, and
applications to real-world problems.

Software, conferences, courses, jobs, and others related announcements
are also welcome.

You can post, read, and reply messages solely on the Web. You can
choose to receive messages as individual emails, daily summaries,
daily full-text digest, or read them on the Web only.

Machine Learning mailing list:
machine-learning@egroups.com

Machine Learning homepage: 
http://www.Recursive-Partitioning.com 

Archives:
http://www.egroups.com/group/machine-learning

To subscribe, send an empty message to
machine-learning-subscribe@egroups.com

To unsubscribe, send an empty message to
machine-learning-unsubscribe@egroups.com

Group Moderator: 
machine-learning-owner@egroups.com

-- 
Tjen-Sien Lim                (608) 262-8181 (Voice)
Dept. of Statistics          (209) 882-7914 (Fax)
Univ. of Wisconsin-Madison   limt@stat.wisc.edu
1210 West Dayton Street      http://www.stat.wisc.edu/~limt
Madison, WI 53706


<webmaster@wavelet.org>