---

PRELIMINARY PROGRAM

---

TUESDAY AFTERNOON

TA S-1: Array Signal Processing

Chair: J.F. Bohme, Ruhr Univ., Germany

Robust Adaptive Beamforming: Experimental Results

Modified Subspace Smoothing Technique for Multipath Direction Finding

Self Calibration in Presence of correlated Discrete Sources

On forward -backward mode for direction of arrival estimation

MVDR beamforming with inverse updating

High Resolution Array Processing for non circular signals

Spatio -Temporal Coding for Radar Array Processing

A method of direction finding operating on a heterogeneous array

Adaptive Separation of Sources and Estimation of their Directions of Arrival

Maximum Likelihood Time -of -Arrival Estimation using Antenna Arrays:
Application to Global Navigation Satellite Systems

Simultaneously Estimating Frequency and direction of Coherent Signal
via an Array Triplet in Motion

TA SS-2: Video Compression and Communications

Organizer: A. Kastaggelos

Chair: A. Kastaggelos, Northwestern Univ., U.S.A.

Error Resilience and Concealment in Video Coding: (Invited Lecture)

Networking Technologies for Future Video Communications

Multiple Description Coding Using Non -Hierarchical Signal Decomposition

Rate control for delay -constrained video transmission

Compression and Packetization for MPEG -2/4 Video and Audio

MPEG -4 over Wireless Network

Optimization of Wireframe Model Adaptation and Motion Estimation in a Rate -Distortion Framework

High -Order Motion Compensation for Low Bit -Rate Video

A Contribution to Video Coding Using Layers

TA S-3a: Multirate Systems

Chair: P.P. Vaidyanathan, California Inst. of Technology, U.S.A.

Results on Cyclic Signal Processing Systems

Non -uniformly subsampled filter banks

A Design Method of 2 -D Axial -Symmetric Paraunitary Filter Banks with a Lattice Structure

Versatile Building Blocks for Multirate Processing of Bandpass Signals

Fast filter bank synthesis by periodical anamorphosis

TA S-3b: Filter Design

Chair: W. Mecklenbrauker, TU Wien, Austria

Constrained Least Squares Design of FIR Filters Using Iterative Reweighted Least Squares

Design of polynomial interpolation filters based on Taylor series

A Design Method for Half -Band Filters

Design of multiplierless elliptic IIR halfband filters and Hilbert transformers

New Realizations of Second -Order All -Pass Transfer Functions

TA S-4a: Face Recognition

Chair: J. Kittler, Univ. of Surrey, U.K.

Linear projection algorithms and morphological dynamic link architecture for frontal face verification

Constantine Kotropoulos and Ioannis Pitas

Nonlinear scale decomposition based features for visual speech recognition

Wire Frame Fitting For Automatic Tracking In Model -Based Video Coding

Region -based segmentation and tracking of human faces

Facial Feature Extraction in Frontal Views Using Biometric Analogies

TA S-4b: Image Restoration

Chair: J. Biemond, Delft Univ., Netherlands

The Dual -Tree Complex Wavelet Transform: A New Efficient Tool for Image Restoration
and Enhancement

Spectral Ampitude Estimation -based X -Ray Image restoration: an extension
of a speech enhancement approach

Bayesian deconvolution of Poissonian point sources

Modelling PSF of Scanning Electron Microscopes for image Restoration

Iterative Blind Image Restoration Using Local Constraints

TA S-5: Speech Recognition

Chair: P. Noll, TU of Berlin, Germany

On the Integration of Dialect and Speaker Adaptation in a Multi -Dialect Speech Recognition System

Combining Bayesian Learning and Vector Field Smoothing for On -Line
Incremental Speaker Adaptation

Identification of Spoken European Languages

A Bayesian Triphone Model with Parameter Tying

Spectral Subtraction and Missing Feature Modeling for Speaker Verification

Blind Equalization in the Cepstral Domain for Robust Telephone Based Speech Recognition

Robust Speech Recognition Algorithms in A Car Noise Environment

Frame Pruning for Speaker Recognition

Integration of Parsing and Incremental Speech Recognition

Optimizing Hidden Markov Models for Chinese An -set Syllables

TA P-6: Pattern Recognition and Neural Networks

Chair: A.N. Diamantaras, Univ. of Thessaloniki, Greece

A unified approach to laterally -connected neural nets

A Class of Fast Complex Domain Neural Networks for Signal Processing Applications

Efficient Layer -Wise Learning of a Feedforward Neural Networks Using the Backpropagation

Recurrent Neural Networks and Filters Adaptation with Stability Control

A Nonexclusive Classification System Based on Co -operative Fuzzy Clustering

Fuzzy Logic control Based on Weighted Rules using Neural Network

The edge detection from images corrupted by mixed noise using fuzzy rules

A Combination of Classical and Fuzzy Classification Techniques on a Self Organized Memories
(SOM). -Type Neural Network Computational Platform

A New Decision Criterion for Feature Selection. Application to the Classification
of Non Destructive Testing Signatures

A Quantitative Matching Technique Based on Eigenvector Approach

Direct Evaluation of Frame -Based Nonstationary Pattern Recognition Methods by
Using Bhattacharyya Distance

Handwritten Farsi Character Recognition Using Evolutionary Fuzzy Clustering

Hidden Markov Model with Nonstationary States

Experimental Car Plate Recognition System by Neural Netwrorks and Image Processing

Parameter Estimation of Conics: Application to handwritten digits

Multisensor Object Identification on Airports using Autoassociative Neural Networks

Karin Haese

Neural Network -Based Event Detection for Surveillance Applications

Noise Elimination in Approximation and Time Series Prediction with Hinging Hyperplanes

Multiscale versus Multiresolution Analysis for Multisensor Image Fusion

Randomized Regression in Differentiators

TA P-7: Architectures I

Chair: G. Glentis, Univ. of Athens, Greece

A Study on Discrete Wavelet Transform Implementation for a High Level Synthesis Tool

A 500 MHz 2d -DWT VLSI Processor

Improving Performance vs. Silicon Size Tradeoffs Using Coprocessors.

Single Chip DSP Array Processor: 100 Million+ Transistors with Multithreading Approach

Asynchronous Timing Model for High -Level Synthesis of DSP Applications

DSP Architecture for real time digital video converter

A PRML Equalizer for Hard Disk Drives with Low Sensitivity to Sampling Phase Variation

A Recursive Algorithm for the Generation of Space -Filling Curves

Automated Parallel -Pipeline Structure of FFT Hardware Design for Real -Time
Multidimensional Signal Processing

Parallel Implementation of a Face Location Algorithm Based on the Hough Transformation

GSLC Architecture for sequence detectors using spatial diversity

Miguel A. Lagunas, Ana I. Perez - Neira and J. Vidal

A pipelined architecture for DLMS algorithm considering both hardware complexity
and output latency

A G -transform based systolic array for Least Squares Problems

VLSI Implementation and Complexity Comparison of Residue Generators Modulo 3

Adaptative filters implementation performances under power dissipation constraint

Behavioral Synthesis of Digital Filters Using Attribute Grammars

The Impact of Data Characteristics on Hardware Selection for Low -Power DSP