PROGRAM

4th WSEAS / IASME International Conference on:
SYSTEM SCIENCE and SIMULATION in ENGINEERING  (ICOSSSE 2005)

Puerto De La Cruz, Tenerife, Canary Islands, Spain
December 16-18, 2005

Friday, December 16, 2005

ROOM C’: 

Plenary Lecture I

 

New Approaches to Information Aggregation
in Intelligent Systems

Prof. Dr. Imre J. Rudas, Rector

Budapest Tech Polytechnical Institution

Bécsi út 96/B, H-1034
Budapest, Hungary

[email protected]

Abstract:- In intelligent systems like fuzzy control systems, decision support and expert systems, where fuzzy rule base and reasoning mechanism

are applied, one of the key issues is the problem of aggregation of information represented by membership functions. Since the pioneering work

of Prof. Lotfi Zadeh dated to 1965 a great number of fuzzy connectives, aggregation operators have been introduced. The results of continuous

research in this area have proved that the generally used t-norms, t-conorms do not follow always the real phenomena. The requirement to develop

more sophisticated intelligent systems demands to find new operators. This paper summarizes the research results of the authors that have been

carried out in recent years on generalization of conventional operators.

ROOM A’:  

Plenary Lecture II

Statistical Learning for Dependence Mining and 

Problem Solving: Fundamentals, Challenges, and a Unified Theory

Professor LEI XU,  IEEE Fellow,  IAPR Fellow,
Chair Professor of 
Chinese University of Hong Kong
Hong Kong

Abstract: An intelligent system is featured by its abilities of discovering knowledge and skills of problem solvinge, both of which are learned from data. Due to observation noise  and a finite size of samples,  the learning task faces two challenges. One is finding  mathematical representations  to  suit dependence structures underlying the world. The other is getting a theory to guide learning  such  that only those  reliable dependence structures are learned into mathematical representations with an appropriate complexity  that matches the size of samples.  This talk consists of two parts. The first provides a  summary of  existing major mathematical representations for dependence structures and typical learning theories, as well as applications, from a unified perspective under the name of Bayesian Ying Yang learning.   The second part briefly introduces the main results by the speaker  for solving several typical problems in statistical learning and pattern recognition.

ROOM C’: 

Plenary Lecture III

A General Concept for the Modelling and
Simulation of MIMO Mobile Fading Channels

Prof. Dr.-Ing. HABIL. M. PÄTZOLD
Mobile Communications Group,
Faculty of Engineering and Science
Agder University College, Grooseveien 36,
NO-4876 Grimstad
Norway

 

Abstract:-For the development, analysis, and parameter optimization of cellular mobile systems, a profound knowledge of the mobile fading channel is indispensable. With regard to the development of future mobile communication systems employing multiple-input multiple-output (MIMO) techniques, there is currently a strong need for new channel models the statistical properties of which are sufficiently close to real-world channels.
In this plenary lecture, a general concept will be presented for the development and simulation of realistic mobile fading channels models. For illustrative purposes, a basic concept will first be introduced and exemplarily applied to design classical Rayleigh and Rice fading channel models often used to study the performance of single-input single-output systems. It will then be shown how the basic concept can be extended to enable the modelling of spatially correlated fading channels. The usefulness of the generalized concept will be demonstrated by hand of elaborated space-time channels proposed for MIMO mobile-to-mobile systems, where it is supposed that both the transmitter and the receiver are moving. For such so-called MIMO mobile-to-mobile channels, we will first introduce a proper geometrical model. Starting from the geometrical model, it will be shown how a non-realizable reference model and an efficient simulation model can be obtained. An overview of commonly used parameter computation methods will be given and their advantages and disadvantages will be discussed. For the important case of isotropic scattering, a closed-form solution of the parameter computation problem will be presented to enable the fitting of the statistical properties of the developed MIMO channel simulator to those of the reference model with arbitrarily precision. The lecture closes with some practice-relevant examples of simulating MIMO mobile-to-mobile fading channels.

ROOM D’: 

SESSION: Intelligent Systems

Chair:  Zeljko Panian, Alexandrou Onea, Kresimir Fertalj

ROOM D’: 

SESSION: Control, Optimization and Optical Engineering

Chair:  Alexandrou Onea, Zeljko Panian, Michel Gratton

Saturday, December 17, 2005

ROOM D’: 

SESSION: Materials, Electronic Circuits and Simulation

Chair:  Ryszard S. Choras, Domenico Guida, Jana Kleckova
 

Sunday, December 18, 2005

ROOM D’: 

SESSION: Medical, Electronics and Communications

Chair:  Dalibor Biolek, Viera Biolkova, Wataru Kase

ROOM B’: 

SESSION: Knowledge management and techniques

Chair:  Dalibor Biolek, Viera Biolkova, Kresimir Fertalj