Aino Ropponen, Risto Ritala
OPTIMAL MEASUREMENT SCHEDULING AND DESIGN WITH DYNAMIC PROGRAMMING

We shall present approaches to optimize the design of a measurement system and schedule dynamically a versatile measurement resource. The analysis is based on expressing the system management task as a dynamic programming problem in which the system state is partially observable. We shall review the well-known linearquadratic-Gaussian case, and discuss and give examples solutions of discrete state systems. Furthermore, we discuss on approximate dynamic programming methods to solve such problems in practice.

Komyo Kariya, Shigeru Takayama
“MEASUREMENT SCIENCE” AND EDUCATION TO CULTIVATE SCIENTIFIC BASIC KNOWLEDGE OF PEOPLE

We had already expressed the importance of “Measurement Science” .on the education to rise up the scientific sense for the people live in the society carried on natural science, technique and information. It is the liberty to apply how “Science” to the Education, an answer is in this paper. And we had reported the contents by some papers. This paper suggests the usefulness of application of “Measurement Science” for the stage 1 and stage 2 education (Compulsory Education) and the stage 3 education (Semi-Compulsory Education), when classify National Education in 5 stages.

V.N. Yakimov, G.N. Dyakonov, E.G. Udaltsova, V.I .Batishchev, A.V. Mashkov
A METHOD OF DISTRIBUTED TEACHING SYSTEM REALIZATION BASED ON FUZZY MEASUREMENT OF USER'S FEATURES

The purpose of the paper is to apply the main ideas and methods of soft (fuzzy) measurements to the problem of network learning. Teaching system is considered to be really distributed. As secondary effect of its content (in some point of view it may be called primary) we suppose a method to increase an effectiveness of the WWW.

Lukáš Kupka, Emmanuel Simeu, Haralampos-G. Stratigopoulos, Libor Rufer, Salvador Mir, Olga Tumová
SIGNATURE ANALYSIS FOR MEMS PSEUDORANDOM TESTING USING NEURAL NETWORKS

The aim of this work is to develop a lowoverhead, low-cost built-in test for Micro Electro Mechanical Systems (MEMS). The proposed method relies on processing the Impulse Response (IR) through trained neural networks, in order to predict a set of MEMS performances, which are otherwise very expensive to measure using the conventional test approach. The use of neural networks allows us to employ a low-dimensional IR signature, which results in a compact built-in test. A MEMS structure combining electro-thermal excitation and piezoresistive sensing was chosen as our case study. A behavioral model of this structure was built using Matlab for the purpose of the experiment. The results demonstrate that the neural network predictions are in excellent agreement with the simulation results of the behavioral model.

Zbigniew Czaja
TESTING OF ANALOG PARTS OF ELECTRONIC EMBEDDED SYSTEMS WITH LIMITED ACCESS TO INTERNAL NODES

A new class of multi-port methods based on extension of input-output two-port methods of soft fault diagnosis of passive elements in analog circuits is presented. It uses accessible internal nodes of the tested analog circuit for additional measurements of circuit time responses. Thanks to this, the fault resolution increases, that is we obtain better fault localization coverage.