Informatica

New formulae providing exact explicit solutions for a class of scalar delayed systems are given.

Teaching of computer programming by electronic mail has been held in Lithuanian schools since 1992. School students of upper grades (8–12) took part in the experiments. The experiments showed that students cope better with first stages of programming (specifying the problem, etc.) than with the last ones (programming, debugging). The analysis of programming products (made by their peers) by students themselves may be expected to be an important component of learning and teaching.

Adaptive Control Distributed Parameter Systems (ACDPS) with adaptive learning algorithms based on orthogonal neural network methodology are presented in this paper. We discuss a modification of orthogonal least squares learning to find appropriate efficient algorithms for solution of ACDPS problems. A two times problem linked with the real time of plant control dynamic processes and the learning time for adjustment of parameters in adaptive control of unknown distributed systems is discussed.

The performance of a computer network is commonly measured by the maximum minimum time required to move a certain amount of data between any 2 nodes in the network. Due to the advances in technology, certain links in the network may be upgraded, for instance to optical fibre links, so that better performance can be achieved. In this paper, we study the LINK UPGRADE problem for networks. We first show that the LINK UPGRADE problem is NP-complete. We also show that, a closely related problem, the MINIMUM COST LINK UPGRADE problem is NP-complete even if the underlying topology of the network is a linear array. However, for certain classes of networks, the LINK UPGRADE problem can be solved in polynomial time. For general networks, we provide effective heuristics for the above problems.

The problem of speaker identification is investigated. Basic segments – pseudostationary intervals of voiced sounds are used for identification. The identification is carried out, comparing average distances between an investigative and comparatives. Coefficients of the linear prediction model (LPC) of a vocal tract, cepstral coefficients and LPC coefficients of an excitation signal are used for identification as features. Three speaker identification methods are presented. Experimental investigation of their performance is discussed.

A method for processing sequential information of states of technological processes or other complicated dynamic systems and for sequential detection of many abrupt or slow changes in several unknown states is considered. The method is based on a sequential nonlinear mapping of many-dimensional vectors of parameters (collection of which describes the present state of dynamic systems) into two-dimensional vectors in order to reflect the states and their changes on the PC screen. The mapping error function is chosen and expressions for sequential nonlinear mapping are obtained. The mapping preserves the inner structure of distances among the vectors. An example is given. A theoretical minimum amount of parameter vectors mapped simultaneously at the very beginning is obtained.

An analytical review of recent publications in the area of digital speech signal processing is presented. The aim of the given paper is the analysis of these publications, where Artificial Neural Networks (ANNs) were successfully employed. Numerous methods of ANNs employment are discussed due to identify when and why they are reliable alternative to the conventional adaptive signal processing techniques.

The problem of dust explosion localization in corn processing enterprises is analysed. Condition of flame spread limiting in pipes is defined. Mathematical models of explosion processes are presented, which enable us to choose a mathematical expression form, with regard to desirable process identification accuracy and the complexity of analytic expressions.

In the present paper, the method of structure analysis for multivariate functions was applied to rational approximation in classification problems. Then the pattern recognition and generalisation ability was investigated experimentally in numerical recognition. A comparison with Hopfield Net was carried out. The overall results of using of new approach may be treated as a success.