Anatoliy Platonov

Optimal identification of regression-type processes under adaptively controlled observations

The paper presents a new approach for the design and analysis of adaptive systems for the optimal identification of physical processes and signals described by linear regression-type equations. Contrary to the traditional methods, a compound model of the observed process is proposed. This model describes an unobservable process that is subject to identification and the observing device (sensor) separately. The introduced adaptive model of the sensor with bounded linear range of its characteristic is more general and adequate than the commonly used ones. It is shown that optimal adaptive control of the sensor parameters and its fit to the statistics of the identified process significantly improve the accuracy of the parameter estimates and increase their convergence rate. Results of the theoretical part of the paper are illustrated by a simple analytic example and confirmed via simulation. >

Optimization of Adaptive Communication Systems with Feedback Channels

The work presents a new approach to the concurrent optimization and design of transmitting and receiving parts of the adaptive communication systems (ACS). There is shown that optimal ACS used in the threshold mode provide, under given BER, a bit rate equal to the capacity of the forward channel.

Analog-digital systems for adaptive measurements and parameter estimation of noisy processes

The theoretical framework and principles of designing a new class of highly effective analog-digital computer-aided measurement systems are presented. The proposed approach enables the designers to determine uniquely the optimal structure, parameters and characteristics of measurement systems working in the presence of noises and disturbances as well as to carry out the comprehensive analysis of their work. The approach is based on the new idea of the nonlinear adaptively controlled (smart) sensor. The algorithms for optimal processing of the measured data and estimating unknown parameters of analyzed processes or objects, coupled with the optimal adaptive control of the sensor parameters, are derived. The superiority of measurement systems with adaptively controlled sensors over the conventional ones employing nonadaptive sensors is shown. The main theoretical results and conclusions are verified and confirmed via simulations taking as an example the string-frame accelerometer.

Particularities of cyclic intelligent ADC design, implementation and adjusting

The paper presents the results of design, implementation and adjusting of the CMOS version of the hardware prototype of new ldquointelligentrdquo cyclic analog-to-digit converter (IC ADC). There are presented the concept of intelligent conversion, main principles of practical implementation, main stages and particularities of design and adjusting the IC ADC.

Particularities of the Cyclic A/D Converters ENOB Definition and Measurement

The paper is devoted to methods of accurate assessment of expected and real effective number of hits (ENOB) of cyclic A/D converters. There is discussed the equivalence of definition of ENOB in IEEE Standard 1241-2000 and analytical definition of ENOB used in our previous works. Generalization of ENOB definition is discussed, which enables its accurate application to analysis and testing of the cyclic and pipe-line ADCs performance. The indirect and direct methods of ENOB assessment are discussed. The results of the paper can be used for development of reliable procedures for analysis and testing of ADC performance

Methods of Additional Improvement of Intelligent CADC Performance

The methods of additional improvement of the quality of sub-optimal intelligent cyclic A/D converters (IC ADC) are presented. There are proposed new approaches, based on utilisation of resources of digital part of converters (software), which extend the functional possibilities and range of IC ADC applications. The methods of definition of most appropriate parameters of the analogue part are discussed. The results of researches permit to design the converters with performance close to theoretically available.

Optimal synthesis of smart measurement systems with adaptive correction of drifts and setting errors of the sensor's working point

This paper deals with a problem in optimal synthesis of smart measurement systems. A nonlinear model of the sensor is introduced, and directly applied to the derivation of optimal algorithms for processing measurement data. The analysis is performed for systems with both noncompensatory and compensatory sensors. It is assumed that a measurement is performed in the presence of random disturbances and noise. The algorithms developed take into account the limited input-output range of the sensor as well as the influence of its internal noise on measurement quality. Simultaneously, they automatically and optimally correct errors in the initial setting of the sensors working point, as well as those caused by its random drift.

A New Approach to Design and Optimisation of Pipeline A/D Converters

Foundations of a new approach to design and optimisation of pipeline analogue-to-digital converters (PADC) are presented in the paper. The approach employs the results of analytical and applied research on the concurrent optimisation of the analogue and digital parts of adaptive estimation systems. Application of the approach allowed us to modify the known architecture and principles of conversion used in conventional PADC in the way significantly improving quality of conversion. Particularities of new PADC work were studied in advanced simulation experiments, which confirmed the results of analytical investigations. Special attention was paid to comparison of the performance of new and known PADC and assessment of potentially achievable benefits.

New Approach to Improvement and Measurement of the Performance of PHY Layer Links of WSN

This paper presents a new approach to the improvement of the sensor node - base station (SNBS) links of physical layer of wireless sensor networks performance. The key idea is the consideration of SNBS as the remote measurement and estimation systems and application of the corresponding analytical tools. It is shown that the main difficulty in the improvement of digital SNBS performance is impossibility to formulate and solve the optimization task using well developed methods of the optimal estimation theory. In turn, the SNBS transmitting the signals formed by the sensors using analog pulse-amplitude transmitter adjusted over the feedback channel permit the formulation of analytical form of the mean square error (MSE) and solution of optimization task. The obtained optimal transmission-reception algorithm enables designing of a low energy-size-cost analog SNBS, which transmit the signals with minimal MSE and a bit rate equal to the capacity of the link. Moreover, the MSE of transmission determines the information characteristics of the links, as well as enables a development of the unified methods of the SNBS real performance evaluation and measurement.

Analog-digital wide-range measurement systems with adaptively adjusted quartz glass sensors

In this paper, a new class of smart analog-digital systems for measurement of acceleration, displacement, size, weight, pressure, and other quantities, is considered. Theoretical basics and practical ways of realizing such systems are presented. Techno-economic and technological characteristics as well as possibilities of production of these systems are also discussed. Their principal feature is that they employ sensors whose parameters can be adjusted adaptively to the statistics of a measured process. Introducing appropriate modifications into conventional sensors, this enables us at no great expense to considerably extend the dynamic range of measurements as well as to increase their accuracy. With a view to superior quality characteristics and low cost, as a reference point for designing this class of measurement systems, electromechanical sensors based on quartz glass technology are chosen.