The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
In this paper an approximation algorithm allowing to calculate parameters of rational and fractional continuous-time transfer functions as well as transfer functions being their series connections with delay from the identified frequency response is proposed. Parameters of the mentioned transfer functions are obtained by an optimisation algorithm utilizing ideas of randomized search. The presented...
In the paper a modification enabling acceleration of the rate of convergence for LMS-like on-line identification and adaptation algorithms is proposed. This is based on an artificial decaying of initial conditions in recursive identification as well as adaptation algorithms. The decaying is done using a set of the most recent measurements. Properties of the algorithms with the proposed modification...
In this paper properties of adaptive algorithms utilising ideas of stochastic gradient search optimisation applied to noise reduction are discussed. A focus on their average rate of convergence as well as on the obtained noise reduction is given. The presented discussion is based on a simulation case study devoted to noise reduction using a feedforward adaptive active noise control system.
In this paper a decomposition of observed mixtures of wide-sense stationary time-series coming from different sources into the corresponding orthogonal components being autoregressive-moving-average (ARMA) time-series is proposed. The first step is to estimate power spectral density or autocorrelation function for the observed mixture. Models of the mentioned orthogonal components are obtained by...
In the paper a unified approach to parametric nonlinear input-output dynamic system identification based on a memetic optimisation algorithm is presented. Efficiency of the presented approach is illustrated by a simulated identification of block-oriented nonlinear input-output dynamic systems. A focus on model identification for the case of very small as well as large number of input-output measurements...
In the paper an approach to parametric continuous-time input-output linear dynamic system identification based on a randomized search method is presented. During identification experiment the excitation signal is generated directly from D/A converter that is equipped only with zero-order or first-order hold filter — the corresponding low-pass analogue reconstruction filter is not used. In this experiment...
In the paper a strategy for discrete-time model identification of linear dynamic subsystem in simple block-oriented nonlinear systems operating under feedback is presented. In the proposed procedure a discrete-time frequency response of this linear part is identified using integrated bispectra or trispectra. Based on the identified frequency response the corresponding rational parametric approximation...
In the paper an approach to AR time-series identification based on observations obtained using data acquisition system equipped with a quantizer having saturation is presented. In the presented approach AR time-series model identification is replaced by an ARMA time-series model identification that returns parameters of AR time-series. A focus on model identification for ultra low- and ultra high-power...
In the paper an approach to discrete-time frequency response identification of a plant operating under feedback in the case of periodic disturbances is presented. In the presented approach to discrete-time frequency response identification integrated bispectra and higher-order non-gaussian multisine excitations are applied. A focus on almost sure convergence of the identified discrete-time frequency...
In the paper an approach to Hammerstein system identification based on quantized low-power observations is presented. In the presented approach realisations of discrete-time multisine random processes are used as the Hammerstein system excitation and the overall Hammerstein system identification problem is decomposed into the linear dynamic control channel subsystem, the corresponding disturbance...
In the paper an approach to discrete-time frequency response as well as the corresponding parametric model identification using deterministic discrete-time multisine excitations and quantized plant output is presented. A focus on model identification in the case of disturbance-free plant output and output signal level comparable with data acquisition system accuracy is given. Convergence of the identified...
In the paper stationarity of random processes obtained from elementary bilinear time-series models is analysed. It is shown in the presented analysis that elementary bilinear time-series models can be interpreted as linear time-series with time varying random parameters. The analysis is illustrated by simulation experiments showing time dependencies of some statistical moments of the discussed random...
In the paper an approach to discrete-time frequency response identification in the case of periodic disturbances is presented. A focus on almost sure convergence of the identified discrete-time frequency response to true plant is given. The presented discussion is illustrated by an example showing effectiveness of the presented approach.
Multisine time-series are sums of discrete-time sines with deterministic amplitudes and phase shifts chosen in a special way. In this paper a new recursive approach to synthesis and simulation of multisine time-series with predefined spectral properties, is presented. On the basis of the power spectral density function and phase shifts chosen with some well-defined properties, the finite discrete...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.