Transfer function estimation from noisy input and output data
The author considers the problem of consistent identification of linear systems where both input and output observations contain white noise. Two new types of bias-eliminating least-squares (BELS) based algorithms are proposed. The original system is time invariant and is characterized by the equation \(x(t)=G(q^{-1})r(t)\), where the input \(r(t)\) is assumed to be a stationary random sequence with rational spectral density, \(x(t)\) denotes the output, \(q^{-1}\) stands for the unit-delay operator, and \(G(q^{-1})\) is a ratio of the form \[ G(q^{-1})= \left(\sum^m_{i=1} b_iq^{-i}\right) \left(1-\sum^n_{i=1} a_iq^{-i}\right)^{-1}. \] It is demonstrated that the proposed algorithms can give consistent parameter estimates without any use of prefiltering. A direct parameter estimation is possible and makes these algorithms more appealing in implementation and in on-line estimation. A numerical example illustrates the comparison of the algorithms.
- Numerically robust transfer function modeling from noisy frequency domain data
- Auxiliary model method for transfer function estimation from noisy input and output data
- On the estimation of transfer functions
- TRANSFER FUNCTION ESTIMATION
- scientific article; zbMATH DE number 4054855
- A note on transfer function model specification with noisy closed-loop input data
- Uncertainty of transfer function modelling using prior estimated noise models
- Determining the transfer function with the use of step response in the presence of noise
- Relations between bias-eliminating least squares, the Frisch scheme and extended compensated least squares methods for identifying errors-in-variables systems
- A covariance matching approach for identifying errors-in-variables systems
- Identification algorithms for linear systems with correlated input and output noise
- Robust identification of systems with both bias and variance disturbances
- Mean square error of the empirical transfer function estimator for stochastic input signals.
- Auxiliary model method for transfer function estimation from noisy input and output data
- A unified framework for EIV identification methods when the measurement noises are mutually correlated
- LPV system identification under noise corrupted scheduling and output signal observations
- The Frisch scheme in multivariable errors-in-variables identification
- Identification of continuous-time errors-in-variables models
- Accuracy analysis of bias-eliminating least squares estimates for errors-in-variables systems
- An improved bias-compensation approach for errors-in-variables model identification
- Errors-in-variables system identification using structural equation modeling
- On a least-squares-based algorithm for identification of stochastic linear systems.
- Bias-eliminating least-squares identification of errors-in-variables models with mutually correlated noises
- Approximative weighting for a covariance-matching approach for identifying errors-in-variables systems
- A note on transfer function model specification with noisy closed-loop input data
- PARAMETRIC IDENTIFICATION OF LINEAR NOISY INPUT-OUTPUT SYSTEMS
- scientific article; zbMATH DE number 3852266 (Why is no real title available?)
- Frisch scheme identification for dynamic diagonal bilinear models
- Parameter consistency and quadratically constrained errors-in-variables least-squares identification
- scientific article; zbMATH DE number 25209 (Why is no real title available?)
- ON UNBIASED PARAMETER ESTIMATION OF LINEAR SYSTEMS USING NOISY MEASUREMENTS
- Parameter estimation of stochastic linear systems with noisy input
- A bias-corrected estimator for nonlinear systems with output-error type model structures
- Algorithms for recursive/semi-recursive bias-compensating least squares system identification within the errors-in-variables framework
- Bias compensation-based parameter estimation for output error moving average systems
- Least-squares parameter estimation of linear systems with noisy input–output data
- A generalised instrumental variable estimator for multivariable errors-in-variables identification problems
- Errors-in-variables methods in system identification
- Recursive identification of errors-in-variables systems based on the correlation analysis
- A confirmatory factor analysis approach for addressing the errors-in-variables problem with colored output noise
- Accuracy analysis of a covariance matching approach for identifying errors-in-variables systems
- A generalized instrumental variable estimation method for errors-in-variables identification problems
- Maximum likelihood identification of noisy input-output models
- Unbiased equation-error based algorithms for efficient system identification using noisy measurements
This page was built for publication: Transfer function estimation from noisy input and output data
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4210692)
