Application of M-Estimation in Robust Recursive System Identification

An iterative least squares method with modified residuals is presented which is dedicated to the robust coefficient estimation of Walsh functions for time series contaminated with noise, some of which may even be outliers. Instead of the mean-square approximation error (MSE), a robust criterion is proposed for estimating the coefficients of the time series. It is minimised by applying the ordinary iterative Gauss-Newton approach so that an arbitrary function, which is absolutely integrable in the interval [0, T), can be properly approximated by the first M Walsh functions. A proof of convergence of the proposed method is provided. Results of simulation confirming robustness and convergence of the robust estimates are included. This method should be of great value in real-life situations.

show abstract

“…11). This is the main advantage of this method over other robust methods [10][11][12] presented before.…”

Section: \Aa)= [R M (T Jomentioning

confidence: 77%

Robust coefficient estimation of Walsh functions

Dai

Sinha

1990

IEE Proc. D Control Theory Appl. UK

Self Cite

View full text Add to dashboard Cite

show abstract

“…This is due to the fact that backward prediction errors appear in the lattice algorithm, preventing us from using directly the elegant methods of robustizing utilized in [ 11] for the forward prediction method.…”

Section: + η τ (η)φ-\η -\)U(n)mentioning

confidence: 99%

Weighted exact least squares lattice algorithm for adaptive noise cancellation

Madhavan

Bruin

1992

Can. J. Electr. Comput. Eng.

View full text Add to dashboard Cite

A new weighted exact least squares lattice algorithm for adaptive noise cancellation is developed. In addition to the forgetting factor, an independent weighting factor is included in the performance criterion to be minimized. Such a weighting scheme can be used to improve noise cancellation performance. We explore this by comparing the features of the weighted and nonweighted algorithms. Using simulations, we demonstrate the superior performance of the weighted exact least squares lattice algorithm.On développe un nouvel algorithme aux moindres carrés en treillis pondérés exactement pour l'annulation du bruit adaptif. En plus du facteur d'oubli, un facteur de pondération indépendant est inclus dans le critère de performance à minimiser. Une technique de pondération de ce type peut être utilisée pour améliorer la performance sur l'annulation du bruit. Cette possibilité est étudiée en comparant les caractéristiques des algorithmes pondérés et non-pondérés. Par simulations, on montre que l'algorithme en treillis aux moindres carrés pondérés offre une performance supérieure.

show abstract

An Improved Recursive Identification Method For Bilinear Systems

Liu

Wang

1993

International Journal of Modelling and Simulation

View full text Add to dashboard Cite

Application of M-Estimation in Robust Recursive System Identification

Cited by 3 publications

References 6 publications

Robust coefficient estimation of Walsh functions

Robust coefficient estimation of Walsh functions

Weighted exact least squares lattice algorithm for adaptive noise cancellation

An Improved Recursive Identification Method For Bilinear Systems

Contact Info

Product

Resources

About