DocumentCode :
1973190
Title :
Robust system identification for non-persistently exciting input signals
Author :
Hull, AW ; Jenkins, W.K.
Author_Institution :
Coordinated Sci. Lab., Illinois Univ., IL, USA
fYear :
1989
fDate :
14-16 Aug 1989
Firstpage :
602
Abstract :
The performance of adaptive identification algorithms is constrained by the spectral characteristics of the input signals. Too few frequency components may result in parameter wander, and possible instability. Direct sequence spread-spectrum techniques may be used to increase the spectral richness of the training signal. Computer simulations of gradient descent algorithms for FIR (finite impulse response) and IR (infinite impulse response) systems indicate that parameter wander is eliminated and the rate of convergence is dramatically increased. The convergence of least squares algorithms is unaffected, but it is conjectured that their numerical properties are improved
Keywords :
adaptive systems; identification; signal processing; FIR system; IIR system; adaptive identification algorithms; convergence rate; direct sequence; finite impulse response; gradient descent algorithms; infinite impulse response; least squares algorithms; nonpersistently exciting input signals; parameter wander; robust system identification; spectral characteristics; spread-spectrum techniques; training signal; Convergence; Frequency; Least squares approximation; Least squares methods; Resonance light scattering; Robustness; Signal processing; Signal processing algorithms; Spread spectrum communication; System identification;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Circuits and Systems, 1989., Proceedings of the 32nd Midwest Symposium on
Conference_Location :
Champaign, IL
Type :
conf
DOI :
10.1109/MWSCAS.1989.101926
Filename :
101926
Link To Document :
بازگشت