Title :
Analysis of stochastic gradient identification of Hermite polynomial systems with memory
Author :
Celka, P. ; Bershad, N.J. ; Vesin, J.M.
Author_Institution :
Dept. of Electr. Eng., Swiss Fed. Inst. of Technol., Lausanne, Switzerland
Abstract :
This paper presents analytical and Monte Carlo results for a stochastic gradient adaptive scheme which identifies an orthogonal polynomial-type nonlinear system with memory. The analysis includes recursions for the mean and fluctuation behavior of the adaptive parameters as well computation of the mean-square error increase due to the adaptation process
Keywords :
Monte Carlo methods; adaptive estimation; filtering theory; gradient methods; identification; mean square error methods; nonlinear systems; polynomials; stochastic processes; Hermite polynomial systems; Monte Carlo results; adaptive parameters; approximate solution; fluctuation behavior; linear filter estimation; mean behavior; mean-square error; memory; orthogonal polynomial-type nonlinear system; statistical analysis; stochastic gradient adaptive scheme; stochastic gradient identification; Adaptive signal processing; Australia; Fluctuations; Nonlinear filters; Polynomials; Signal processing algorithms; Stochastic processes; Stochastic systems; Taylor series; Vectors;
Conference_Titel :
Signal Processing and Its Applications, 1999. ISSPA '99. Proceedings of the Fifth International Symposium on
Conference_Location :
Brisbane, Qld.
Print_ISBN :
1-86435-451-8
DOI :
10.1109/ISSPA.1999.815840