Title :
Identification of nonminimum phase MA systems using cepstral operations on slices of higher order spectra
Author :
Alshebeili, Saleh A. ; Çetin, A. Enis ; Venetsanopoulos, A.N.
Author_Institution :
Dept. of Electr. Eng., King Saud Univ., Riyadh, Saudi Arabia
fDate :
9/1/1992 12:00:00 AM
Abstract :
Recursive and least-squares methods for reconstructing the impulse response of a nonminimum phase, linear time-invariant (NMP-LTI) MA system from its (k+1)st-order spectrum (k>1) are presented. The methods are nonparametric. They are based on computing the complex cepstrum of the impulse response sequences of the unknown system from the diagonal slice of the (k+1)st-order spectrum. The computational complexity of the system identification problem can be considerably reduced by using slices of higher-order spectra
Keywords :
computational complexity; identification; least squares approximations; signal processing; transient response; cepstral operations; computational complexity; higher order spectra; impulse response sequences; least-squares methods; linear time-invariant; moving average system; nonminimum phase MA systems; recursive methods; system identification; Cepstral analysis; Cepstrum; Computational complexity; Councils; Least squares methods; Random processes; Signal processing; Signal processing algorithms; System identification; Transfer functions;
Journal_Title :
Circuits and Systems II: Analog and Digital Signal Processing, IEEE Transactions on