Title :
Identification of nonminimum phase systems using the second- and third-order statistics
Author :
ALshebeili, S. ; Venetsanopoulos, A. ; Cetin, A.
Author_Institution :
Dept. of Electr. Eng., Toronto Univ., Ont., Canada
Abstract :
The authors present a method for reconstruction of the impulse response of a nonminimum phase, linear, time-invariant (NMP-LTI) system from the second- and third-order statistics of the output sequence. The method is based on computing the complex cepstrum of the unknown impulse response from the power spectrum and a one-dimensional (1D) slice of the bispectrum. This method has the advantages of being independent of the type of the system (MA, AR, ARMA) and its computational complexity is also independent of the order of the utilized spectrum.<>
Keywords :
T invariance; identification; linear systems; spectral analysis; statistical analysis; transient response; bispectrum; complex cepstrum; computational complexity; impulse response; linear time-invariant system; nonminimum phase systems; output sequence; power spectrum; second-order statistics; signal processing; spectral analysis; system identification; third-order statistics; Cepstrum; Data processing; Fourier transforms; Phase estimation; Power engineering computing; Reconstruction algorithms; Signal processing; Sonar; Statistics; System identification;
Conference_Titel :
Spectrum Estimation and Modeling, 1990., Fifth ASSP Workshop on
Conference_Location :
Rochester, NY, USA
DOI :
10.1109/SPECT.1990.205538
