Title :
AR model order selection based on bispectral cross-correlation
Author :
Noonan, J. ; Premus, V. ; Irza, J.
Author_Institution :
Dept. of Electr. Eng., Tufts Univ., Medford, MA, USA
Abstract :
A new method is presented for optimal model order selection for autoregressive bispectrum estimation. Simulation results are reviewed which demonstrate the method´s performance for the case of quadratically coupled sinusoids embedded in white Gaussian noise.<>
Keywords :
correlation theory; optimisation; parameter estimation; spectral analysis; white noise; AR model order selection; autoregressive bispectrum estimation; bispectral cross-correlation; optimal model order selection; quadratically coupled sinusoids; white Gaussian noise; Autocorrelation; Frequency estimation; Gaussian noise; Noise figure; Optimization methods; Power engineering and energy; Power measurement; Signal to noise ratio; Spectral analysis; TV;
Conference_Titel :
Spectrum Estimation and Modeling, 1990., Fifth ASSP Workshop on
Conference_Location :
Rochester, NY, USA
DOI :
10.1109/SPECT.1990.205581
