Non-minimum phase identification based on higher order spectrum slices
Résumé
A new family of methods, named PEP (Phase Estimation using Polyspectrum slices), for the reconstruction of the Fourier phase of a complex LTI system excited by a white non-Gaussian input is proposed. More precisely, we propose two subfamilies of methods, the q-PEP (q>2) and (q1, q2)-PEP (q2>q1>2) algorithms. The q-PEP methods exploit the best Two-Dimensional (2D) slice of the data q-th order spectrum. The originality of the (q1, q2)-PEP methods consists of exploiting simultaneously one 1D slice of the q1-th order spectrum and one 2D slice of the q2-th order spectrum. These new algorithms are easy both to implement and to use. Moreover, the asymptotic unbiasedness and consistency of these methods are demonstrated. Eventually, computer simulations show that the PEP algorithms exhibit in general better performances than classical methods especially for band-limited systems.
Mots clés
Non-minimum phase system identification
blind deconvolution
polyspectra
Single Input Single Output (SISO)
Fourier analysis
linear systems phase estimation
signal reconstruction
spectral analysis
time-varying systems
Fourier phase reconstruction
asymptotic unbiasedness
band-limited systems
higher order spectrum slices
linear time invariant system
method consistency
nonminimum phase identification
phase estimation
polyspectrum slices
white non-Gaussian input
Loading...