All-Pole Estimation in Spectral Domain

Autoregressive (AR) modeling is a popular spectral analysis method commonly resolved in the time domain. This paper presents a novel AR analysis framework dealing with the estimation of poles directly from spectral samples. The basis of the method lies on a minimizing functional built with a certain mapping of the spectral residue. The optimization mechanism is based on the multivariate Newton-Raphson algorithm. Two different mappings are considered, namely, linear and logarithmic. The linear case results in a nonquadratic convex functional, whose global minimum is equivalent to that of the time-domain autocorrelation method. The logarithmic case under the maximum likelihood criterion turns out equivalent to the Whittle likelihood, proven here to be suitable for frequency selective estimation. The statistical and convergence performance of the method is demonstrated with simulations on stochastic and deterministic harmonic signals.

[1]  Piet M. T. Broersen,et al.  Time series analysis in a frequency subband , 2003, IEEE Trans. Instrum. Meas..

[2]  Paavo Alku,et al.  All-pole modeling technique based on weighted sum of LSP polynomials , 2003, IEEE Signal Processing Letters.

[3]  Fredrik Gustafsson,et al.  Frequency-domain continuous-time AR modeling using non-uniformly sampled measurements , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[4]  Rik Pintelon,et al.  Time series analysis in the frequency domain , 1999, IEEE Trans. Signal Process..

[5]  Tryphon T. Georgiou,et al.  Kullback-Leibler approximation of spectral density functions , 2003, IEEE Trans. Inf. Theory.

[6]  J. Makhoul,et al.  Linear prediction: A tutorial review , 1975, Proceedings of the IEEE.

[7]  Arye Nehorai,et al.  Adaptive pole estimation , 1990, IEEE Trans. Acoust. Speech Signal Process..

[8]  H. J. Orchard,et al.  The Laguerre method for finding the zeros of polynomials , 1989 .

[9]  Steven Kay,et al.  Modern Spectral Estimation: Theory and Application , 1988 .

[10]  Piet M. T. Broersen,et al.  Time-series analysis if data are randomly missing , 2006, IEEE Transactions on Instrumentation and Measurement.

[11]  M. Niedzwiecki,et al.  Generalized adaptive notch and comb filters for identification of quasi-periodically varying systems , 2005, IEEE Transactions on Signal Processing.

[12]  Alexander J. Smola,et al.  Learning with kernels , 1998 .

[13]  Gerald Matz,et al.  TFARMA models: order estimation and stabilization , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[14]  Yves Rolain,et al.  Box-Jenkins identification revisited - Part II: Applications , 2006, Autom..

[15]  Steven Kay,et al.  The effects of noise on the autoregressive spectral estimator , 1979 .

[16]  Carlos E. Davila,et al.  On the noise-compensated Yule-Walker equations , 2001, IEEE Trans. Signal Process..

[17]  J. Schoukens,et al.  Frequency domain system identification using arbitrary signals , 1997, IEEE Trans. Autom. Control..

[18]  J. Schoukens,et al.  Box-Jenkins identification revisited - Part I: Theory , 2006, Autom..

[19]  Eric Moulines,et al.  Estimation of the spectral envelope of voiced sounds using a penalized likelihood approach , 2001, IEEE Trans. Speech Audio Process..

[20]  F. Itakura.  Line spectrum representation of linear predictor coefficients of speech signals , 1975 .

[21]  R. Gray,et al.  Distortion measures for speech processing , 1980 .

[22]  J. L. Roux,et al.  A fixed point computation of partial correlation coefficients , 1977 .

[23]  Luis Weruaga,et al.  Adaptive chirp-based time-frequency analysis of speech signals , 2006, Speech Commun..

[24]  Denis Donnelly,et al.  The fast Fourier transform for experimentalists, part IV: autoregressive spectral analysis , 2005, Comput. Sci. Eng..