Multiwindow estimators of correlation

This paper is concerned with the structure of estimators of correlation matrices and correlation sequences. We argue that reasonable estimators of the correlation matrix are quadratic in the data and nonnegative definite. We also specify the structure of the estimator when the data are modulated: a property we call modulation covariance. We state a representation theorem for estimators that have these attributes. We also derive a representation for estimators that have the additional requirement that the estimated matrix be Toeplitz. These representation theorems admit estimators that use multiwindowed copies of the data. In many circumstances, this multiwindow structure is superior to the conventional sum of lagged-products or outer-product estimators.

[1]  A. Schuster,et al.  On Lunar and Solar Periodicities of Earthquakes , 1897, Nature.

[2]  D. Thomson.  Quadratic-inverse spectrum estimates: applications to palaeoclimatology , 1990, Philosophical Transactions of the Royal Society of London. Series A: Physical and Engineering Sciences.

[3]  Thomas Kailath,et al.  ESPRIT-estimation of signal parameters via rotational invariance techniques , 1989, IEEE Trans. Acoust. Speech Signal Process..

[4]  Petre Stoica,et al.  MUSIC, maximum likelihood, and Cramer-Rao bound , 1989, IEEE Transactions on Acoustics, Speech, and Signal Processing.

[5]  Charles R. Johnson,et al.  Matrix analysis , 1985, Statistical Inference for Engineers and Data Scientists.

[6]  Georges Bienvenu,et al.  Adaptivity to background noise spatial coherence for high resolution passive methods , 1980, ICASSP.

[7]  M. P. Clark.  Frequency-wavenumber spectrum analysis using quadratic estimators , 1992, [1992] IEEE Sixth SP Workshop on Statistical Signal and Array Processing.

[8]  Henri Clergeot,et al.  Choix entre les différentes méthodes quadratiques d’estimation du spectre de puissance , 1984 .

[9]  D. Thomson,et al.  Spectrum estimation and harmonic analysis , 1982, Proceedings of the IEEE.

[10]  Petre Stoica,et al.  MUSIC, maximum likelihood and Cramer-Rao bound: further results and comparisons , 1989, International Conference on Acoustics, Speech, and Signal Processing,.

[11]  R. O. Schmidt,et al.  Multiple emitter location and signal Parameter estimation , 1986 .

[12]  C. T. Mullis,et al.  Quadratic Estimators of the Power Spectrum , 1989 .

[13]  U. Grenander,et al.  Statistical Spectral Analysis of Time Series Arising from Stationary Stochastic Processes , 1953 .

[14]  J. W. Tukey,et al.  The Measurement of Power Spectra from the Point of View of Communications Engineering , 1958 .

[15]  L.L. Scharf,et al.  Quadratic Estimators Of The CorrelationMatrix , 1994, IEEE Seventh SP Workshop on Statistical Signal and Array Processing.

[16]  Louis L. Scharf,et al.  Multiwindow estimators of correlation , 1994, Proceedings of 1994 28th Asilomar Conference on Signals, Systems and Computers.