Polyspectra of ordered signals

Polyspectra are related to Fourier transforms of moment or cumulant functions of any order of random signals. They play an important role in many problems of signal analysis and processing. However, there are only a few statistical models giving explicitly the expression of polyspectra. Ordered signals are signals for which the explicit expression of the moment functions requires that the time instants appearing in these moments are put in an increasing order. There are many examples of such signals, the best known being the random telegraph signal constructed from a Poisson process. Some of these examples are presented and analyzed. The origin of the ordering structure is related with the point that real time is an oriented variable making a difference between past and future. This especially appears in Markov processes. The calculation of polyspectra is difficult because ordering is not adapted to Fourier analysis. By an appropriate grouping of various terms, the explicit expression of spectral moment functions is obtained. It shows in particular that many ordered signals present a normal density on the normal manifolds of the frequency domain and another contribution on the stationary manifold that is explicitly calculated. The analysis of the structure of this expression allows us to discuss some relationships with normal distribution, central limit theorem, and time reversibility.

[1]  W. M. Wonham,et al.  Probability Densities of the Smoothed ‘ Random Telegraph Signal ’† , 1958 .

[2]  Hagit Messer,et al.  On the principal domain of the discrete bispectrum of a stationary signal , 1995, IEEE Trans. Signal Process..

[3]  R. Mortensen.  Random Signals and Systems , 2019, Model-Based Processing.

[4]  J. A. McFadden.  The probability density of the output of a filter when the input is a random telegraphic signal: Differential-equation method , 1959, IRE Trans. Inf. Theory.

[5]  Messaoud Benidir,et al.  Polyspectrum modeling using linear or quadratic filters , 1993, IEEE Trans. Signal Process..

[6]  D. Brillinger.  An Introduction to Polyspectra , 1965 .

[7]  Vinod Chandran,et al.  A general procedure for the derivation of principal domains of higher-order spectra , 1994, IEEE Trans. Signal Process..

[8]  R. Barakat.  Another random telegraph signal , 1978, Proceedings of the IEEE.

[9]  Bernard C. Picinbono,et al.  On circularity , 1994, IEEE Trans. Signal Process..

[10]  John W. Dalle Molle,et al.  Trispectral analysis of stationary random time series , 1995 .

[11]  Roger F. Dwyer,et al.  Fourth‐order spectra of Gaussian amplitude‐modulated sinusoids , 1991 .

[12]  Bernard C. Picinbono.  Ergodicity and fourth-order spectral moments , 1997, IEEE Trans. Inf. Theory.

[13]  Sang-Won Nam,et al.  Application of higher order spectral analysis to cubically nonlinear system identification , 1994, IEEE Trans. Signal Process..

[14]  B. Picinbono,et al.  Higher-Order Coherence Functions of Optical Fields and Phase Fluctuations , 1968 .

[15]  G. Weiss.  TIME-REVERSIBILITY OF LINEAR STOCHASTIC PROCESSES , 1975 .

[16]  Claude Lefèvre,et al.  On time-reversibility and the uniqueness of moving average representations for non-Gaussian stationary time series , 1988 .

[17]  Albert N. Shiryaev,et al.  On a Method of Calculation of Semi-Invariants , 1959 .