Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

IEEE International Conference on Acoustics Speech and Signal Processing

Marginals vs. covariance in joint distribution theory

Richard G. Baraniuk

Published Nov 19, 2002• 13 citations
Publisher

Abstract

Cohen (1966, 1995) has proposed a method for constructing joint distributions of arbitrary physical quantities, in direct generalization of joint time-frequency representations. We investigate the covariance properties of this procedure and caution that in its present form it cannot generate all possible distributions. Using group theory, we extend Cohen's construction to a more general form that can be customized to satisfy specific marginal and covariance requirements.

DOI
10.1109/icassp.1995.480407

Related papers

Beyond time-frequency analysis: energy densities in one and many dimensions

Richard G. Baraniuk

Given a unitary operator A representing a physical quantity of interest, we employ concepts from group representation theory to define two natural signal energy densities for A. The first is invariant to A and proves useful when the effect of A is to be ignored; the second is covariant to A and measures the "A" content of signals. We also consider joint densities for multiple operators and, in the process, provide an alternative interpretation of Cohen's (see Englewood Cliffs, NJ: Prentice-Hall, 1995) general construction for joint distributions of arbitrary variables.

IEEE Transactions on Signal Processing•Jan 1, 1998•23 citations

Joint distributions for arbitrary variables

Leon W. Cohen, Richard G. Baraniuk

We compare the methods of Cohen (1971) and of Baraniuk and Jones (see Proc. IEEE ICASSP-93, p.320-323, 1993) for obtaining joint distributions for arbitrary variables. We show that the two methods produce identical results for variables whose associated operators are obtained via a unitary transformation of the time and frequency operators. In addition, we generalize this result and show that all pairs of variables connected by a unitary transformation have joint distributions that are identical in functional form. This result shows how pairs of variables can be grouped into classes whose joint distributions are functionally equivalent and, therefore, share equivalent properties.>

IEEE-SP International Symposium on Time-Frequency and Time-Scale Analysis•Dec 17, 2002•2 citations

A general approach for obtaining joint representations in signal analysis. I. Characteristic function operator method

Leon W. Cohen

A general approach for obtaining joint representations for arbitrary physical quantities is presented. The characteristic function operator method of Moyal (1949) and Ville (1960) and generalized by Cohen (1966, 1976) and Scully and Cohen (1987) for the time-frequency case is developed for arbitrary variables.

IEEE Transactions on Signal Processing•May 1, 1996•33 citations

variant Time-Frequency Distributions ased on Conjugate Operators

Franz Hlawatsch, Helmut Bölcskei

We propose classes of quadratic time-frequency distributions that retain the inner structure of Cohen's class. Each of these classes is based on a pair of conjugate unitary operators producing time-frequency displacements. The classes satisfy covariance and marginal properties corresponding to these operators. For each class, we define a central member generalizing the Wigner distribution and the Q-distribution, and we specify a transformation by which the class can be derived from Cohen's class.

•Jan 1, 1996•0 citations

Properties of the positive time-frequency distribution functions

Leon W. Cohen

Joint positive distributions of time and frequency which satisfy the correct marginals exist and can be readily obtained. In this presentation we generalize the continuous case previously considered to the discrete time case and the two dimensional situation.

IEEE International Conference on Acoustics Speech and Signal Processing•Mar 23, 2005•2 citations

Beyond time-frequency analysis: energy densities in one and many dimensions

Richard G. Baraniuk

Given a unitary operator A representing a physical quantity of interest, we employ concepts from group representation theory to define two natural signal energy densities for A. The first is invariant to A and proves useful when the effect of A is to be ignored; the second is covariant to A and measures the "A" content of signals. The construction is quite general and is also easily extended to the multi-operator case, which generalizes previously derived joint densities such as the time-frequency and time-scale distributions.>

IEEE International Conference on Acoustics Speech and Signal Processing•Dec 17, 2002•26 citations
Cohen (1966, 1995) has proposed a method for constructing joint distributions of arbitrary physical quantities, in direct generalization of joint time-frequency representations. We investigate the covariance properties of this procedure and caution that in its present form it cannot generate all possible distributions. Using group theory, we extend Cohen's construction to a more general form that can be customized to satisfy specific marginal and covariance requirements.
Given a unitary operator A representing a physical quantity of interest, we employ concepts from group representation theory to define two natural signal energy densities for A. The first is invariant to A and proves useful when the effect of A is to be ignored; the second is covariant to A and measures the "A" content of signals. We also consider joint densities for multiple operators and, in the process, provide an alternative interpretation of Cohen's (see Englewood Cliffs, NJ: Prentice-Hall, 1995) general construction for joint distributions of arbitrary variables.
We compare the methods of Cohen (1971) and of Baraniuk and Jones (see Proc. IEEE ICASSP-93, p.320-323, 1993) for obtaining joint distributions for arbitrary variables. We show that the two methods produce identical results for variables whose associated operators are obtained via a unitary transformation of the time and frequency operators. In addition, we generalize this result and show that all pairs of variables connected by a unitary transformation have joint distributions that are identical in functional form. This result shows how pairs of variables can be grouped into classes whose joint distributions are functionally equivalent and, therefore, share equivalent properties.>
We propose classes of quadratic time-frequency distributions that retain the inner structure of Cohen's class. Each of these classes is based on a pair of conjugate unitary operators producing time-frequency displacements. The classes satisfy covariance and marginal properties corresponding to these operators. For each class, we define a central member generalizing the Wigner distribution and the Q-distribution, and we specify a transformation by which the class can be derived from Cohen's class.
Given a unitary operator A representing a physical quantity of interest, we employ concepts from group representation theory to define two natural signal energy densities for A. The first is invariant to A and proves useful when the effect of A is to be ignored; the second is covariant to A and measures the "A" content of signals. The construction is quite general and is also easily extended to the multi-operator case, which generalizes previously derived joint densities such as the time-frequency and time-scale distributions.>