Time and Spectral Domain Relative Entropy: A New Approach to Multivariate Spectral Estimation

DOI10.1109/TAC.2012.2190153MaRDI QIDQ5352975zbMATH OpenOpenAlexFDO

Authors Augusto Ferrante, Chiara Masiero, Michele Pavon

Publication date 8 September 2017

Published in IEEE Transactions on Automatic Control (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1103.5602

Mathematics Subject Classification ID

Statistical aspects of information-theoretic topics (62B10) Signal detection and filtering (aspects of stochastic processes) (60G35) Inference from stochastic processes and spectral analysis (62M15) Estimation and detection in stochastic control theory (93E10)

Abstract: The concept of spectral relative entropy rate is introduced for jointly stationary Gaussian processes. Using classical information-theoretic results, we establish a remarkable connection between time and spectral domain relative entropy rates. This naturally leads to a new spectral estimation technique where a multivariate version of the Itakura-Saito distance is employed}. It may be viewed as an extension of the approach, called THREE, introduced by Byrnes, Georgiou and Lindquist in 2000 which, in turn, followed in the footsteps of the Burg-Jaynes Maximum Entropy Method. Spectral estimation is here recast in the form of a constrained spectrum approximation problem where the distance is equal to the processes relative entropy rate. The corresponding solution entails a complexity upper bound which improves on the one so far available in the multichannel framework. Indeed, it is equal to the one featured by THREE in the scalar case. The solution is computed via a globally convergent matricial Newton-type algorithm. Simulations suggest the effectiveness of the new technique in tackling multivariate spectral estimation tasks, especially in the case of short data records.

Cited in

(20)

• The Schrödinger problem on the non-commutative Fisher-Rao space
• Mean-square consistency of the f-truncated M^2-periodogram
• Space and spectral domain relative entropy for homogeneous random fields
• Adaptive spectral estimation for nonstationary multivariate time series
• A Well-Posed Multidimensional Rational Covariance and Generalized Cepstral Extension Problem
• A fast robust numerical continuation solver to a two-dimensional spectral estimation problem
• Empirical Bayesian learning in AR graphical models
• First-order methods for nonnegative trigonometric matrix polynomials
• A maximum entropy approach to the realizability of spin correlation matrices
• Line spectral analysis using the G-filter: an atomic norm minimization approach
• An interpretation of the dual problem of the THREE-like approaches
• M^2 spectral estimation: a flexible approach ensuring rational solutions
• Maximum entropy spectral estimation for regular time series of degenerate rank
• M^2-spectral estimation: a relative entropy approach
• On the estimation of structured covariance matrices
• scientific article; zbMATH DE number 3988553 (Why is no real title available?)
• Autoregressive identification of Kronecker graphical models
• On adaptive covariance and spectrum estimation of locally stationary multivariate processes
• Rational approximations of spectral densities based on the Alpha divergence
• Necessary and sufficient condition for modified Nevanlinna-Pick interpolation for closed-loop pole placement

This page was built for publication: Time and Spectral Domain Relative Entropy: A New Approach to Multivariate Spectral Estimation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5352975)