On the existence of a solution to a spectral estimation problem \emph{à la} Byrnes-Georgiou-Lindquist

Abstract

A parametric spectral estimation problem in the style of Byrnes, Georgiou, and Lindquist was posed in \cite{FPZ-10}, but the existence of a solution was only proved in a special case. Based on their results, we show that a solution indeed exists given an arbitrary matrix-valued prior density. The main tool in our proof is the topological degree theory.

6 pages of two-column draft, accepted for publication in IEEE-TAC

References in corpus (1)

• On the Factorization of Rational Discrete-Time Spectral Densities

Cited by in corpus (5)

• On a probabilistic approach to synthesize control policies from example datasets
• Probabilistic design of optimal sequential decision-making algorithms in learning and control
• On the Uniqueness Result of Theorem 6 in "Relative Entropy and the Multivariable Multidimensional Moment Problem"
• Discrete fully probabilistic design: towards a control pipeline for the synthesis of policies from examples
• M${}^2$-Spectral Estimation: A Flexible Approach Ensuring Rational Solutions