Identification of low rank vector processes

DOI10.1016/J.AUTOMATICA.2023.110938zbMATH Open1520.93099arXiv2111.10899MaRDI QIDQ6164044FDOQ6164044

Authors: Wenqi Cao, Giorgio Picci, Anders Lindquist 

Publication date: 30 June 2023

Published in: Automatica (Search for Journal in Brave)

Abstract: We study modeling and identification of stationary processes with a spectral density matrix of low rank. Equivalently, we consider processes having an innovation of reduced dimension for which Prediction Error Methods (PEM) algorithms are not directly applicable. We show that these processes admit a special feedback structure with a deterministic feedback channel which can be used to split the identification in two steps, one of which can be based on standard algorithms while the other is based on a deterministic least squares fit. Identifiability of the feedback system is analyzed and a unique identifiable structure is characterized. Simulations show that the proposed procedure works well in some simple examples.

Full work available at URL: https://arxiv.org/abs/2111.10899

Recommendations

• Identification of rank one rational spectral densities from noisy observations: a stochastic realization approach
• When are two multivariate random processes indistinguishable
• Structured low-rank approximation and its applications
• Non-full-rank causal approximations of full-rank multivariate stationary processes with rational spectrum
• Sparse plus low rank network identification: a nonparametric approach

zbMATH Keywords

multivariable system identificationfeedback representationlow-rank process identificationrank-reduced output noise

Mathematics Subject Classification ID

System identification (93B30) Feedback control (93B52)

Cites Work

• Fast projection methods for minimal design problems in linear system theory
• On covariance function tests used in system identification
• A Bayesian approach to sparse dynamic network identification
• Robust dynamical network structure reconstruction
• Singular arma systems: a structure theory
• Generalized linear dynamic factor models: an approach via singular autoregressions
• Minimal Degree Coprime Factorization of Rational Matrices
• Title not available (Why is that?)
• Identifiability of linear dynamic networks
• Representations of jointly stationary stochastic feedback processes†
• Title not available (Why is that?)
• AR Identification of Latent-Variable Graphical Models
• Algorithms for Triangular Decomposition of Block Hankel and Toeplitz Matrices with Application to Factoring Positive Matrix Polynomials
• Feedback between stationary stochastic processes
• Identification of processes in closed loop-identifiability and accuracy aspects
• Linear Stochastic Systems
• Prediction error identification of linear dynamic networks with rank-reduced noise
• Low Rank and Structured Modeling of High-Dimensional Vector Autoregressions
• Learning Latent Variable Dynamic Graphical Models by Confidence Sets Selection
• Modeling Complex Systems by Generalized Factor Analysis

Cited In (3)

• Dealing with collinearity in large-scale linear system identification using Gaussian regression
• Spectral Factorization of Rank-Deficient Rational Densities
• Independent Vector Analysis: Identification Conditions and Performance Bounds

This page was built for publication: Identification of low rank vector processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6164044)