Global convergence of the EM algorithm for ARX models with uncertain communication channels

被引：13

作者：

Chen, Jing ^{[1
]}

Huang, Biao ^{[2
]}

Zhu, Quanmin ^{[3
]}

Liu, Yanjun ^{[1
]}

Li, Lun ^{[1
]}

机构：

[1] Jiangnan Univ, Key Lab Adv Proc Control Light Ind, Minist Educ, Wuxi 214122, Jiangsu, Peoples R China

[2] Univ Alberta, Dept Chem & Mat Engn, Edmonton, AB T6G 2G6, Canada

[3] Univ West England, Dept Engn Design & Math, Bristol BS16 1QY, Avon, England

来源：

SYSTEMS & CONTROL LETTERS | 2020年 / 136卷

基金：

中国国家自然科学基金;

关键词：

Parameter estimation; ARX model; EM algorithm; Modified Kalman filter; Kullback-Leibler divergence; MAXIMUM-LIKELIHOOD-ESTIMATION; RECURSIVE-IDENTIFICATION; SYSTEM-IDENTIFICATION; NONLINEAR-SYSTEMS; VARYING SYSTEMS; LINEAR-SYSTEM; QUANTIZATION;

D O I：

10.1016/j.sysconle.2019.104614

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

An expectation maximization (EM) algorithm is presented for ARX modeling with uncertain communication channels. The considered model consists of two parts: a dynamic model which is expressed by an ARX model, and an output model, both subject to white Gaussian noises. Since the true outputs of the ARX model are assumed to be unknown, a modified Kalman filter is derived to estimate the output, and then the parameters are estimated by the EM algorithm using the estimated outputs. The Kullback-Leibler divergence and the submartingale are used to prove that the parameter estimates can converge to the true values with the EM algorithm. Furthermore, a simulation example is presented to verify the theoretical results. (C) 2020 Elsevier B.V. All rights reserved.

引用

页数：8

共 50 条

[31] Singularity and Slow Convergence of the EM algorithm for Gaussian Mixtures
Hyeyoung Park
Tomoko Ozeki
[J]. Neural Processing Letters, 2009, 29 : 45 - 59
[32] Improved convergence guarantees for learning Gaussian mixture models by EM and gradient EM
Segol, Nimrod
Nadler, Boaz
[J]. ELECTRONIC JOURNAL OF STATISTICS, 2021, 15 (02): : 4510 - 4544
[33] On convergence analysis of an accelerated EM algorithm for SPECT imaging
Wang, Jinping
Bu, Jing
Guo, Yao
[J]. APPLICABLE ANALYSIS, 2021, 100 (15) : 3233 - 3243
[34] Convergence of stochastic gradient estimation algorithm for multivariable ARX-like systems
Liu, Yanjun
Sheng, Jie
Ding, Ruifeng
[J]. COMPUTERS & MATHEMATICS WITH APPLICATIONS, 2010, 59 (08) : 2615 - 2627
[35] COMMUNICATION CONSTRAINED LEARNING WITH UNCERTAIN MODELS
Hare, James Zachary
Uribe, Cesar A.
Kaplan, Lance M.
Jadbabaie, Ali
[J]. 2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, 2020, : 8609 - 8613
[36] A note on EM algorithm for mixture models
Yao, Weixin
[J]. STATISTICS & PROBABILITY LETTERS, 2013, 83 (02) : 519 - 526
[37] Recursive identification of switched ARX hybrid models: Exponential convergence and persistence of excitation
Vidal, R
Anderson, BDO
[J]. 2004 43RD IEEE CONFERENCE ON DECISION AND CONTROL (CDC), VOLS 1-5, 2004, : 32 - 37
[38] Communication-efficient distributed EM algorithm
Liu, Xirui
Wu, Mixia
Xu, Liwen
[J]. STATISTICAL PAPERS, 2024,
[39] The very uncertain prospect of "global" convergence in corporate governance
Branson, DM
[J]. CORNELL INTERNATIONAL LAW JOURNAL, 2001, 34 (02) : 321 - 362
[40] Stabilization of nested systems with uncertain subsystem communication channels
Yadav, V
Voulgaris, PG
Salapaka, MV
[J]. 42ND IEEE CONFERENCE ON DECISION AND CONTROL, VOLS 1-6, PROCEEDINGS, 2003, : 2853 - 2858

← 1 2 3 4 5 →