On Iterative Parameter Identification of FIR Systems with Batched Possibly Incorrect Binary-Valued Observations

Jian Guo, Wenchao Xue*, Ting Wang, Ji Feng Zhang, Yanjun Zhang

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper considers the problem of parameter identification for a binary output finite impulse response (FIR) system with measurement error, where the measurement error makes the binary measurement values take opposite values with a certain probability. First, the maximum likelihood estimation (MLE) of the parameters is given and an iterative algorithm with projection based on the Expectation-Maximization algorithm is presented to calculate the MLE. Furthermore, the necessary and sufficient condition for the likelihood function to have a unique maximum point is obtained. It is proved that the iterative estimation error converges to zero at an exponential rate under persistently excitation input conditions. Finally, some numerical simulation results based on a typical system show the effectiveness of the proposed algorithm.

Original languageEnglish
Title of host publication2023 62nd IEEE Conference on Decision and Control, CDC 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages4936-4941
Number of pages6
ISBN (Electronic)9798350301243
DOIs
Publication statusPublished - 2023
Event62nd IEEE Conference on Decision and Control, CDC 2023 - Singapore, Singapore
Duration: 13 Dec 202315 Dec 2023

Publication series

NameProceedings of the IEEE Conference on Decision and Control
ISSN (Print)0743-1546
ISSN (Electronic)2576-2370

Conference

Conference62nd IEEE Conference on Decision and Control, CDC 2023
Country/TerritorySingapore
CitySingapore
Period13/12/2315/12/23

Keywords

  • Binary-valued observation
  • exponential rate
  • maximum likelihood estimate
  • strongly convex
  • system identification

Fingerprint

Dive into the research topics of 'On Iterative Parameter Identification of FIR Systems with Batched Possibly Incorrect Binary-Valued Observations'. Together they form a unique fingerprint.

Cite this