eprintid: 2471
rev_number: 5
eprint_status: archive
userid: 6
dir: disk0/00/00/24/71
datestamp: 2015-01-13 14:01:48
lastmod: 2015-01-13 14:01:48
status_changed: 2015-01-13 14:01:48
type: article
metadata_visibility: show
creators_name: Cerone, Vito
creators_name: Piga, Dario
creators_name: Regruto, Diego
creators_id: 
creators_id: dario.piga@imtlucca.it
creators_id: 
title: Computational load reduction in bounded error identification of Hammerstein systems
ispublished: pub
subjects: QA75
divisions: CSA
full_text_status: none
keywords: Convex relaxation; Hammerstein systems; parameter bounds; set-membership identification
abstract: In this technical note we present a procedure for the identification of Hammerstein systems from measurements affected by bounded noise. First, we show that computation of tight parameter bounds requires the solution to nonconvex optimization problems where the number of decision variables increases with the length of the experimental data sequence. Then, in order to reduce the computational burden of the identification problem, we propose a procedure to relax the formulated problem into a collection of polynomial optimization problems where the number of variables does not depend on the number of measurements. Advantages of the presented approach with respect to previously published results are discussed and highlighted by means of a simulation example.
date: 2013-05
date_type: published
publication: IEEE Transactions on Automatic Control 
volume: 58
number: 5
publisher: IEEE 
pagerange: 1317-1322
id_number: 10.1109/TAC.2012.2223334
refereed: TRUE
issn: 0018-9286
official_url: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6327597&isnumber=6504746
citation:   Cerone, Vito and Piga, Dario and Regruto, Diego  Computational load reduction in bounded error identification of Hammerstein systems.  IEEE Transactions on Automatic Control , 58 (5).  pp. 1317-1322.  ISSN 0018-9286      (2013)