Some Classical and Some New Ideas for Identification of Linear Systems
2013 (English)In: Journal of Control, Automation and Electrical Systems, ISSN 2195-3880, EISSN 2195-3899, Vol. 24, no 1-2, 3-10 p.Article in journal (Refereed) Published
This paper gives an overview of the identificationof linear systems. It covers the classical approach ofparametric methods by means of maximum likelihood andpredicion error methods, as well all classical non-parametricmethods through spectral analysis. It also covers very recenttechniques dealing with convex formulations by regularizationof FIR and ARX models, as well as new alternatives tospectral analysis, through local linear models. An example of identification of aircraft dynamics illustrates the approaches.
Place, publisher, year, edition, pages
Springer, 2013. Vol. 24, no 1-2, 3-10 p.
System identification, Maximimum likelihood, Prediction error methods, Spectral analysis, Regularization, Local polynomial methods
IdentifiersURN: urn:nbn:se:liu:diva-96749DOI: 10.1007/s40313-013-0004-7OAI: oai:DiVA.org:liu-96749DiVA: diva2:643191