Bias Compensated Least Squares Estimation of Continuous Time Output Error Models in the Case of Stochastic Sampling Time Jitter
2006 (English)In: Proceedings of the 14th IFAC Symposium on System Identification, 2006, 612-617 p.Conference paper (Refereed)
This work investigates how stochastic unmeasureable sampling jitternoise aﬀects the result of system identiﬁcation, and proposes a modiﬁcation ofknown approaches to mitigate the eﬀects of sampling jitter. By just assumingconventional additive measurement noise, the analysis shows that the identiﬁedmodel will get a bias in the transfer function amplitude that increases for higherfrequencies. A frequency domain approach with a continuous time system modelallows an analysis framework for sampling jitter noise. This leads to a biascompensated (weighted) least squares algorithm. A continuous time output errormodel is used for numerical illustration.
Place, publisher, year, edition, pages
2006. 612-617 p.
System identification, Stochastic systems, Least-squares estimation, Maximum likelihood, Frequency domain
Engineering and Technology Control Engineering
IdentifiersURN: urn:nbn:se:liu:diva-36628DOI: 10.3182/20060329-3-AU-2901.00094Local ID: 31929ISBN: 978-3-902661-02-9OAI: oai:DiVA.org:liu-36628DiVA: diva2:257477
14th IFAC Symposium on System Identification, Newcastle, Australia, March, 2006