The problem of identification of a linear nonsta-tionary stochastic process is considered and solved using theapproach based on functional series approximation of time-varying parameter trajectories. The proposed fast basis func-tion estimators are computationally attractive and yield resultsthat are better than those provided by the local least squaresalgorithms. It is shown that two important design parameters –the number of basis functions and the size of the local analysisinterval – can be selected on-line in an adaptive way.
Authors
Additional information
- DOI
- Digital Object Identifier link open in new tab 10.23919/eusipco.2019.8902739
- Category
- Aktywność konferencyjna
- Type
- publikacja w wydawnictwie zbiorowym recenzowanym (także w materiałach konferencyjnych)
- Language
- angielski
- Publication year
- 2019
