DocumentCode
1194419
Title
Pole identification of ARMA processes by extended Levinson recursion
Author
Martinelli, G. ; Orlandi, G. ; Burrascano, P.
Volume
33
Issue
3
fYear
1986
fDate
3/1/1986 12:00:00 AM
Firstpage
348
Lastpage
350
Abstract
Very simple algorithms are available in the technical literature for identifying an AR process. The most famous of them is the Levinson recursion. The simplicity of this recursion is a consequence of the peculiar nature of the Yule-Walker equations which it solves. A similar situation does not exist in the case of ARMA processes, also for their AR parts, in spite of the fact that these parts satisfy the extended Yule-Walker equations. The closeness of the Yule-Walker equations to their extended version suggests the possibility of extending the Levinson procedure for determining the AR part of an ARMA process. A further motivation to this extension is represented by the results obtained in [1]. In the present letter we will show such an extension and we will illustrate the resulting algorithm by an example regarding a radar application.
Keywords
Autoregressive moving-average processes; Poles and zeros, linear systems; Acoustic signal processing; Attenuation; Digital filters; Equations; Filter bank; Frequency response; IIR filters; Passband; Signal processing algorithms; Speech processing;
fLanguage
English
Journal_Title
Circuits and Systems, IEEE Transactions on
Publisher
ieee
ISSN
0098-4094
Type
jour
DOI
10.1109/TCS.1986.1085905
Filename
1085905
Link To Document