1996
DOI: 10.1016/0165-1684(96)00096-5
|View full text |Cite
|
Sign up to set email alerts
|

A stability condition for adaptive recursive second-order polynomial filters

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
11
0

Year Published

1999
1999
2023
2023

Publication Types

Select...
6

Relationship

0
6

Authors

Journals

citations
Cited by 22 publications
(11 citation statements)
references
References 4 publications
0
11
0
Order By: Relevance
“…As noted, for example, in [19], recursive models (even in the linear case) may be unstable unless the ANC algorithm is carefully designed, by suitably setting the step size or adopting specific algorithmic modifications, such as the introduction of a 1426 D. DELVECCHIO AND L. PIRODDI leakage factor. BIBO stability conditions have been derived for recursive Volterra filters [34], bilinear models [35], second order polynomial models [36,37], recursive FLANN filters [22], essentially relying on the asymptotic stability of the linear submodel and ensuring the boundedness of the nonlinear terms. Interestingly enough, only the former assumption is necessary for recursive FLANN filters, given the specific bounded structure of the nonlinear terms [22].…”
Section: Models and Methods For Nanc In The Linear Sp Casementioning
confidence: 99%
“…As noted, for example, in [19], recursive models (even in the linear case) may be unstable unless the ANC algorithm is carefully designed, by suitably setting the step size or adopting specific algorithmic modifications, such as the introduction of a 1426 D. DELVECCHIO AND L. PIRODDI leakage factor. BIBO stability conditions have been derived for recursive Volterra filters [34], bilinear models [35], second order polynomial models [36,37], recursive FLANN filters [22], essentially relying on the asymptotic stability of the linear submodel and ensuring the boundedness of the nonlinear terms. Interestingly enough, only the former assumption is necessary for recursive FLANN filters, given the specific bounded structure of the nonlinear terms [22].…”
Section: Models and Methods For Nanc In The Linear Sp Casementioning
confidence: 99%
“…[32], a sufficient, but not necessary, condition to ensure a bounded output requires x(n) to be bounded. Therefore, in nonlinear system identification experiments, the Gaussian input signal is zero-mean Gaussian signal.…”
Section: Nonlinear System Identificationmentioning
confidence: 99%
“…(10) By calling the term on the right of (10), and introducing the 's defined above, we have (11) Since the poles 's are inside the unit circle, by induction on the order of the system, one can easily show that this last expression can also be written as follows [9]:…”
Section: Sufficient Stability Condition For General Order Recursimentioning
confidence: 99%
“…In [9], the authors of this letter presented simple sufficient stability conditions for recursive quadratic Volterra filters. In this letter, such stability conditions are extended to the more general recursive Volterra systems of arbitrary order, i.e., the polynomial systems described by the following input-output relationship: (1) It is shown in this letter that, provided that the stability of the linear part is guaranteed, the output of the filters in (1) is bounded for every if the input signal is bounded by a certain value which can be very efficiently computed from the filter coefficients.…”
Section: Introductionmentioning
confidence: 99%