Jump to content

Wiener series: Difference between revisions

From Wikipedia, the free encyclopedia
Content deleted Content added
Rocapp (talk | contribs)
Added an explanation for the connection between the Wiener Series and the Wiener Filter, which are two different but related concepts.
Tags: Mobile edit Mobile web edit
Citation bot (talk | contribs)
Add: pmid, bibcode. | Use this bot. Report bugs. | Suggested by Abductive | Category:Mathematical series | #UCB_Category 15/84
 
(18 intermediate revisions by 12 users not shown)
Line 1: Line 1:
In mathematics, the '''Wiener series''' (or ''Wiener G-functional expansion'') originates from the 1958 book of [[Norbert Wiener]]. It is an orthogonal expansion for nonlinear [[functional (mathematics)|functionals]] closely related to the [[Volterra series]] and having the same relation to it as an orthogonal Hermite polynomial expansion has to a power series. The analogue of the coefficients are referred to as '''Wiener kernels'''. The terms of the series are orthogonal (uncorrelated) with respect to a statistical input of [[white noise]]. This property allows the terms to be identified in applications by the ''Lee-Schetzen method''.
In mathematics, the '''Wiener series''', or '''Wiener G-functional expansion''', originates from the 1958 book of [[Norbert Wiener]]. It is an orthogonal expansion for nonlinear [[functional (mathematics)|functionals]] closely related to the [[Volterra series]] and having the same relation to it as an orthogonal [[Hermite polynomials|Hermite polynomial]] expansion has to a [[power series]]. For this reason it is also known as the '''Wiener–Hermite expansion'''. The analogue of the coefficients are referred to as '''Wiener kernels'''. The terms of the series are orthogonal (uncorrelated) with respect to a statistical input of [[white noise]]. This property allows the terms to be identified in applications by the ''Lee–Schetzen method''.


The Wiener series is important in [[system identification|nonlinear system identification]]. In this context, the series approximates the functional relation of the output to the entire history of system input at any time. The Wiener series has been applied mostly to the identification of biological systems, especially in neuroscience.
The Wiener series is important in [[system identification|nonlinear system identification]]. In this context, the series approximates the functional relation of the output to the entire history of system input at any time. The Wiener series has been applied mostly to the identification of biological systems, especially in [[neuroscience]].


The name Wiener series is almost exclusively used in system theory. In the mathematical literature it occurs as the Ito expansion (1951) which is entirely equivalent to it. (The Wiener series should not be confused with the [[Wiener filter]], which is another algorithm developed by Norbert Wiener that has found use in signal processing).
The name Wiener series is almost exclusively used in [[system theory]]. In the mathematical literature it occurs as the Itô expansion (1951) which has a different form but is entirely equivalent to it.
The Wiener series should not be confused with the [[Wiener filter]], which is another algorithm developed by Norbert Wiener used in signal processing.


==Wiener G-functional expressions==
==Wiener G-functional expressions==
Given a system with an input/output pair <math>(x(t),y(t))</math> where the input is white noise with zero mean value and power A, we can write the output of the system as sum of a series of Wiener G-functionals
Given a system with an input/output pair <math>(x(t),y(t))</math> where the input is white noise with zero mean value and power A, we can write the output of the system as sum of a series of Wiener G-functionals
<math>
<math>
y(n) = \sum_{p}{(G_p x)(n)}
y(n) = \sum_p (G_p x)(n)
</math>
</math>


In the following the expressions of the G-functionals up to the fifth order will be given:
In the following the expressions of the G-functionals up to the fifth order will be given:{{clarify|reason = what are the functions k_n? This seems to be the first time they are introduced. Are they the 'Wiener kernels' mentioned earlier?|date=May 2023}}
<nowiki>{{Clarify}}</nowiki>

<math>
: <math>
(G_0 x)(n) = k_0 = E\left\{ {y(n)} \right\};
(G_0 x)(n) = k_0 = E\left\{ y(n) \right\};
</math>
</math>


<math>
: <math>
(G_1 x)(n) = \sum_{\tau _1 = 0}^{N_1 - 1} {k_1 (\tau _1 )x(n - \tau _1 )};
(G_1 x)(n) = \sum_{\tau _1 = 0}^{N_1 - 1} k_1 (\tau _1 )x(n - \tau _1 );
</math>
</math>


<math>
: <math>
(G_2 x)(n) = \sum_{\tau _1, \tau_2 = 0}^{N_2 - 1} {k_2 (\tau _1 ,\tau _2 )x(n - \tau _1 )x(n - \tau _2)} - A\sum_{\tau _1 = 0}^{N_2 - 1} {k_2 (\tau _1 ,\tau _1 )};
(G_2 x)(n) = \sum_{\tau _1, \tau_2 = 0}^{N_2 - 1} k_2 (\tau _1 ,\tau _2 )x(n - \tau _1 )x(n - \tau _2) - A\sum_{\tau _1 = 0}^{N_2 - 1} k_2 (\tau _1 ,\tau _1 );
</math>
</math>


<math>
: <math>
(G_3 x)(n) = \sum\limits_{\tau _1,\ldots,\tau_3 = 0}^{N_3 - 1} {k_3 (\tau _1 ,\tau _2 ,\tau _3 ) x(n - \tau _1 )x(n - \tau _2)x(n - \tau _3)}
(G_3 x)(n) = \sum_{\tau _1,\ldots,\tau_3 = 0}^{N_3 - 1} k_3 (\tau _1 ,\tau _2 ,\tau _3 ) x(n - \tau _1 )x(n - \tau _2)x(n - \tau _3)
- 3A \sum_{\tau _1 = 0}^{N_3 - 1} \sum_{\tau _2 = 0}^{N_3 - 1}k_3 (\tau _1 ,\tau _2 ,\tau _2 ) x(n - \tau _1 );
- 3A \sum_{\tau _1 = 0}^{N_3 - 1} \sum_{\tau _2 = 0}^{N_3 - 1}k_3 (\tau _1 ,\tau _2 ,\tau _2 ) x(n - \tau _1 );
</math>
</math>


<math>
: <math>
\begin{align}
(G_4 x)(n) = \sum\limits_{\tau _1,\ldots,\tau_4 = 0}^{N_4 - 1} k_4 (\tau _1 ,\tau _2 ,\tau _3 ,\tau _4 )
(G_4 x)(n) = {} & \sum_{\tau_1,\ldots,\tau_4 = 0}^{N_4 - 1} k_4 (\tau_1 ,\tau_2 ,\tau_3 ,\tau_4 )
x(n - \tau _1 )x(n - \tau _2 )x(n - \tau _3 )x(n - \tau _4 ) +
x(n - \tau_1 )x(n - \tau_2 )x(n - \tau_3 )x(n - \tau_4 ) + {} \\[6pt]
& {} - 6A \sum_{\tau _1,\tau _2 = 0}^{N_4 - 1} \sum_{\tau_3 = 0}^{N_4 - 1} k_4 (\tau_1, \tau_2, \tau_3 ,\tau_3) x(n - \tau_1 )x(n - \tau_2) + 3A^2 \sum_{\tau_1,\tau_2 = 0}^{N_4 - 1} k_4 (\tau_1 ,\tau_1 ,\tau_2 ,\tau_2 ) ;
\end{align}
</math>
</math>


<math>
: <math>
\begin{align}
- 6A \sum\limits_{\tau _1,\tau _2 = 0}^{N_4 - 1}\sum\limits_{\tau _3 = 0}^{N_4 - 1}{k_4 (\tau _1 ,\tau _2 ,\tau _3 ,\tau _3 ) x(n - \tau _1 )x(n - \tau _2 )}
+ 3A^2 \sum\limits_{\tau _1,\tau _2 = 0}^{N_4 - 1} {k_4 (\tau _1 ,\tau _1 ,\tau _2 ,\tau _2 ) };
(G_5 x)(n) = {} & \sum_{\tau _1,\ldots,\tau _5 = 0}^{N_5 - 1} k_5 (\tau_1, \tau_2, \tau_3, \tau_4, \tau_5 )
x(n - \tau_1 )x(n - \tau_2 )x(n - \tau_3 )x(n - \tau_4 )x(n - \tau_5 ) + {} \\[6pt]
</math>
& {} - 10A\sum_{\tau _1,\ldots,\tau _3 = 0}^{N_5 - 1} \sum_{\tau _4 = 0}^{N_5 - 1} k_5 (\tau_1, \tau _2 ,\tau_3, \tau_4, \tau_4 ) x(n - \tau_1 )x(n - \tau_2 )x(n - \tau_3 ) \\[6pt]

& {} + 15A^2 \sum_{\tau _1 = 0}^{N_5 - 1} \sum_{\tau_2,\tau_3 = 0}^{N_5 - 1} k_5 (\tau_1, \tau_2, \tau_2 ,\tau_3 ,\tau_3 ) x(n - \tau_1 ).
<math>
\end{align}
(G_5 x)(n) = \sum\limits_{\tau _1\ldots\tau _5 = 0}^{N_5 - 1} k_5 (\tau _1 ,\tau _2 ,\tau _3 ,\tau _4 ,\tau _5 )
x(n - \tau _1 )x(n - \tau _2 )x(n - \tau _3 )x(n - \tau _4 )x(n - \tau _5 ) +
</math>

<math>
- 10A\sum\limits_{\tau _1,\ldots\tau _3 = 0}^{N_5 - 1} \sum\limits_{\tau _4 = 0}^{N_5 - 1} k_5 (\tau _1 ,\tau _2 ,\tau _3 ,\tau _4 ,\tau _4 ) x(n - \tau _1 )x(n - \tau _2 )x(n - \tau _3 )
</math>

<math>
+ 15A^2 \sum\limits_{\tau _1 = 0}^{N_5 - 1} \sum\limits_{\tau _2,\tau _3 = 0}^{N_5 - 1} k_5 (\tau _1 ,\tau _2 ,\tau _2 ,\tau _3 ,\tau _3 ) x(n - \tau _1 ).
</math>
</math>


Line 60: Line 56:
==References==
==References==
* {{cite book |title=Nonlinear Problems in Random Theory |first=Norbert |last=Wiener |year=1958 |publisher= Wiley and MIT Press}}
* {{cite book |title=Nonlinear Problems in Random Theory |first=Norbert |last=Wiener |year=1958 |publisher= Wiley and MIT Press}}
* {{cite journal |doi=10.1080/00207176508905543 |title=Measurement of the Wiener kernels of a non-linear system by cross-correlation |author=Lee and Schetzen |journal=International Journal of Control |series=First |volume=2 |pages=237&ndash;254 |year=1965 |last2=Schetzen‡ |first2=M. |issue=3}}
* {{cite journal |doi=10.1080/00207176508905543 |title=Measurement of the Wiener kernels of a non-linear system by cross-correlation |author=Lee and Schetzen |journal=International Journal of Control |series=First |volume=2 |pages=237&ndash;254 |year=1965 |last2=Schetzen‡ |first2=M. |issue=3}}
* Itô K "A multiple Wiener integral" J. Math. Soc. Jpn. 3 1951 157–169
* {{cite journal |doi=10.1126/science.175.4027.1276 |title=White-noise analysis of a neuron chain: an application of the Wiener theory |last=Marmarelis |first=P.Z. |author2=Naka, K. |journal=[[Science (journal)|Science]] |volume=175 |pages=1276&ndash;1278 |year=1972 |pmid=5061252 |issue=4027}}
* {{cite journal |doi=10.1126/science.175.4027.1276 |title=White-noise analysis of a neuron chain: an application of the Wiener theory |last=Marmarelis |first=P.Z. |author2=Naka, K. |journal=[[Science (journal)|Science]] |volume=175 |pages=1276&ndash;1278 |year=1972 |pmid=5061252 |issue=4027|bibcode=1972Sci...175.1276M }}
* {{cite book |title=The Volterra and Wiener Theories of Nonlinear Systems |first=Martin |last=Schetzen |year=1980 |isbn=978-0-471-04455-0 |publisher=John Wiley and Sons}}
* {{cite book |title=The Volterra and Wiener Theories of Nonlinear Systems |first=Martin |last=Schetzen |year=1980 |isbn=978-0-471-04455-0 |publisher=John Wiley and Sons}}
* {{cite journal |title=Wiener Analysis of Nonlinear Feedback |last=Marmarelis |first=P.Z. |journal=Sensory Systems Annals of Biomedical Engineering |volume=19 |pages=345&ndash;382 |year=1991}}
* {{cite journal |title=Wiener Analysis of Nonlinear Feedback |last=Marmarelis |first=P.Z. |journal=Sensory Systems Annals of Biomedical Engineering |volume=19 |issue=4 |pages=345&ndash;382 |year=1991|doi=10.1007/BF02584316 |pmid=1741522 }}
* {{cite journal |doi=10.1162/neco.2006.18.12.3097 |title=A unifying view of Wiener and Volterra theory and polynomial kernel regression |last=Franz |first=M |author2=Schölkopf, B. |journal=[[Neural Computation (journal)|Neural Computation]]|volume=18 |pages=3097&ndash;3118 |year=2006 |issue=12}}
* {{cite journal |doi=10.1162/neco.2006.18.12.3097 |title=A unifying view of Wiener and Volterra theory and polynomial kernel regression |last=Franz |first=M |author2=Schölkopf, B. |journal=[[Neural Computation (journal)|Neural Computation]]|volume=18 |pages=3097&ndash;3118 |year=2006 |issue=12|pmid=17052160 }}
* L.A. Zadeh On the representation of nonlinear operators. IRE Westcon Conv. Record pt.2 1957 105-113.


[[Category:Mathematical series]]
[[Category:Mathematical series]]

Latest revision as of 21:45, 11 July 2024

In mathematics, the Wiener series, or Wiener G-functional expansion, originates from the 1958 book of Norbert Wiener. It is an orthogonal expansion for nonlinear functionals closely related to the Volterra series and having the same relation to it as an orthogonal Hermite polynomial expansion has to a power series. For this reason it is also known as the Wiener–Hermite expansion. The analogue of the coefficients are referred to as Wiener kernels. The terms of the series are orthogonal (uncorrelated) with respect to a statistical input of white noise. This property allows the terms to be identified in applications by the Lee–Schetzen method.

The Wiener series is important in nonlinear system identification. In this context, the series approximates the functional relation of the output to the entire history of system input at any time. The Wiener series has been applied mostly to the identification of biological systems, especially in neuroscience.

The name Wiener series is almost exclusively used in system theory. In the mathematical literature it occurs as the Itô expansion (1951) which has a different form but is entirely equivalent to it.

The Wiener series should not be confused with the Wiener filter, which is another algorithm developed by Norbert Wiener used in signal processing.

Wiener G-functional expressions

[edit]

Given a system with an input/output pair where the input is white noise with zero mean value and power A, we can write the output of the system as sum of a series of Wiener G-functionals

In the following the expressions of the G-functionals up to the fifth order will be given:[clarification needed]

{{Clarify}}

See also

[edit]

References

[edit]
  • Wiener, Norbert (1958). Nonlinear Problems in Random Theory. Wiley and MIT Press.
  • Lee and Schetzen; Schetzen‡, M. (1965). "Measurement of the Wiener kernels of a non-linear system by cross-correlation". International Journal of Control. First. 2 (3): 237–254. doi:10.1080/00207176508905543.
  • Itô K "A multiple Wiener integral" J. Math. Soc. Jpn. 3 1951 157–169
  • Marmarelis, P.Z.; Naka, K. (1972). "White-noise analysis of a neuron chain: an application of the Wiener theory". Science. 175 (4027): 1276–1278. Bibcode:1972Sci...175.1276M. doi:10.1126/science.175.4027.1276. PMID 5061252.
  • Schetzen, Martin (1980). The Volterra and Wiener Theories of Nonlinear Systems. John Wiley and Sons. ISBN 978-0-471-04455-0.
  • Marmarelis, P.Z. (1991). "Wiener Analysis of Nonlinear Feedback". Sensory Systems Annals of Biomedical Engineering. 19 (4): 345–382. doi:10.1007/BF02584316. PMID 1741522.
  • Franz, M; Schölkopf, B. (2006). "A unifying view of Wiener and Volterra theory and polynomial kernel regression". Neural Computation. 18 (12): 3097–3118. doi:10.1162/neco.2006.18.12.3097. PMID 17052160.
  • L.A. Zadeh On the representation of nonlinear operators. IRE Westcon Conv. Record pt.2 1957 105-113.