Namespaces
Variants
Actions

Difference between revisions of "Wiener-Hopf equation"

From Encyclopedia of Mathematics
Jump to: navigation, search
m (tex encoded by computer)
 
Line 1: Line 1:
 +
<!--
 +
w0979001.png
 +
$#A+1 = 23 n = 0
 +
$#C+1 = 23 : ~/encyclopedia/old_files/data/W097/W.0907900 Wiener\ANDHopf equation
 +
Automatically converted into TeX, above some diagnostics.
 +
Please remove this comment and the {{TEX|auto}} line below,
 +
if TeX found to be correct.
 +
-->
 +
 +
{{TEX|auto}}
 +
{{TEX|done}}
 +
 
An integral equation on the half-line with a kernel which depends on the difference between the arguments:
 
An integral equation on the half-line with a kernel which depends on the difference between the arguments:
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w0979001.png" /></td> <td valign="top" style="width:5%;text-align:right;">(1)</td></tr></table>
+
$$ \tag{1 }
 +
u ( x) - \int\limits _ { 0 } ^  \infty  k ( x- s) u ( s)  ds  = \
 +
f ( x),\  0 \leq  x < \infty .
 +
$$
  
 
Equations of this type often appear in problems of mathematical physics, e.g. in the theory of radiative transfer (Milne's problem); in the theory of diffraction (diffraction on a half-plane, the problem of boundary refraction).
 
Equations of this type often appear in problems of mathematical physics, e.g. in the theory of radiative transfer (Milne's problem); in the theory of diffraction (diffraction on a half-plane, the problem of boundary refraction).
  
The first studies of equation (1) are due to N. Wiener and E. Hopf ([[#References|[1]]] and [[#References|[2]]]), and deal with a factorization method (see [[Wiener–Hopf method|Wiener–Hopf method]]). It was the idea of factorization which proved to be the determining factor in the construction of the theory of integral equations such as (1). V.A. Fok (also written as V.A. Fock) [[#References|[3]]] studied the Wiener–Hopf equation on the assumption that the kernel <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w0979002.png" /> is even and decreases exponentially.
+
The first studies of equation (1) are due to N. Wiener and E. Hopf ([[#References|[1]]] and [[#References|[2]]]), and deal with a factorization method (see [[Wiener–Hopf method|Wiener–Hopf method]]). It was the idea of factorization which proved to be the determining factor in the construction of the theory of integral equations such as (1). V.A. Fok (also written as V.A. Fock) [[#References|[3]]] studied the Wiener–Hopf equation on the assumption that the kernel $  k( x) $
 +
is even and decreases exponentially.
  
 
The formal scheme for solving the Wiener–Hopf equation is the following. Let
 
The formal scheme for solving the Wiener–Hopf equation is the following. Let
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w0979003.png" /></td> </tr></table>
+
$$
 +
v ( x)  = \left \{
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w0979004.png" /></td> </tr></table>
+
$$
 +
n ( x)  = \left \{
  
 
Equation (1) can then be written on the whole line as:
 
Equation (1) can then be written on the whole line as:
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w0979005.png" /></td> <td valign="top" style="width:5%;text-align:right;">(2)</td></tr></table>
+
$$ \tag{2 }
 +
v ( x) - \int\limits _ {- \infty } ^  \infty  k ( x - s) v ( x)  ds  = \
 +
f ( x) + n ( x),\ \
 +
- \infty < x < \infty .
 +
$$
  
 
If the conditions for the existence of the Fourier transforms of all functions forming part of equation (2), i.e.
 
If the conditions for the existence of the Fourier transforms of all functions forming part of equation (2), i.e.
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w0979006.png" /></td> </tr></table>
+
$$
 +
V ( \lambda )  = \
 +
\int\limits _ { 0 } ^  \infty  u ( x) e ^ {i \lambda x }  dx,\ \
 +
K ( \lambda )  = \int\limits _ {- \infty } ^  \infty  k ( x)
 +
e ^ {i \lambda x }  dx,
 +
$$
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w0979007.png" /></td> </tr></table>
+
$$
 +
F ( \lambda )  = \int\limits _ { 0 } ^  \infty  f ( x) e ^ {i \lambda x }
 +
dx,\  N ( \lambda )  = \int\limits _ {- \infty } ^ { 0 }  n ( x) e ^ {i \lambda x }  dx,
 +
$$
  
 
are met, then, using the [[Fourier transform|Fourier transform]], equation (2) is reduced to the functional equation
 
are met, then, using the [[Fourier transform|Fourier transform]], equation (2) is reduced to the functional equation
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w0979008.png" /></td> <td valign="top" style="width:5%;text-align:right;">(3)</td></tr></table>
+
$$ \tag{3 }
 +
[ 1 - K ( \lambda )] V ( \lambda )  = F ( \lambda ) + N ( \lambda ),
 +
$$
  
where <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w0979009.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w09790010.png" /> are unknown functions. The Wiener–Hopf method makes it possible to solve equation (3) for a certain class of functions. The condition <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w09790011.png" /> must be met in this context. The index of the equation,
+
where $  V ( \lambda ) $
 +
and $  N ( \lambda ) $
 +
are unknown functions. The Wiener–Hopf method makes it possible to solve equation (3) for a certain class of functions. The condition $  1 - K ( \lambda ) \neq 0 $
 +
must be met in this context. The index of the equation,
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w09790012.png" /></td> <td valign="top" style="width:5%;text-align:right;">(4)</td></tr></table>
+
$$ \tag{4 }
 +
\nu  = -  \mathop{\rm ind} [ 1 - K ( \lambda )]  = \
 +
-
 +
\frac{1}{2 \pi }
 +
\int\limits _ {- \infty } ^  \infty    d _  \lambda  [ 1 - K ( \lambda )],
 +
$$
  
plays a special role in the theory of equation (1) for a non-symmetric kernel. If <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w09790013.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w09790014.png" />, then: if <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w09790015.png" />, the inhomogeneous equation (1) has a unique solution; if <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w09790016.png" />, the homogeneous equation (1) has <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w09790017.png" /> linearly independent solutions; if <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w09790018.png" />, the inhomogeneous equation (1) has either no solution, or else has a unique solution if the following condition is met:
+
plays a special role in the theory of equation (1) for a non-symmetric kernel. If $  k \in L _ {1} (- \infty , \infty ) $
 +
and $  1 - K( \lambda ) \neq 0 $,  
 +
then: if $  \nu = 0 $,  
 +
the inhomogeneous equation (1) has a unique solution; if $  \nu > 0 $,  
 +
the homogeneous equation (1) has $  \nu $
 +
linearly independent solutions; if $  \nu < 0 $,  
 +
the inhomogeneous equation (1) has either no solution, or else has a unique solution if the following condition is met:
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w09790019.png" /></td> </tr></table>
+
$$
 +
\int\limits _ { 0 } ^  \infty  f ( x) \psi _ {k} ( x)  dx  = 0,\ \
 +
k = 0 \dots | \nu | - 1,
 +
$$
  
where <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w09790020.png" /> are the linearly independent solutions of the transposed homogeneous equation to (1):
+
where $  \psi _ {k} ( x) $
 +
are the linearly independent solutions of the transposed homogeneous equation to (1):
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w09790021.png" /></td> </tr></table>
+
$$
 +
\psi ( x) - \int\limits _ { 0 } ^  \infty  k ( x- s) \psi ( s)  ds  = 0.
 +
$$
  
 
====References====
 
====References====
 
<table><TR><TD valign="top">[1]</TD> <TD valign="top">  N. Wiener,  E. Hopf,  "Ueber eine Klasse singulärer Integralgleichungen"  ''Sitzungber. Akad. Wiss. Berlin''  (1931)  pp. 696–706</TD></TR><TR><TD valign="top">[2]</TD> <TD valign="top">  E. Hopf,  "Mathematical problems of radiative equilibrium" , Cambridge Univ. Press  (1934)</TD></TR><TR><TD valign="top">[3]</TD> <TD valign="top">  V.A. Fok,  "On some integral equations of mathematical physics"  ''Mat. Sb.'' , '''14 (56)''' :  1–2  (1944)  pp. 3–50  (In Russian)  (French abstract)</TD></TR><TR><TD valign="top">[4]</TD> <TD valign="top">  B. Noble,  "Methods based on the Wiener–Hopf technique for the solution of partial differential equations" , Pergamon  (1958)</TD></TR></table>
 
<table><TR><TD valign="top">[1]</TD> <TD valign="top">  N. Wiener,  E. Hopf,  "Ueber eine Klasse singulärer Integralgleichungen"  ''Sitzungber. Akad. Wiss. Berlin''  (1931)  pp. 696–706</TD></TR><TR><TD valign="top">[2]</TD> <TD valign="top">  E. Hopf,  "Mathematical problems of radiative equilibrium" , Cambridge Univ. Press  (1934)</TD></TR><TR><TD valign="top">[3]</TD> <TD valign="top">  V.A. Fok,  "On some integral equations of mathematical physics"  ''Mat. Sb.'' , '''14 (56)''' :  1–2  (1944)  pp. 3–50  (In Russian)  (French abstract)</TD></TR><TR><TD valign="top">[4]</TD> <TD valign="top">  B. Noble,  "Methods based on the Wiener–Hopf technique for the solution of partial differential equations" , Pergamon  (1958)</TD></TR></table>
 
 
  
 
====Comments====
 
====Comments====
The theorems about the solution of the Wiener–Hopf integral equation referred to above appeared in [[#References|[a1]]], which treats equation (1) in a number of different function spaces of Banach or Hilbert type. The matrix-valued version of the theory is due to [[#References|[a2]]]. Explicit solutions for the case when <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w09790022.png" /> is a rational matrix function may be found in [[#References|[a3]]]. For a recent exposition of the theory of Wiener–Hopf integral equations, including the Fredholm theory and the state-space method for the case of rational <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/w/w097/w097900/w09790023.png" />, see [[#References|[a4]]].
+
The theorems about the solution of the Wiener–Hopf integral equation referred to above appeared in [[#References|[a1]]], which treats equation (1) in a number of different function spaces of Banach or Hilbert type. The matrix-valued version of the theory is due to [[#References|[a2]]]. Explicit solutions for the case when $  K( \lambda ) $
 +
is a rational matrix function may be found in [[#References|[a3]]]. For a recent exposition of the theory of Wiener–Hopf integral equations, including the Fredholm theory and the state-space method for the case of rational $  K( \lambda ) $,  
 +
see [[#References|[a4]]].
  
 
====References====
 
====References====
 
<table><TR><TD valign="top">[a1]</TD> <TD valign="top">  M.G. Krein,  "Integral equations on a half-line with kernel depending upon the difference of the arguments"  ''Transl. Amer. Math. Soc. (2)'' , '''22'''  (1962)  pp. 163–288  ''Uspekhi Mat. Nauk'' , '''13''' :  5  (1958)  pp. 3–120</TD></TR><TR><TD valign="top">[a2]</TD> <TD valign="top">  I. [I.Ts. Gokhberg] Gohberg,  M.G. Krein,  "Systems of integral equations on a half line with kernels depending on the difference of arguments"  ''Transl. Amer. Math. Soc. (2)'' , '''14'''  (1960)  pp. 217–287  ''Uspekhi Mat. Nauk'' , '''13''' :  2 (80)  (1958)  pp. 3–72</TD></TR><TR><TD valign="top">[a3]</TD> <TD valign="top">  H. Bart,  I. Gohberg,  M.A. Kaashoek,  "Minimal factorization of matrix and operation functions" , Birkhäuser  (1979)</TD></TR><TR><TD valign="top">[a4]</TD> <TD valign="top">  I.C. Gohberg,  S. Goldberg,  M.A. Kaashoek,  "Classes of linear operators" , '''1''' , Birkhäuser  (1990)  pp. Chapts. XI-XII</TD></TR><TR><TD valign="top">[a5]</TD> <TD valign="top">  H. Hochstadt,  "Integral equations" , Wiley  (1973)</TD></TR></table>
 
<table><TR><TD valign="top">[a1]</TD> <TD valign="top">  M.G. Krein,  "Integral equations on a half-line with kernel depending upon the difference of the arguments"  ''Transl. Amer. Math. Soc. (2)'' , '''22'''  (1962)  pp. 163–288  ''Uspekhi Mat. Nauk'' , '''13''' :  5  (1958)  pp. 3–120</TD></TR><TR><TD valign="top">[a2]</TD> <TD valign="top">  I. [I.Ts. Gokhberg] Gohberg,  M.G. Krein,  "Systems of integral equations on a half line with kernels depending on the difference of arguments"  ''Transl. Amer. Math. Soc. (2)'' , '''14'''  (1960)  pp. 217–287  ''Uspekhi Mat. Nauk'' , '''13''' :  2 (80)  (1958)  pp. 3–72</TD></TR><TR><TD valign="top">[a3]</TD> <TD valign="top">  H. Bart,  I. Gohberg,  M.A. Kaashoek,  "Minimal factorization of matrix and operation functions" , Birkhäuser  (1979)</TD></TR><TR><TD valign="top">[a4]</TD> <TD valign="top">  I.C. Gohberg,  S. Goldberg,  M.A. Kaashoek,  "Classes of linear operators" , '''1''' , Birkhäuser  (1990)  pp. Chapts. XI-XII</TD></TR><TR><TD valign="top">[a5]</TD> <TD valign="top">  H. Hochstadt,  "Integral equations" , Wiley  (1973)</TD></TR></table>

Latest revision as of 08:29, 6 June 2020


An integral equation on the half-line with a kernel which depends on the difference between the arguments:

$$ \tag{1 } u ( x) - \int\limits _ { 0 } ^ \infty k ( x- s) u ( s) ds = \ f ( x),\ 0 \leq x < \infty . $$

Equations of this type often appear in problems of mathematical physics, e.g. in the theory of radiative transfer (Milne's problem); in the theory of diffraction (diffraction on a half-plane, the problem of boundary refraction).

The first studies of equation (1) are due to N. Wiener and E. Hopf ([1] and [2]), and deal with a factorization method (see Wiener–Hopf method). It was the idea of factorization which proved to be the determining factor in the construction of the theory of integral equations such as (1). V.A. Fok (also written as V.A. Fock) [3] studied the Wiener–Hopf equation on the assumption that the kernel $ k( x) $ is even and decreases exponentially.

The formal scheme for solving the Wiener–Hopf equation is the following. Let

$$ v ( x) = \left \{ $$ n ( x) = \left \{

Equation (1) can then be written on the whole line as:

$$ \tag{2 } v ( x) - \int\limits _ {- \infty } ^ \infty k ( x - s) v ( x) ds = \ f ( x) + n ( x),\ \ - \infty < x < \infty . $$

If the conditions for the existence of the Fourier transforms of all functions forming part of equation (2), i.e.

$$ V ( \lambda ) = \ \int\limits _ { 0 } ^ \infty u ( x) e ^ {i \lambda x } dx,\ \ K ( \lambda ) = \int\limits _ {- \infty } ^ \infty k ( x) e ^ {i \lambda x } dx, $$

$$ F ( \lambda ) = \int\limits _ { 0 } ^ \infty f ( x) e ^ {i \lambda x } dx,\ N ( \lambda ) = \int\limits _ {- \infty } ^ { 0 } n ( x) e ^ {i \lambda x } dx, $$

are met, then, using the Fourier transform, equation (2) is reduced to the functional equation

$$ \tag{3 } [ 1 - K ( \lambda )] V ( \lambda ) = F ( \lambda ) + N ( \lambda ), $$

where $ V ( \lambda ) $ and $ N ( \lambda ) $ are unknown functions. The Wiener–Hopf method makes it possible to solve equation (3) for a certain class of functions. The condition $ 1 - K ( \lambda ) \neq 0 $ must be met in this context. The index of the equation,

$$ \tag{4 } \nu = - \mathop{\rm ind} [ 1 - K ( \lambda )] = \ - \frac{1}{2 \pi } \int\limits _ {- \infty } ^ \infty d _ \lambda [ 1 - K ( \lambda )], $$

plays a special role in the theory of equation (1) for a non-symmetric kernel. If $ k \in L _ {1} (- \infty , \infty ) $ and $ 1 - K( \lambda ) \neq 0 $, then: if $ \nu = 0 $, the inhomogeneous equation (1) has a unique solution; if $ \nu > 0 $, the homogeneous equation (1) has $ \nu $ linearly independent solutions; if $ \nu < 0 $, the inhomogeneous equation (1) has either no solution, or else has a unique solution if the following condition is met:

$$ \int\limits _ { 0 } ^ \infty f ( x) \psi _ {k} ( x) dx = 0,\ \ k = 0 \dots | \nu | - 1, $$

where $ \psi _ {k} ( x) $ are the linearly independent solutions of the transposed homogeneous equation to (1):

$$ \psi ( x) - \int\limits _ { 0 } ^ \infty k ( x- s) \psi ( s) ds = 0. $$

References

[1] N. Wiener, E. Hopf, "Ueber eine Klasse singulärer Integralgleichungen" Sitzungber. Akad. Wiss. Berlin (1931) pp. 696–706
[2] E. Hopf, "Mathematical problems of radiative equilibrium" , Cambridge Univ. Press (1934)
[3] V.A. Fok, "On some integral equations of mathematical physics" Mat. Sb. , 14 (56) : 1–2 (1944) pp. 3–50 (In Russian) (French abstract)
[4] B. Noble, "Methods based on the Wiener–Hopf technique for the solution of partial differential equations" , Pergamon (1958)

Comments

The theorems about the solution of the Wiener–Hopf integral equation referred to above appeared in [a1], which treats equation (1) in a number of different function spaces of Banach or Hilbert type. The matrix-valued version of the theory is due to [a2]. Explicit solutions for the case when $ K( \lambda ) $ is a rational matrix function may be found in [a3]. For a recent exposition of the theory of Wiener–Hopf integral equations, including the Fredholm theory and the state-space method for the case of rational $ K( \lambda ) $, see [a4].

References

[a1] M.G. Krein, "Integral equations on a half-line with kernel depending upon the difference of the arguments" Transl. Amer. Math. Soc. (2) , 22 (1962) pp. 163–288 Uspekhi Mat. Nauk , 13 : 5 (1958) pp. 3–120
[a2] I. [I.Ts. Gokhberg] Gohberg, M.G. Krein, "Systems of integral equations on a half line with kernels depending on the difference of arguments" Transl. Amer. Math. Soc. (2) , 14 (1960) pp. 217–287 Uspekhi Mat. Nauk , 13 : 2 (80) (1958) pp. 3–72
[a3] H. Bart, I. Gohberg, M.A. Kaashoek, "Minimal factorization of matrix and operation functions" , Birkhäuser (1979)
[a4] I.C. Gohberg, S. Goldberg, M.A. Kaashoek, "Classes of linear operators" , 1 , Birkhäuser (1990) pp. Chapts. XI-XII
[a5] H. Hochstadt, "Integral equations" , Wiley (1973)
How to Cite This Entry:
Wiener-Hopf equation. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=Wiener-Hopf_equation&oldid=49215
This article was adapted from an original article by V.I. Dmitriev (originator), which appeared in Encyclopedia of Mathematics - ISBN 1402006098. See original article