Namespaces
Variants
Actions

Difference between revisions of "Malgrange-Ehrenpreis theorem"

From Encyclopedia of Mathematics
Jump to: navigation, search
m (AUTOMATIC EDIT (latexlist): Replaced 66 formulas out of 66 by TEX code with an average confidence of 2.0 and a minimal confidence of 2.0.)
Line 1: Line 1:
Briefly said: Every constant-coefficients [[Linear partial differential equation|linear partial differential equation]] on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m1200901.png" /> can be solved.
+
<!--This article has been texified automatically. Since there was no Nroff source code for this article,
 +
the semi-automatic procedure described at https://encyclopediaofmath.org/wiki/User:Maximilian_Janisch/latexlist
 +
was used.
 +
If the TeX and formula formatting is correct, please remove this message and the {{TEX|semi-auto}} category.
  
Let <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m1200902.png" /> be a [[Polynomial|polynomial]] in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m1200903.png" /> variables, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m1200904.png" />, where the sum is a finite one, each <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m1200905.png" /> is a multi-index <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m1200906.png" />, and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m1200907.png" />. To <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m1200908.png" /> one associates a constant-coefficient linear partial differential operator <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m1200909.png" />, obtained by replacing <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009010.png" /> by <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009011.png" /> (<img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009012.png" />).
+
Out of 66 formulas, 66 were replaced by TEX code.-->
  
Although several special cases had been known for a long time, it is only in the mid 1950s that it has been shown that for every non-zero constant-coefficient linear partial differential operator the equation <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009013.png" /> can be solved for all functions or distributions (cf. also [[Generalized function|Generalized function]]) <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009014.png" /> on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009015.png" />. In particular, there is a distribution <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009016.png" /> such that <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009017.png" />, where <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009018.png" /> is the Dirac mass at <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009019.png" /> (cf. also [[Dirac delta-function|Dirac delta-function]]). Such an <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009020.png" /> is called a fundamental solution. This is the celebrated Malgrange–Ehrenpreis theorem (established independently by B. Malgrange [[#References|[a8]]] and L. Ehrenpreis [[#References|[a1]]]). In case of the [[Laplace operator|Laplace operator]], the [[Newton potential|Newton potential]] (up to some factor) is a fundamental solution.
+
{{TEX|semi-auto}}{{TEX|done}}
 +
Briefly said: Every constant-coefficients [[Linear partial differential equation|linear partial differential equation]] on ${\bf R} ^ { n }$ can be solved.
  
If <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009021.png" /> is any distribution on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009022.png" /> with compact support, by using a fundamental solution <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009023.png" /> it is immediate to solve the equation <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009024.png" />. Indeed (<img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009025.png" /> denoting convolution, cf. also [[Convolution of functions|Convolution of functions]]),
+
Let $P$ be a [[Polynomial|polynomial]] in $n$ variables, $P ( \xi ) = \sum _ { J } a _ { J } \xi ^ { J }$, where the sum is a finite one, each $J$ is a multi-index $J = ( j _ { 1 } , \ldots , j _ { n } ) \in \mathbf{N} ^ { n }$, and $\xi ^ { J } = \xi _ { 1 } ^ { j_1 } \ldots \xi _ { n } ^ { j_n }$. To $P$ one associates a constant-coefficient linear partial differential operator $P ( D )$, obtained by replacing $\xi_j$ by $- i \partial / \partial x _ { j }$ ($i = \sqrt { - 1 }$).
  
<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/m/m120/m120090/m12009026.png" /></td> </tr></table>
+
Although several special cases had been known for a long time, it is only in the mid 1950s that it has been shown that for every non-zero constant-coefficient linear partial differential operator the equation $P ( D ) ( u ) = g$ can be solved for all functions or distributions (cf. also [[Generalized function|Generalized function]]) $g$ on ${\bf R} ^ { n }$. In particular, there is a distribution $E$ such that $P ( D ) ( E ) = \delta _ { 0 }$, where $\delta _ { 0 }$ is the Dirac mass at $0$ (cf. also [[Dirac delta-function|Dirac delta-function]]). Such an $E$ is called a fundamental solution. This is the celebrated Malgrange–Ehrenpreis theorem (established independently by B. Malgrange [[#References|[a8]]] and L. Ehrenpreis [[#References|[a1]]]). In case of the [[Laplace operator|Laplace operator]], the [[Newton potential|Newton potential]] (up to some factor) is a fundamental solution.
  
The Malgrange–Ehrenpreis theory goes further. One can solve the equation for non-compactly supported right-hand side <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009027.png" /> by exhausting <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009028.png" /> by balls, and by solving the equation on each ball. The process is made to converge by corrections based on the approximation of local solutions to the homogeneous equation <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009029.png" /> by global solutions. The space of solutions to a linear homogeneous ordinary differential equation, with constant coefficients (for example, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009030.png" />), is spanned by exponential solutions <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009031.png" /> (in the example, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009032.png" /> is a root of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009033.png" />), or by exponential and possibly polynomial-exponential solutions (<img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009034.png" /> in the case of a double root). Similarly, the solutions to <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009035.png" /> can be approximated, on convex sets, by linear combinations of polynomial-exponential solutions, functions of the type <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009036.png" />, where <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009037.png" /> is a polynomial and necessarily <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009038.png" />. See [[#References|[a2]]], [[#References|[a8]]], [[#References|[a5]]], (7.3.6). More generally, convolution equations can be considered.
+
If $g$ is any distribution on ${\bf R} ^ { n }$ with compact support, by using a fundamental solution $E$ it is immediate to solve the equation $P ( D ) ( u ) = g$. Indeed ($*$ denoting convolution, cf. also [[Convolution of functions|Convolution of functions]]),
  
Most often, the Malgrange–Ehrenpreis theorem is attacked by using the [[Fourier transform|Fourier transform]]. If <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009039.png" /> is a <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009040.png" />-function on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009041.png" />, say with compact support, its Fourier transform is defined by
+
\begin{equation*} P ( D ) ( E ^ { * } g ) = ( P ( D ) ( E ) ) ^ { * } g = \delta _ { 0 } * g = g. \end{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/m/m120/m120090/m12009042.png" /></td> </tr></table>
+
The Malgrange–Ehrenpreis theory goes further. One can solve the equation for non-compactly supported right-hand side $g$ by exhausting ${\bf R} ^ { n }$ by balls, and by solving the equation on each ball. The process is made to converge by corrections based on the approximation of local solutions to the homogeneous equation $P ( D ) u = 0$ by global solutions. The space of solutions to a linear homogeneous ordinary differential equation, with constant coefficients (for example, $y ^ { \prime \prime } + b y ^ { \prime } + c y = 0$), is spanned by exponential solutions $x \mapsto e ^ { r x }$ (in the example, $r$ is a root of $r ^ { 2 } + b r + c = 0$), or by exponential and possibly polynomial-exponential solutions ($x e ^ { rx }$ in the case of a double root). Similarly, the solutions to $P ( D ) u = 0$ can be approximated, on convex sets, by linear combinations of polynomial-exponential solutions, functions of the type $Q ( x ) e ^ { i \xi x }$, where $Q$ is a polynomial and necessarily $P ( \xi ) = 0$. See [[#References|[a2]]], [[#References|[a8]]], [[#References|[a5]]], (7.3.6). More generally, convolution equations can be considered.
  
and one has <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009043.png" />. Looking for a fundamental solution basically consists in solving the equation <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009044.png" />. This is of course totally elementary in case the function <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009045.png" /> is integrable on <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009046.png" />, since then, by inverse Fourier transformation, one can define (classically) <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009047.png" /> by
+
Most often, the Malgrange–Ehrenpreis theorem is attacked by using the [[Fourier transform|Fourier transform]]. If $\phi$ is a $C ^ { \infty }$-function on ${\bf R} ^ { n }$, say with compact support, its Fourier transform is defined by
  
<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/m/m120/m120090/m12009048.png" /></td> </tr></table>
+
\begin{equation*} \hat { \phi } ( \xi ) = \int _ { \mathbf{R} ^ { n } } \phi ( x ) e ^ { - i \xi x } d x, \end{equation*}
  
(For example, with <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009049.png" /> denoting the usual Laplacian, the operator <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009050.png" />, for <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009051.png" />, for which <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009052.png" />.)
+
and one has $( P ( D ) ( \phi ) )_{ \widehat{}} ( \xi ) = P ( \xi ) \widehat { \phi } ( \xi )$. Looking for a fundamental solution basically consists in solving the equation $\hat { E } = 1 / P ( \xi )$. This is of course totally elementary in case the function $1 / P ( \xi )$ is integrable on ${\bf R} ^ { n }$, since then, by inverse Fourier transformation, one can define (classically) $E$ by
  
In general, one has to face two problems: the lack of integrability at infinity of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009053.png" />, and the real zeros of the polynomial <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009054.png" />. Roughly speaking, since the two problems are not disjoint, one overcomes the first difficulty by the calculus of distributions, and the second one by switching to the complex domain. Both Malgrange and Ehrenpreis used the [[Hahn–Banach theorem|Hahn–Banach theorem]]. By now (1998) things have been much simplified, and a quick  "constructive"  proof of the Malgrange–Ehrenpreis theorem can be found in [[#References|[a5]]], (7.3.10), see also [[#References|[a3]]], (1.56). A formula claimed to be more explicit was recently given by H. König [[#References|[a7]]].
+
\begin{equation*} E ( x ) = \frac { 1 } { ( 2 \pi ) ^ { n } } \int _ { \mathbf{R} ^ { n } } \frac { 1 } { P ( \xi ) } e ^ { i \xi x } d \xi . \end{equation*}
  
The solvability of constant-coefficients linear partial differential operators can also be studied with pure <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009055.png" /> methods. There is a remarkable fundamental result by L.V. Hörmander [[#References|[a4]]], (2.6), whose proof is totally elementary, brief, and a masterpiece:
+
(For example, with $\Delta$ denoting the usual Laplacian, the operator $P ( D ) = I + ( - \Delta ) ^ { N }$, for $N &gt; n / 2$, for which $P ( \xi ) = 1 + | \xi | ^ { 2 N }$.)
  
If <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009056.png" /> is a bounded set in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009057.png" />, and if <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009058.png" /> is non-zero, there exists a constant <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009059.png" /> such that for every <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009060.png" />-function <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009061.png" /> with compact support in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009062.png" />,
+
In general, one has to face two problems: the lack of integrability at infinity of $1 / P ( \xi )$, and the real zeros of the polynomial $P$. Roughly speaking, since the two problems are not disjoint, one overcomes the first difficulty by the calculus of distributions, and the second one by switching to the complex domain. Both Malgrange and Ehrenpreis used the [[Hahn–Banach theorem|Hahn–Banach theorem]]. By now (1998) things have been much simplified, and a quick  "constructive" proof of the Malgrange–Ehrenpreis theorem can be found in [[#References|[a5]]], (7.3.10), see also [[#References|[a3]]], (1.56). A formula claimed to be more explicit was recently given by H. König [[#References|[a7]]].
  
<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/m/m120/m120090/m12009063.png" /></td> </tr></table>
+
The solvability of constant-coefficients linear partial differential operators can also be studied with pure $L^{2}$ methods. There is a remarkable fundamental result by L.V. Hörmander [[#References|[a4]]], (2.6), whose proof is totally elementary, brief, and a masterpiece:
  
It then follows, from elementary [[Hilbert space|Hilbert space]] theory, that the adjoint operator <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009064.png" /> (associated to the polynomial <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009065.png" />) is solvable in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/m/m120/m120090/m12009066.png" />. From there, one can recover the Malgrange–Ehrenpreis theorem. See [[#References|[a10]]], or [[#References|[a9]]] for an elementary exposition. However, the problem being not only to show the existence of a fundamental solution, but to get solutions with good properties, the construction in [[#References|[a5]]], Chap. 7, seems close to being optimal, see [[#References|[a6]]], Chap. 10.
+
If $\Omega$ is a bounded set in ${\bf R} ^ { n }$, and if $P ( D )$ is non-zero, there exists a constant $C &gt; 0$ such that for every $C ^ { \infty }$-function $\phi$ with compact support in $\Omega$,
 +
 
 +
\begin{equation*} \| P ( D ) ( \phi ) \| _ { 2 } \geq C \| \phi \| _ { 2 } ( L ^ { 2 } \text { norms } ). \end{equation*}
 +
 
 +
It then follows, from elementary [[Hilbert space|Hilbert space]] theory, that the adjoint operator $P ^ { * } ( D )$ (associated to the polynomial $\overline { P ( - \xi ) }$) is solvable in $L ^ { 2 } ( \Omega )$. From there, one can recover the Malgrange–Ehrenpreis theorem. See [[#References|[a10]]], or [[#References|[a9]]] for an elementary exposition. However, the problem being not only to show the existence of a fundamental solution, but to get solutions with good properties, the construction in [[#References|[a5]]], Chap. 7, seems close to being optimal, see [[#References|[a6]]], Chap. 10.
  
 
The situation is radically different for operators with variable coefficients (see [[Lewy operator and Mizohata operator|Lewy operator and Mizohata operator]]).
 
The situation is radically different for operators with variable coefficients (see [[Lewy operator and Mizohata operator|Lewy operator and Mizohata operator]]).
  
 
====References====
 
====References====
<table><TR><TD valign="top">[a1]</TD> <TD valign="top">  L. Ehrenpreis,  "Solutions of some problems of division I"  ''Amer. J. Math.'' , '''76'''  (1954)  pp. 883–903</TD></TR><TR><TD valign="top">[a2]</TD> <TD valign="top">  L. Ehrenpreis,  "Solutions of some problems of division II"  ''Amer. J. Math.'' , '''78'''  (1956)  pp. 685–715</TD></TR><TR><TD valign="top">[a3]</TD> <TD valign="top">  G.B. Folland,  "Introduction to partial differential equations" , Princeton Univ. Press  (1995)</TD></TR><TR><TD valign="top">[a4]</TD> <TD valign="top">  L. Hörmander,  "On the theory of general partial differential operators"  ''Acta Math.'' , '''94'''  (1955)  pp. 161–258</TD></TR><TR><TD valign="top">[a5]</TD> <TD valign="top">  L. Hörmander,  "The analysis of linear partial differential operators I" , ''Grundl. Math. Wissenschaft.'' , '''256''' , Springer  (1983)</TD></TR><TR><TD valign="top">[a6]</TD> <TD valign="top">  L. Hörmander,  "The analysis of linear partial differential operators II" , ''Grundl. Math. Wissenschaft.'' , '''257''' , Springer  (1983)</TD></TR><TR><TD valign="top">[a7]</TD> <TD valign="top">  H. König,  "An explicit formula for fundamental solutions of linear partial differential equations with constant coefficients"  ''Proc. Amer. Math. Soc.'' , '''120'''  (1994)  pp. 1315–1318</TD></TR><TR><TD valign="top">[a8]</TD> <TD valign="top">  B. Malgrange,  "Existence et approximation des solutions des équations aux dérivées partielles et des équations de convolution"  ''Ann. Inst. Fourier (Grenoble)'' , '''6'''  (1955/6)  pp. 271–355</TD></TR><TR><TD valign="top">[a9]</TD> <TD valign="top">  J-P. Rosay,  "A very elementary proof of the Malgrange–Ehrenpreis theorem"  ''Amer. Math. Monthly'' , '''98'''  (1991)  pp. 518–523</TD></TR><TR><TD valign="top">[a10]</TD> <TD valign="top">  F. Treves,  "Thèse d'Hörmander"  ''Sém. Bourbaki'' , '''Exp. 130'''  (1956)</TD></TR></table>
+
<table><tr><td valign="top">[a1]</td> <td valign="top">  L. Ehrenpreis,  "Solutions of some problems of division I"  ''Amer. J. Math.'' , '''76'''  (1954)  pp. 883–903</td></tr><tr><td valign="top">[a2]</td> <td valign="top">  L. Ehrenpreis,  "Solutions of some problems of division II"  ''Amer. J. Math.'' , '''78'''  (1956)  pp. 685–715</td></tr><tr><td valign="top">[a3]</td> <td valign="top">  G.B. Folland,  "Introduction to partial differential equations" , Princeton Univ. Press  (1995)</td></tr><tr><td valign="top">[a4]</td> <td valign="top">  L. Hörmander,  "On the theory of general partial differential operators"  ''Acta Math.'' , '''94'''  (1955)  pp. 161–258</td></tr><tr><td valign="top">[a5]</td> <td valign="top">  L. Hörmander,  "The analysis of linear partial differential operators I" , ''Grundl. Math. Wissenschaft.'' , '''256''' , Springer  (1983)</td></tr><tr><td valign="top">[a6]</td> <td valign="top">  L. Hörmander,  "The analysis of linear partial differential operators II" , ''Grundl. Math. Wissenschaft.'' , '''257''' , Springer  (1983)</td></tr><tr><td valign="top">[a7]</td> <td valign="top">  H. König,  "An explicit formula for fundamental solutions of linear partial differential equations with constant coefficients"  ''Proc. Amer. Math. Soc.'' , '''120'''  (1994)  pp. 1315–1318</td></tr><tr><td valign="top">[a8]</td> <td valign="top">  B. Malgrange,  "Existence et approximation des solutions des équations aux dérivées partielles et des équations de convolution"  ''Ann. Inst. Fourier (Grenoble)'' , '''6'''  (1955/6)  pp. 271–355</td></tr><tr><td valign="top">[a9]</td> <td valign="top">  J-P. Rosay,  "A very elementary proof of the Malgrange–Ehrenpreis theorem"  ''Amer. Math. Monthly'' , '''98'''  (1991)  pp. 518–523</td></tr><tr><td valign="top">[a10]</td> <td valign="top">  F. Treves,  "Thèse d'Hörmander"  ''Sém. Bourbaki'' , '''Exp. 130'''  (1956)</td></tr></table>

Revision as of 17:00, 1 July 2020

Briefly said: Every constant-coefficients linear partial differential equation on ${\bf R} ^ { n }$ can be solved.

Let $P$ be a polynomial in $n$ variables, $P ( \xi ) = \sum _ { J } a _ { J } \xi ^ { J }$, where the sum is a finite one, each $J$ is a multi-index $J = ( j _ { 1 } , \ldots , j _ { n } ) \in \mathbf{N} ^ { n }$, and $\xi ^ { J } = \xi _ { 1 } ^ { j_1 } \ldots \xi _ { n } ^ { j_n }$. To $P$ one associates a constant-coefficient linear partial differential operator $P ( D )$, obtained by replacing $\xi_j$ by $- i \partial / \partial x _ { j }$ ($i = \sqrt { - 1 }$).

Although several special cases had been known for a long time, it is only in the mid 1950s that it has been shown that for every non-zero constant-coefficient linear partial differential operator the equation $P ( D ) ( u ) = g$ can be solved for all functions or distributions (cf. also Generalized function) $g$ on ${\bf R} ^ { n }$. In particular, there is a distribution $E$ such that $P ( D ) ( E ) = \delta _ { 0 }$, where $\delta _ { 0 }$ is the Dirac mass at $0$ (cf. also Dirac delta-function). Such an $E$ is called a fundamental solution. This is the celebrated Malgrange–Ehrenpreis theorem (established independently by B. Malgrange [a8] and L. Ehrenpreis [a1]). In case of the Laplace operator, the Newton potential (up to some factor) is a fundamental solution.

If $g$ is any distribution on ${\bf R} ^ { n }$ with compact support, by using a fundamental solution $E$ it is immediate to solve the equation $P ( D ) ( u ) = g$. Indeed ($*$ denoting convolution, cf. also Convolution of functions),

\begin{equation*} P ( D ) ( E ^ { * } g ) = ( P ( D ) ( E ) ) ^ { * } g = \delta _ { 0 } * g = g. \end{equation*}

The Malgrange–Ehrenpreis theory goes further. One can solve the equation for non-compactly supported right-hand side $g$ by exhausting ${\bf R} ^ { n }$ by balls, and by solving the equation on each ball. The process is made to converge by corrections based on the approximation of local solutions to the homogeneous equation $P ( D ) u = 0$ by global solutions. The space of solutions to a linear homogeneous ordinary differential equation, with constant coefficients (for example, $y ^ { \prime \prime } + b y ^ { \prime } + c y = 0$), is spanned by exponential solutions $x \mapsto e ^ { r x }$ (in the example, $r$ is a root of $r ^ { 2 } + b r + c = 0$), or by exponential and possibly polynomial-exponential solutions ($x e ^ { rx }$ in the case of a double root). Similarly, the solutions to $P ( D ) u = 0$ can be approximated, on convex sets, by linear combinations of polynomial-exponential solutions, functions of the type $Q ( x ) e ^ { i \xi x }$, where $Q$ is a polynomial and necessarily $P ( \xi ) = 0$. See [a2], [a8], [a5], (7.3.6). More generally, convolution equations can be considered.

Most often, the Malgrange–Ehrenpreis theorem is attacked by using the Fourier transform. If $\phi$ is a $C ^ { \infty }$-function on ${\bf R} ^ { n }$, say with compact support, its Fourier transform is defined by

\begin{equation*} \hat { \phi } ( \xi ) = \int _ { \mathbf{R} ^ { n } } \phi ( x ) e ^ { - i \xi x } d x, \end{equation*}

and one has $( P ( D ) ( \phi ) )_{ \widehat{}} ( \xi ) = P ( \xi ) \widehat { \phi } ( \xi )$. Looking for a fundamental solution basically consists in solving the equation $\hat { E } = 1 / P ( \xi )$. This is of course totally elementary in case the function $1 / P ( \xi )$ is integrable on ${\bf R} ^ { n }$, since then, by inverse Fourier transformation, one can define (classically) $E$ by

\begin{equation*} E ( x ) = \frac { 1 } { ( 2 \pi ) ^ { n } } \int _ { \mathbf{R} ^ { n } } \frac { 1 } { P ( \xi ) } e ^ { i \xi x } d \xi . \end{equation*}

(For example, with $\Delta$ denoting the usual Laplacian, the operator $P ( D ) = I + ( - \Delta ) ^ { N }$, for $N > n / 2$, for which $P ( \xi ) = 1 + | \xi | ^ { 2 N }$.)

In general, one has to face two problems: the lack of integrability at infinity of $1 / P ( \xi )$, and the real zeros of the polynomial $P$. Roughly speaking, since the two problems are not disjoint, one overcomes the first difficulty by the calculus of distributions, and the second one by switching to the complex domain. Both Malgrange and Ehrenpreis used the Hahn–Banach theorem. By now (1998) things have been much simplified, and a quick "constructive" proof of the Malgrange–Ehrenpreis theorem can be found in [a5], (7.3.10), see also [a3], (1.56). A formula claimed to be more explicit was recently given by H. König [a7].

The solvability of constant-coefficients linear partial differential operators can also be studied with pure $L^{2}$ methods. There is a remarkable fundamental result by L.V. Hörmander [a4], (2.6), whose proof is totally elementary, brief, and a masterpiece:

If $\Omega$ is a bounded set in ${\bf R} ^ { n }$, and if $P ( D )$ is non-zero, there exists a constant $C > 0$ such that for every $C ^ { \infty }$-function $\phi$ with compact support in $\Omega$,

\begin{equation*} \| P ( D ) ( \phi ) \| _ { 2 } \geq C \| \phi \| _ { 2 } ( L ^ { 2 } \text { norms } ). \end{equation*}

It then follows, from elementary Hilbert space theory, that the adjoint operator $P ^ { * } ( D )$ (associated to the polynomial $\overline { P ( - \xi ) }$) is solvable in $L ^ { 2 } ( \Omega )$. From there, one can recover the Malgrange–Ehrenpreis theorem. See [a10], or [a9] for an elementary exposition. However, the problem being not only to show the existence of a fundamental solution, but to get solutions with good properties, the construction in [a5], Chap. 7, seems close to being optimal, see [a6], Chap. 10.

The situation is radically different for operators with variable coefficients (see Lewy operator and Mizohata operator).

References

[a1] L. Ehrenpreis, "Solutions of some problems of division I" Amer. J. Math. , 76 (1954) pp. 883–903
[a2] L. Ehrenpreis, "Solutions of some problems of division II" Amer. J. Math. , 78 (1956) pp. 685–715
[a3] G.B. Folland, "Introduction to partial differential equations" , Princeton Univ. Press (1995)
[a4] L. Hörmander, "On the theory of general partial differential operators" Acta Math. , 94 (1955) pp. 161–258
[a5] L. Hörmander, "The analysis of linear partial differential operators I" , Grundl. Math. Wissenschaft. , 256 , Springer (1983)
[a6] L. Hörmander, "The analysis of linear partial differential operators II" , Grundl. Math. Wissenschaft. , 257 , Springer (1983)
[a7] H. König, "An explicit formula for fundamental solutions of linear partial differential equations with constant coefficients" Proc. Amer. Math. Soc. , 120 (1994) pp. 1315–1318
[a8] B. Malgrange, "Existence et approximation des solutions des équations aux dérivées partielles et des équations de convolution" Ann. Inst. Fourier (Grenoble) , 6 (1955/6) pp. 271–355
[a9] J-P. Rosay, "A very elementary proof of the Malgrange–Ehrenpreis theorem" Amer. Math. Monthly , 98 (1991) pp. 518–523
[a10] F. Treves, "Thèse d'Hörmander" Sém. Bourbaki , Exp. 130 (1956)
How to Cite This Entry:
Malgrange-Ehrenpreis theorem. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=Malgrange-Ehrenpreis_theorem&oldid=50373
This article was adapted from an original article by Jean-Pierre Rosay (originator), which appeared in Encyclopedia of Mathematics - ISBN 1402006098. See original article