Namespaces
Variants
Actions

Friedrichs inequality

From Encyclopedia of Mathematics
Jump to: navigation, search
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.


An inequality of the form

$$ \tag{1 } \int\limits _ \Omega f ^ { 2 } \ d \Omega \leq C \left \{ \int\limits _ \Omega \sum _ {i = 1 } ^ { n } \left ( \frac{\partial f }{\partial x _ {i} } \right ) ^ {2} \ d \Omega + \int\limits _ \Gamma f ^ { 2 } d \Gamma \right \} , $$

where $ \Omega $ is a bounded domain of points $ x = x ( x _ {1} \dots x _ {n} ) $ in an $ n $- dimensional Euclidean space with an $ ( n - 1) $- dimensional boundary $ \Gamma $ satisfying a local Lipschitz condition, and the function $ f \equiv f ( x) \in W _ {2} ^ {1} ( \Omega ) $( a Sobolev space).

The right-hand side of the Friedrichs inequality gives an equivalent norm in $ W _ {2} ^ {1} ( \Omega ) $. Using another equivalent norm in $ W _ {2} ^ {1} ( \Omega ) $, one obtains (see [2]) a modification of the Friedrichs inequality of the form

$$ \tag{2 } \int\limits _ \Omega f ^ { 2 } \ d \Omega \leq C \left \{ \int\limits _ \Omega \sum _ {i = 1 } ^ { n } \left ( \frac{\partial f }{\partial x _ {i} } \right ) ^ {2} d \Omega + \left ( \int\limits _ \Gamma f d \Gamma \right ) ^ {2} \right \} . $$

There are generalizations (see [3][5]) of the Friedrichs inequality to weighted spaces (see Weighted space; Imbedding theorems). Suppose that $ \Gamma \subset C ^ {(} l) $ and that the numbers $ r $, $ p $ and $ \alpha $ are real, with $ r $ being a natural number and $ 1 \leq p < \infty $. One says that $ f \in W _ {p, \alpha } ^ {r} ( \Omega ) $ if the norm

$$ \| f \| _ {W _ {p, \alpha } ^ {r} ( \Omega ) } = \ \| f \| _ {L _ {p} ( \Omega ) } + \| f \| _ {\omega _ {p, \alpha } ^ {r} ( \Omega ) } $$

is finite, where

$$ \| f \| _ {L _ {p} ( \Omega ) } = \ \left ( \int\limits _ \Omega | f | ^ {p} \ d \Omega \right ) ^ {1/p} , $$

$$ \| f \| _ {\omega _ {p, \alpha } ^ {r} ( \Omega ) } = \sum _ {| k | = r } \| \rho ^ \alpha f ^ {(} k) \| _ {L _ {p} ( \Omega ) } , $$

$$ f ^ { ( k) } = \frac{\partial ^ {| k | } f }{\partial x _ {1} ^ {k _ {1} } \dots \partial x _ {n} ^ {k _ {n} } } ,\ \ | k | = \sum _ {i = 1 } ^ { n } k _ {i} , $$

and $ \rho = \rho ( x) $ is distance function from $ x \in \Omega $ to $ \Gamma $.

Suppose that $ s _ {0} $ is a natural number such that

$$ r - \alpha - { \frac{1}{p} } \leq \ s _ {0} < r - \alpha + 1 - { \frac{1}{p} } . $$

Then, if $ \Gamma \subset C ^ {( s _ {0} + 1) } $, $ - p ^ {-} 1 < \alpha < r - p ^ {-} 1 $, $ r/2 \leq s _ {0} $, for $ f \in W _ {p, \alpha } ^ {r} ( \Omega ) $ the following inequality holds:

$$ \| f \| _ {L _ {p} ( \Omega ) } \leq \ C \left \{ \sum _ {l + s < r/2 } \left \| \left ( \left . \frac{\partial ^ {s} f }{\partial n ^ {s} } \ \right | _ \Gamma \right ) ^ {(} l) \ \right \| _ {L _ {p} ( \Gamma ) } + \| f \| _ {\omega _ {p, \alpha } ^ {r} ( \Omega ) } \right \} , $$

where $ ( \partial ^ {s} f/ \partial n ^ {s} ) \mid _ \Gamma $ is the derivative of order $ s $ with respect to the interior normal to $ \Gamma $ at the points of $ \Gamma $.

One can also obtain an inequality of the type (2), which has in the simplest case the form

$$ \| f \| _ {L _ {p} ( \Omega ) } ^ {p} \leq \ C \left ( \| f \| _ {\omega _ {p, \alpha } ^ {1} ( \Omega ) } ^ {p} + \left | \int\limits _ \Gamma u \tau d \Gamma \ \right | ^ {p} \right ) , $$

where

$$ p , \gamma > 1,\ \ - \frac{1}{p} < \alpha < 1 - \frac{1}{p} - \frac{1} \gamma , $$

$$ \tau \in L _ \gamma ( \Gamma ),\ \int\limits _ \Gamma \tau d \Gamma \neq 0. $$

The constant $ C $ is independent of $ f $ throughout.

The inequality is named after K.O. Friedrichs, who obtained it for $ n = 2 $, $ f \in C ^ {(} 2) ( \overline \Omega \; ) $( see [1]).

References

[1] K.O. Friedrichs, "Eine invariante Formulierung des Newtonschen Gravititationsgesetzes und des Grenzüberganges vom Einsteinschen zum Newtonschen Gesetz" Math. Ann. , 98 (1927) pp. 566–575
[2] S.L. Sobolev, "Applications of functional analysis in mathematical physics" , Amer. Math. Soc. (1963) (Translated from Russian)
[3] S.M. Nikol'skii, P.I. Lizorkin, "On some inequalities for weight-class functions and boundary-value problems with a strong degeneracy at the boundary" Soviet Math. Dokl. , 5 (1964) pp. 1535–1539 Dokl. Akad. Nauk SSSR , 159 : 3 (1964) pp. 512–515
[4] S.M. Nikol'skii, "Approximation of functions of several variables and imbedding theorems" , Springer (1975) (Translated from Russian)
[5] D.F. Kalinichenko, "Some properties of functions in the spaces and " Mat. Sb. , 64 : 3 (1964) pp. 436–457 (In Russian)
[6] R. Courant, D. Hilbert, "Methods of mathematical physics. Partial differential equations" , 2 , Interscience (1965) (Translated from German)
[7] L. Nirenberg, "On elliptic partial differential equations" Ann. Scuola Norm. Sup. Pisa Ser. 3 , 13 : 2 (1959) pp. 115–162
[8] L. Sandgren, "A vibration problem" Medd. Lunds Univ. Mat. Sem. , 13 (1955) pp. 1–84
How to Cite This Entry:
Friedrichs inequality. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=Friedrichs_inequality&oldid=46991
This article was adapted from an original article by D.F. KalinichenkoN.V. Miroshin (originator), which appeared in Encyclopedia of Mathematics - ISBN 1402006098. See original article