What are the optimal constants for the Sobolev inequalities in 3D?

  • Context: Graduate 
  • Thread starter Thread starter Gerenuk
  • Start date Start date
  • Tags Tags
    Inequality
Click For Summary
SUMMARY

The discussion focuses on the optimal constants for Sobolev inequalities in three-dimensional space, specifically relating the integral of the gradient of a function to the integral of the function itself. For functions that are zero on a Lipschitz boundary, the relevant inequality is derived from Poincaré's inequality for W0,2 spaces, as outlined in Evans' PDE text on page 265. The optimal constant is dimension-dependent and requires careful tracking through the proofs of relevant estimates. For functions not zero on the boundary, an additional average value adjustment is necessary, leading to another form of Poincaré's inequality as noted on page 275 of Evans' work.

PREREQUISITES
  • Understanding of Sobolev inequalities
  • Familiarity with Poincaré's inequality
  • Knowledge of functional spaces, specifically W0,2
  • Basic concepts of partial differential equations (PDEs)
NEXT STEPS
  • Study the proofs of Sobolev inequalities in Evans' PDE, focusing on pages 265-275
  • Research the implications of dimension-dependent constants in Sobolev spaces
  • Explore the applications of Poincaré inequalities in various boundary conditions
  • Investigate the role of Lipschitz boundaries in Sobolev inequalities
USEFUL FOR

Mathematicians, researchers in functional analysis, and students studying partial differential equations who are looking to deepen their understanding of Sobolev inequalities and their applications in various contexts.

Gerenuk
Messages
1,027
Reaction score
5
I'm searching for an inequality between
[tex]\iiint_\infty |\nabla f|^2 \mathrm{d}^3r[/tex]
and
[tex]\iiint_\infty |f|^2 \mathrm{d}^3r[/tex]

I saw similar inequalities that they called Sobolev inequalities. What would be the correct form and optimal constant for this 3D case?
 
Physics news on Phys.org
For functions that are zero on a lipschitz boundary, this is Poincare's inequality for W01,2. A straightforward proof is found in Evans PDE page 265. If you want to find the exact constant, you could go through the proofs of the relevant estimates on the previous several pages, carefully keeping track of the constants. I haven't done this. The constant will be dimension-dependent.

For functions that are not zero on the boundary, you have to subtract off the average value,
||u-avg(u)||L2 <= C||Du||L2

Which is another "Poincare inequality" (Evand p. 275)
 
Last edited:

Similar threads

  • · Replies 6 ·
Replies
6
Views
4K
  • · Replies 1 ·
Replies
1
Views
1K
Replies
1
Views
7K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 13 ·
Replies
13
Views
2K
  • · Replies 26 ·
Replies
26
Views
2K
Replies
3
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
Replies
4
Views
2K
  • · Replies 0 ·
Replies
0
Views
2K