Solutions of the Cheeger problem via torsion functions
Analysis of PDEs
2011-07-14 v2
Abstract
The Cheeger problem for a bounded domain Ω⊂RN, N>1 consists in minimizing the quotients ∣∂E∣/∣E∣ among all smooth subdomains E⊂Ω and the Cheeger constant h(Ω) is the minimum of these quotients. Let ϕp∈C1,α(Ωˉ) be the p-torsion function, that is, the solution of torsional creep problem −Δpϕp=1 in Ω, ϕp=0 on ∂Ω, where Δpu:=div(∣∇u∣p−2∇u) is the p-Laplacian operator, p>1. The paper emphasizes the connection between these problems. We prove that limp→1+(∥ϕp∥L∞(Ω))1−p=h(Ω)=limp→1+(∥ϕp∥L1(Ω))1−p. Moreover, we deduce the relation limp→1+∥ϕp∥L1(Ω)≥CNlimp→1+∥ϕp∥L∞(Ω) where CN is a constant depending only of N and h(Ω), explicitely given in the paper. An eigenfunction u∈BV(Ω)∩L∞(Ω) of the Dirichlet 1-Laplacian is obtained as the strong L1 limit, as p→1+, of a subsequence of the family {ϕp/∥ϕp∥L1(Ω)}p>1. Almost all t-level sets Et of u are Cheeger sets and our estimates of u on the Cheeger set ∣E0∣ yield ∣B1∣h(B1)N≤∣E0∣h(Ω)N, where B1 is the unit ball in RN. For Ω convex we obtain u=∣E0∣−1χE0.
Cite
@article{arxiv.1011.3070,
title = {Solutions of the Cheeger problem via torsion functions},
author = {Hamilton Bueno and Grey Ercole},
journal= {arXiv preprint arXiv:1011.3070},
year = {2011}
}
Comments
Typos were corrected