Perturbation and energy estimates (Q4237660)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 1274869
Language Label Description Also known as
default for all languages
No label defined
    English
    Perturbation and energy estimates
    scientific article; zbMATH DE number 1274869

      Statements

      Perturbation and energy estimates (English)
      0 references
      0 references
      13 December 1999
      0 references
      sufficiency of the \((\psi)\) condition of Nirenberg-Trèves
      0 references
      local solvability of the pseudodifferential operators of principal type
      0 references
      The paper is devoted to the problem of the sufficiency of the \((\psi)\) condition of Nirenberg-Trèves for the local solvability of the pseudodifferential operators of principal type; see on the subject the recent paper of \textit{N. Lerner} [Ann. Math., II. Ser. 139, No. 2, 363-393 (1994; Zbl 0818.35152)]. Here, the author proves the following. Let \(P\) be a pseudodifferential operator with symbol in \(S^1_{1,0}\); assume it admits a principal part satisfying condition \((\psi)\). Then there exists an \(L^2\)-bounded operator \(R\) such that the equation \((P+R)u= f\) has a local solution in \(H^{-1}\) for \(f\) in \(L^2\). In general \(R\) has to be taken with non-classical symbol in \(S^0_{1/2,1/2}\). The paper is written in a very clear way and, despite the involved technicality, it is readable by non-experts. We observe the following. It is known that there exists \(P\) satisfying \((\psi)\), for which \(Pu= f\in L^2\), or \((P+R)u= f\in L^2\), do not admit \(L^2\) solutions; that is, a loss of regularity \(u\in H^{-1}\) is expected in general. It is likely, in view of previous counterexamples, that there exists \(P\) satisfying \((\psi)\), for which \((P+R)u= f\in L^2\) do not admit solutions \(u\in H^{-1-N}\), \(N>0\), if the \(L^2\) bounded operator \(R\) is arbitrary; that is, one has to be very careful in choosing \(R\), to obtain \(u\in H^{-1}\). A natural guess is that for every \(P\) satisfying \((\psi)\) and for every \(N\) there exists \(R_N\) regularizing of order \(N\) such that \((P+R_N)u= f\in L^2\) admits solutions \(u\in H^{-1-N}\).
      0 references

      Identifiers