Let $U$ be a bounded subset of $\mathbb{R}^n$. Let $p>n$. Let $W^{2,1}(U_T)$ be the Banach space of functions $u:U\times[0,T]\rightarrow\mathbb{R}$ with the norm $\|u\|_{W^{2,1}(U_T)}=\sum_{2s+|\kappa|\leq 2}\|\partial_tD^\kappa_xu\|_{L^p(U_T)}$.

For a continuous function $u:U_T\rightarrow\mathbb{R}$ and $0<\alpha<1$. Define $$\langle u\rangle_{C^{\alpha,\alpha/2}(U_T)} = \sup_{x\not=y,t}\frac{|u(x,t)-u(y,t)|}{|x-y|^\alpha} + \sup_{x,t\not=s}\frac{|u(x,t)-u(x,s)|}{|t-s|^{\alpha/2}}.$$

Are there constants $C>0$, $T_0$ and $\delta>0$ such that $\langle u\rangle_{C^{\alpha,\alpha/2}(U_{T_0})} \leq CT_0^\delta\|u\|_{W^{2,1}_p(U_{T_0})}$ for all $u\in W^{2,1}_p(U_{T_0})$?

  • 1
    $\begingroup$ If $U$ is a nice enough domain there are embedding theorems for anysothropic Sobolev spaces into anysotropic Holder spaces, see Besov, Ilin, Nikolskii "Integral representations of functions and imbedding theorems". $\endgroup$ – Andrew Dec 30 '18 at 9:33
  • $\begingroup$ I found the book you mentioned before reading your comment. As you mentioned, the book is really nice. Anyway, thanks. $\endgroup$ – Ngu Dec 30 '18 at 10:49
  • $\begingroup$ Is constant $T_0$ essential in the estimate? Putting $C_1=T_0^\delta$ removes it. $\endgroup$ – Andrew Dec 30 '18 at 17:34
  • $\begingroup$ The constant in the right-hand side depends on $T$. It is bounded for $T$ small. You are right. We have a uniform constant $C$ for all $T\leq T_0$. $\endgroup$ – Ngu Dec 31 '18 at 1:48

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.