On the planar dual Minkowski problem (Q1649325)

The article under review develops the Brunn-Minkowski theory of convex bodies and deals with a novel notion of dual curvature measures introduced recently in [\textit{Y. Huang} et al., Acta Math. 216, No. 2, 325--388 (2016; Zbl 1372.52007)]. The main attention is paid to the following question called \textit{the dual Minkowski problem}. Given a real number \(q\) and a finite Borel measure \(\mu\) on the sphere \(\mathbb S^n\), does there exist a convex body \(\Omega\) in \(\mathbb R^{n+1}\) such that the \(q\)-th dual curvature measure of \(\Omega\) coincides with \(\mu\)? In the case of \(q=0\) the problem under consideration is equivalent to the classical Alexandrov problem concerning the characterisation of the integral Gauss curvature. The case of \(q<0\) and arbitrary \(n\) is solved in [\textit{Y. Zhao}, Calc. Var. Partial Differ. Equ. 56, No. 2, Paper No. 18, 16 p. (2017; Zbl 1392.52005)]. As for the case of \(q>0\), it was studied only under the additional assumption that \(\mu\) is even so that \(\Omega\) is central-symmetric. The main result of the present paper solves the problem in the case of \(q>0\), \(n=2\), without any assumptions concerning the evenness of \(\mu\). Theorem. 1. Let \(\mu\) be a finite Borel measure on \(\mathbb S^1\) whose density \(f\) is a smooth positive function on \(\mathbb S^1\). Let \(q\) be a positive real number. Then there exists a smooth and strictly convex planar body \(\Omega\subset \mathbb R^2\) whose \(q\)-th dual curvature measure coincides with \(\mu\). 2. Let \(\mu\) be a finite Borel measure on \(\mathbb S^1\) whose density \(f\in L^\infty (\mathbb S^1)\) satisfies \(f\geq \lambda_0\) for some constant \(\lambda_0>0\). Then there exists a convex planar body \(\Omega\subset \mathbb R^2\) whose \(q\)-th dual curvature measure coincides with \(\mu\). Analytically, the problem in the planar case reduces to the equation \[ u_{\theta\theta}+u = \frac{\sqrt{u^2+u^2_\theta}^{\,\, 2-q}}{u} f \] for a function \(u\) depending on an arc-length parameter \(\theta\) on \(\mathbb S^1\). The proof of its solvability is based on specific degree-theoretic arguments including delicate a priori estimates.