On a non-homogeneous version of a problem of Firey

TL;DR

This paper explores the uniqueness of solutions to a Monge-Ampère type equation on the sphere, extending previous results by analyzing a broader class of functions G and employing novel methods.

Contribution

It introduces new techniques to establish (near) uniqueness for solutions of the Monge-Ampère equation for a wider range of functions G, beyond previously studied cases.

Findings

Proved uniqueness results for a broader class of functions G.

Extended understanding of solutions to the Monge-Ampère type equation on spheres.

Developed new analytical methods differing from prior approaches.

Abstract

We investigate the uniqueness for the Monge-Amp\`{e}re type equation \begin{equation} \label{eq-abstract} det(u_{ij}+\delta_{ij}u)_{i,j=1}^{n-1}=G(u),\ \ \ \ \ \ \ (*)\end{equation}on $S^{n-1}$, where $u$ is the restriction of the support function on the sphere $S^{n-1}$ of a convex body that contains the origin in its interior and $G:(0,\infty)\to(0,\infty)$ is a continuous function. The problem was initiated by Firey (1974) who, in the case $G(\theta)=\theta^{-1}$, asked if $u\equiv 1$ is the unique solution to (*). Recently, Brendle, Choi and Daskalopoulos $[9]$ proved that if $G(\theta)=\theta^{-p}$, $p>-n-1$, then $u$ has to be constant, providing in particular a complete solution to Firey's problem. Our primary goal is to obtain uniqueness (or nearly uniqueness) results for (*) for a broader family of functions $G$. Our approach is very different than the techniques developed in $[9]$.