Conant's Metric Spectra Problem

Abstract

In this paper, we try to minimize the scope of possible unique metric spectra up to equivalence. While it is well known that every spectra $S\subseteq \mathbb{R}^+$ is equivalent to a spectra $T\subseteq \mathbb{N}$, it has remained open if $T$ could also maintain a desirable combinatorial form. Conant questioned if $T= \{t_1,...,t_n \}_{<}$ could be taken such that $2^i-1 \leq t_i \leq 2^n -1$. In this paper, we come to two partial answers. The first is that the largest element $t_n$ can be chosen such that $t_n\leq 2^n $. Approximating a full solution, we also observe $T$ with the combinatorial form $2^i \leq t_i \leq 2^{n+1}$. Our methods are rather unique in the field as we utilize linear optimization and polygonal geometry to achieve our results. Our work aims to approach a full characterization of metric spectra, and simplify future computational endeavors in the field.