arXiv++ Combinatorics

In this paper, using the method of compression, we recover the lower bound for the Erdős unit distance problem and provide an alternative proof to the distinct distance conjecture. In particular, in $\mathbb{R}^k$ for all $k\geq 2$, we have \begin{align} \# \bigg\{||\vec{x_j}-\vec{x_t}||:~||\vec{x_j}-\vec{x_t}||=1,~1\leq t,j \leq n,~\vec{x_j},~\vec{x}_t \in \mathbb{R}^k\bigg\}\geq C\frac{\sqrt{k}}{2}n^{1+o(1)}\nonumber \end{align}for some $C>0$. We also show that \begin{align} \# \bigg\{d_j:d_j=||\vec{x_s}-\vec{y_t}||,~d_j\neq d_i,~1\leq s,t\leq n\bigg\}\geq D\frac{\sqrt{k}}{2}n^{\frac{2}{k}-o(1)}\nonumber \end{align}for some $D>0$. These lower bounds generalizes the lower bounds of the Erdős unit distance and the distinct distance problem to higher dimensions.