Research notes on every problem, and a library of the papers behind them. Built from the open erdos repository.
Updated
Source. Published pp. 6–7, Proposition 11
(canonical PDF);
arXiv Proposition 3.9.
Statement. If Y is a regular expansion of a finite Euclidean
configuration X, then Y embeds into an m-regular polygonal
torus for some integer m≥2. The radii of its factors need not
be equal.
Proof. A singleton is immediate, so write
X={x1,…,xn} with n≥2, and choose α>0
such that
∥yi−yj∥2=∥xi−xj∥2+α2(i=j).
Put δ=α2/n2. By
Proposition 8, there are m≥2, r>0
and an injective δ-embedding f:X→Tm,rk.
Define the symmetric errors
eij=∥xi−xj∥2−∥f(xi)−f(xj)∥2,∣eij∣<δ.
Set aii=0 and aij=α2+eij for
i=j. These roots are positive because
δ<α2. To verify the hypothesis of Lemma 4, let
A2=maxi<jaij2. Then
The middle strict inequality uses n>1.
Lemma 4 constructs an almost-regular simplex
Z={z1,…,zn} with distances aij.
Proposition 5 embeds Z into an
m-regular torus T0, with the same m already chosen above;
write this isometry as h.
For 1≤i≤n, put yi′=(f(xi),h(zi)) in
Tm,rk×T0. This product is m-regular. For i=j,
For i=j both sides are zero. The matching of labels is thus an
isometric embedding of Y. □
Source precision. The source calls the residual array almost regular
without giving inequality (1); this is the full bound for its exact
choice δ=α2/n2. Its final display is introduced for all
i,j, but the added α2 applies only when i=j.
The diagonal case is separated here. The corrected sufficient parameter
choice in Lemma 7 changes neither the tolerance nor the conclusion of
this argument. No Euclidean realization of the residual is assumed
before Lemma 4 supplies it.