Wiki
Wiki

Research notes on every problem, and a library of the papers behind them. Built from the open erdos repository.

Updated


Source. The unnumbered Theorem, p. 180, of G. Csizmadia, On the Independence Number of Minimum Distance Graphs, Discrete Comput. Geom. 20 (1998), 179--187, DOI 10.1007/PL00009381; see the source card.

Statement

Setting (p. 179). For a set XX of nn points in the plane with minimum distance 11, the minimum distance graph G(X)G(X) has vertex set XX, two points being adjacent exactly when their distance is 11. The paper puts F(n)=min⁡α(G)F(n)=\min\alpha(G), the minimum of the independence number taken over all minimum distance graphs on nn vertices.

Theorem (p. 180, quoted). "Given nn points in the plane with minimum distance 11, we can always choose at least 935n\frac{9}{35}n of them so that their minimum distance is greater than 11."

In the paper's notation this is F(n)≥935nF(n)\ge\frac{9}{35}n (p. 180). The bound holds for every nn; no largeness assumption is made.

Context the paper gives (p. 179). Erdős asked in 1983 for bounds on F(n)F(n). The vertex set of ⌊n/3⌋\lfloor n/3\rfloor widely spaced unit triangles shows F(n)≤⌈13n⌉F(n)\le\lceil\frac13n\rceil, and a construction of Pach and Tóth (1996) gives F(n)≤⌈516n⌉F(n)\le\lceil\frac{5}{16}n\rceil for large nn. Pollack (1985) proved F(n)≥⌈14n⌉F(n)\ge\lceil\frac14n\rceil from planarity and the four color theorem; the paper notes that 14n\frac14n cannot be improved for planar graphs in general.

Proof pointer

The Theorem follows from Lemma 1 (p. 180): applying the lemma repeatedly, each time deleting the independent set it supplies together with its neighbours, and taking the union of the sets removed gives an independent set of at least 935n\frac{9}{35}n vertices, since each round keeps at least 9/359/35 of the vertices it removes. Lemma 1 is proved in Section 2 (pp. 180--183) from two auxiliary lemmas (Lemmas 2 and 3) proved in Section 3 (pp. 183--187).

Dependencies

Lemma 1 of the same paper. Read depth: claims checked; the statement, its hypotheses and the deduction from Lemma 1 were read clause by clause on the print. The proof of Lemma 1 was not checked step by step.

Bears on

  • Problem 1066: the problem's graphs are the minimum distance graphs of nn plane points pairwise at least 11 apart, and its g(n)g(n) is the paper's F(n)F(n) (a point set with no pair at distance 11 has an edgeless graph). The Theorem gives g(n)≥935ng(n)\ge\frac{9}{35}n for every nn. It does not determine g(n)g(n) or the limit of g(n)/ng(n)/n; the upper bounds the paper recalls come from other work.
  • Problem 1070: no bound. The Theorem needs minimum distance 11, and gives no lower bound for arbitrary nn-point sets in the plane, whose unit distance graphs that problem concerns.