On the finite field Kakeya problem in two dimensions (Q877936)

From MaRDI portal
scientific article
Language Label Description Also known as
English
On the finite field Kakeya problem in two dimensions
scientific article

    Statements

    On the finite field Kakeya problem in two dimensions (English)
    0 references
    0 references
    4 May 2007
    0 references
    The Euclidean Kakeya problem asks for the minimum Hausdorff (or Minkowski) dimension of a Besicovitch set, that is, a compact subset of \(\mathbb{R}^n\) which contains a unit line segment in every direction. Whether or not this minimum dimension equals \(n\) for \(n>2\) remains an important open question in analysis [see \textit{T. Tao}, Notices Am. Math. Soc. 48, No. 3, 294--303 (2001; Zbl 0992.42002)]. The finite field analogue of the Kakeya problem was originally suggested by \textit{T. Wolff} [Rossi, Hugo (ed.), Prospects in mathematics. Invited talks on the occasion of the 250th anniversary of Princeton University. Papers from the conference, Princeton, NJ, USA, March 17--21, 1996. Providence, RI: American Mathematical Society. 129--162 (1999; Zbl 0934.42014)] and asks for the smallest subset of \(\mathbb{F}_q^n\) that contains a line segment in every direction. This question has very recently been completely resolved through a surprisingly simple and elegant argument by \textit{Z. Dvir} [\url{http://arxiv.org/abs/0803.2336} (2008)], who showed that such a set must have cardinality at least \(C_n q^n\), with the constant \(C_n\) depending only on \(n\). The present paper was written before this important breakthrough and focuses on the special case \(\mathbb{F}_q^2\). The starting point is given by the so-called Incidence Formula, which states that if \(B\) is a Besicovitch set of the form \(B=\bigcup_{i \in \mathbb{F}_q^n \cup \infty}l(i,b_i)\), where \(l(b,m)\) denotes the line \(y=mx+b\), then \[ | B| =\frac{q(q+1)}{2}+\sum_{P \in B} \frac{(m_P-1)(m_P-2)}{2}, \] where \(m_P\) denotes the number of lines passing through the point \(P \in \mathbb{F}_q^2\). The author conjectures that the second term in this expression is bounded below by \(\frac{q-1}{2}\). He goes on to prove the lower bound of \(\frac{q}{3}\) for any odd \(q\), and the conjectured lower bound of \(\frac{q-1}{2}\) in the special case that there exists an index \(j \in \mathbb{F}_q \cup \infty\) such that every point \(P \in B\) with \(m_P \geq 3\) lies on the line \(l(j, b_j)\). The author also determines the expected cardinality of a Besicovitch set formed by the union of \(q+1\) randomly chosen lines with distinct slopes, which turns out to equal \((1-e^{-1})q^2\).
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    Kakeya problem
    0 references
    finite field
    0 references
    Besicovitch set
    0 references
    0 references
    0 references