You need to know: Space , notation
for norm in
, origin
, ball
with centre in the origin and radius R, volume
of this ball, notation
for the number of elements in finite set A, scalar product
in
, set
of complex numbers,
, Fourier transform
of an integrable function
.
Background: A set S of points in such that
for all distinct
is called a sphere packing. The center density of a sphere packing S is
. The upper density of S is
.
A function is called admissible if there exist positive constants
,
,
such that
and
for all
.
The Theorem: On 1st October 2001, Henry Cohn and Noam Elkies submitted to the arxiv and Annals of Mathematics a paper in which they proved the following result. Suppose is an admissible function, is not identically zero, and such that (i)
for
, and (ii)
(which implies that
is real) for all
. Then the center density of any sphere packing in
is bounded above by
.
Short context: Finding the densest possible sphere packing in is a fundamental problem in geometry, which, before 2001, was solved only in dimensions
. The Theorem provides a general method for proving upper bounds for sphere packing densities. In the same paper, Cohn and Elkies used the Theorem to derive improved upper bounds in dimensions
. In particular, for
, they derive bound
, where
is the density of densest known packing. In later works (see here and here), the Theorem was used to fully solve the sphere packing problem in dimensions 8 and 24.
Links: Free arxiv version of the original paper is here, journal version is here. See also Section 3.4 of this book for an accessible description of the Theorem.
One thought on “Any sphere packing in R^8 has density at most 1.000001 pi^4/384”