MSC: | 11P70, 05E15, | |
11B50 | ||
Abstract:
In the present thesis, we prove a conjecture belonging to a subbranch of additive
combinatorics that is called combinatorial zero{sum theory. Denoting by p an arbitrary prime
number, it has been known for about forty years that every sequence (P1, P2,...,P2p-1)
consisting of 2p - 1 points from the affine plane Fp2
possesses a non-empty subsequence
the sum of whose elements equals zero. This fact has first been shown independently by
Kruyswijk and Olson and is nowadays known to be an easy consequence of Alon's combinatorial
Nullstellensatz. A less obvious question asks for a classification of all those sequences
(P1, P2,...,P2p-1) of length 2p - 2 whose only zero-sum subsequence is the empty
one. This problem has been investigated by several researchers and in this respect a certain
conjecture implying in particular that any such sequence contains p - 2 equal points has
attracted a great deal of attention in recent years. A precise version of this conjecture will be
given below. By definition, the prime number p has property B if and only if it behaves in
accordance with the conjecture under discussion; by developing for the first time a powerful
method for tackling such inverse problems over Fp2
, we prove the result alluded to in the title.