This research was supported by NSERC and OTKA grants.
On distinct consecutive differences
József Solymosi
Department of Mathematics, University of British Columbia, Vancouver E-mail address : solymosi@math.ubc.ca
-
Abstract.
We show that if
is a monotone increasing set of numbers, and the differences of the consecutive elements are all distinct, then
for any finite set of numbers
. The bound is tight up to the constant multiplier.
1 introduction
Given two sets of numbers,
and
, the sumset of
and
, denoted by
is
Let
be a finite set of real numbers with the property that
|
(1)
|
for any
Sets with this property are said to be convex sets. Answering a question of Erdős, Hegyvári [5] proved that if
is convex then
Hegyvári's result was later improved by Elekes, Nathanson, and Ruzsa [3] . They proved that if
is convex, then
for any set
with
In this paper we extend this result for sets with distinct consecutive differences. Set
has distinct consecutive differences if for any
implies
Theorem 1.
Let
and
be finite sets of real numbers with
and
If
has distinct consecutive differences, then
In particular, if
then
The basic idea behind the proof is the following. The sumset
consists of
translates of
. The translates of two consecutive elements of
are typically not ”far” from each other in the sumset
Also, from a translate of two consecutive elements,
,
we can recover the value of
, since all of the consecutive differences are distinct. Then the number of ”close” pairs in
should be large, around
, therefore
is also large.
In the second part of the paper we extend the result for two sets. As an application we show that for any convex function
, and finite sets of real numbers,
and
, if
then
Along the lines of the proof one can prove the ”statistical” version of Theorem 1 and Theorem 3. We state the analogue of Theorem 1 without working out the details of the proof.
Theorem 2.
Let
is a monotone increasing set of numbers. If the difference set of the consecutive elements,
, is large,
, then
for any finite set of numbers
, where
depends on
only.
2 Distinct consecutive differences
of Theorem 1. Since
the result is immediate for
so we can assume that
and
Let
and
, where
and
. Let
where
and
Let
and
A pair is a two-element subset of
of the form
|
(2)
|
Suppose that
is a pair and
. Since the set
has distinct consecutive differences, there is a unique integer
such that
It follows that there is a unique integer
such that
Therefore, if
then
and
, and so the sumset
contains exactly
pairs.
Let
be integers such that
We partition the set
into
pairwise disjoint sets
as follows:
and, in general, for
Then
Fix an integer
, and consider the increasing sequence
Let
denote the number of elements of this sequence that belong to the set
Then
If
then the set
contains exactly
pairs of the form ( 2 ), and so the number of pairs with fixed
contained in the sets
is
Since the set
contains exactly
distinct pairs, it follows that the total number of pairs contained in the sets
is at least
We can obtain a simple upper bound for the total number of pairs contained in the sets
as follows: For
the set
contains at most
pairs, and so the number of pairs contained in the sets
is at most
Therefore,
We specialize this inequality as follows: Let
and
where
Then
Choose the integers
such that
and
Then
and so
This implies that
If
and
then
This completes the proof.
3 Distinct pairs of consecutive differences
Let
and
be nonempty sets of real numbers, where
and
. Let
for
and
for
The sets
and
have distinct pairs of consecutive differences if there exists a one-to-one map
such that the
ordered pairs
are distinct.
Theorem 3.
Let
and
be nonempty finite sets of real numbers such that
and the sets
and
have distinct pairs of consecutive differences. Let
, and
be nonempty finite sets of real numbers with
, and
Then
If
then
Let
and
, where
and
. Let
for
and
for
. Let
be a one-to-one map such that the
ordered pairs
are distinct. Let
and
, where
and
.
Let
,
and
We consider quadruples of the form
|
(3)
|
Suppose that
,
and
and that
Then
and
Therefore,
The sets
and
have matching consecutive differences with permutation
, and so
Since
it follows that
and
.
Similarly,
This implies that there are
distinct quadruples of the form ( 3 ).
Consider the sumsets
and
.
where
and
Let
where
and
Let
be integers such that
We partition the set
into
pairwise disjoint sets
as follows:
and, in general, for
Then
Similarly, let
be integers such that
and partition the set
into
pairwise disjoint sets
as follows:
for
Fix an integer
, and consider the increasing sequence
These
numbers belong to sumset
Let
denote the number of elements of this sequence that belong to the set
Then
If
then the set
contains exactly
subsets of the form
|
(4)
|
and so the number of pairs with fixed
contained in the sets
is
There are
pairs of the form ( 4 ), and so the number of such pairs that do not belong to one of the sets
is at most
. Therefore, for each
the number of quadruples of the form ( 3 ) whose first two coordinates do not belong to one of the sets
is at most
Summing over
, we conclude that the number of quadruples of the form ( 3 ) whose first two coordinates do not belong to one of the sets
is at most
Similarly, the number of quadruples of the form ( 3 ) whose third and fourth coordinates do not belong to one of the sets
is at most
Therefore, the number of quadruples of the form ( 3 ) whose first pair of coordinates does not belong to a set
or whose last pair of coordinates does not belong to a set
is at most
It follows that the number of quadruples of the form ( 3 ) whose first pair of coordinates belongs to one of sets
and whose last pair of coordinates also belongs to one of the sets
is bounded below by
On the other hand, for each
and
the number of quadruples
such that
and
for
, and also
and
is exactly
It follows that a simple upper bound for the number of quadruples of the form ( 3 ) whose first pair of coordinates belongs to one of sets
and whose last pair of coordinates also belongs to one of the sets
is
Therefore,
|
(5)
|
If we let
and
and
then
and so
If
then
4 Remarks
A simple consequence of Theorem 3 is the following result, which was first proved by Elekes, Nathanson, and Ruzsa [3] .
Theorem 4.
For any strictly convex real function
, and finite sets of real numbers,
and
, if
then
In particular,
The two sets
and
have distinct pairs of consecutive differences with
the identity map. For the second inequality set
and
A construction of Ruzsa [4] shows that the bounds in Theorem 1 and 3 are tight.
However, as we will see, in his construction
and
have very different structures.
It is possible, that if
in Theorem 1, then
is much larger, maybe close to
Here we sketch Ruzsa's construction. Let
be set such that all the differences in
are distinct,
and
is odd. Then there is a list
of the elements of
with repetitions, consists of
elements, such that the consecutive elements have distinct differences. (
where
implies that
) Now we are ready to define
The set
has the property that the consecutive differences are distinct, and it is not difficult to see that
where
denotes the first
natural numbers.
The example shows that Theorem 1 is sharp, however Erdős' original question is still wide open; What is the smallest possible size of
if
is a convex set of integers?
Acknowledgement: I thank Mel Nathanson for his help on writing up the paper.
References
-
Gy. Elekes, On the number of sums and products. Acta Arith. 81 (1997), no. 4, 365–367.
-
Gy. Elekes, Sums versus products in number theory, algebra and Erdős geometry. in: Paul Erdős and his Mathematics. II, Budapest, Bolyai Society Mathematical Studies, 11. (2002), .
-
Elekes, M. Nathanson, and I, Ruzsa, Convexity and sumsets. J. Number Theory 83 (2000), no. 2, 194–201.
-
I. Ruzsa, personal communication (2003)
-
N. Hegyvári, On consecutive sums in sequences. Acta Math. Hungar. 48 (1986), no. 1-2, 193–200.
Department of Mathematics, University of British Columbia, Vancouver E-mail address : solymosi@math.ubc.ca