116 96 3MB
English Pages 181 [196] Year 2021
AMS / MAA THE CARUS MATHEMATICAL MONOGRAPHS
VOL 37
The Finite Field Distance Problem
David J. Covert
10.1090/car/037
The Finite Field Distance Problem
AMS/MAA
THE CARUS MATHEMATICAL MONOGRAPHS
VOL 37
The Finite Field Distance Problem David Covert
2019â2020 Editorial Committee Bruce P. Palka, Editor Francis Bonahon Steven J. Miller
Annalisa Crannel Carol Schumacher
Alex Iosevich Henry Sagerman
2020 Mathematics Subject Classification. Primary 05-02, 52-02, 52C10, 11E08, 11B13, 11T23. Cover image © 2018 by William Banks For additional information and updates on this book, visit www.ams.org/bookpages/car-37 Library of Congress Cataloging-in-Publication Data Names: Covert, David J., 1984â author. Title: The finite field distance problem / David J. Covert. Description: Providence, Rhode Island: MAA Press, an imprint of the American Mathematical Society, [2021] | Series: The Carus mathematical monographs, 0069-0813; volume 37 | Includes bibliographical references and index. Identifiers: LCCN 2020055010 | ISBN 9781470460310 (paperback) | (ebook) Subjects: LCSH: Finite fields (Algebra) | Combinatorial analysis. | Commutative rings. | AMS: Convex and discrete geometry â Research exposition (monographs, survey articles). | Number theory â Forms and linear algebraic groups â Quadratic forms over local rings and fields. | Number theory â Sequences and sets â Additive bases, including sumsets. | Number theory â Finite fields and commutative rings (number-theoretic aspects) â Exponential sums. Classification: LCC QA247.3 .C68 2021 | DDC 512/.3âdc23 LC record available at https://lccn.loc.gov/2020055010 Copying and reprinting. Individual readers of this publication, and nonprofit libraries acting for them, are permitted to make fair use of the material, such as to copy select pages for use in teaching or research. Permission is granted to quote brief passages from this publication in reviews, provided the customary acknowledgment of the source is given. Republication, systematic copying, or multiple reproduction of any material in this publication is permitted only under license from the American Mathematical Society. Requests for permission to reuse portions of AMS publication content are handled by the Copyright Clearance Center. For more information, please visit www.ams.org/publications/ pubpermissions. Send requests for translation rights and licensed reprints to reprint-permission@ ams.org. © 2021 by the American Mathematical Society. All rights reserved. The American Mathematical Society retains all rights except those granted to the United States Government. Printed in the United States of America. â The paper used in this book is acid-free and falls within the guidelines â
established to ensure permanence and durability. Visit the AMS home page at https://www.ams.org/ 10 9 8 7 6 5 4 3 2 1
26 25 24 23 22 21
To Patrick and Leo
Contents Preface
ix
Acknowledgments
xi
Chapter 1 Background 1.1 Equivalence relations and the pigeonhole principle 1.2 Algebra and finite fields 1.3 Basic inequalities 1.4 Notation 1.5 Exercises: Chapter 1
1 1 8 16 17 18
Chapter 2 The distance problem 2.1 Introduction to the distance problem 2.2 Falconer distance problem 2.3 Finite field distance problem 2.4 Exercises: Chapter 2
21 21 25 28 32
Chapter 3 The Iosevich-Rudnev bound 3.1 Counting-method 3.2 The ð¿2 -method 3.3 Finite field spherical averages 3.4 Size and decay estimates for spheres 3.5 Finite field counterexample 3.6 Relations to the Falconer problem 3.7 Exercises: Chapter 3
35 35 41 42 52 57 59 60
Chapter 4 Wolffâs exponent 4.1 Introduction 4.2 Proof of ð¿2 estimate for ð(ð¡)
63 63 65 vii
viii 4.3 Restriction and extension theory 4.4 Exercises: Chapter 4
Contents 70 73
Chapter 5 Rings and generalized distances 5.1 Distances in finite rings 5.2 Distances between two sets 5.3 Generalized distances 5.4 Pinned distances 5.5 Exercises: Chapter 5
75 75 91 94 98 103
Chapter 6 Configurations and group actions 6.1 Finite configurations 6.2 The Elekes-Sharir framework 6.3 Triangles: The â7/4â bound 6.4 Triangles: The â8/5â bound 6.5 Distance graph 6.6 Exercises: Chapter 6
105 105 111 113 117 120 126
Chapter 7 Combinatorics in finite fields 7.1 Incidence theory 7.2 Sum-product phenomena 7.3 Kakeya conjecture 7.4 Waringâs theorem 7.5 Rothâs theorem and the cap-set problem 7.6 The spectral gap theorem 7.7 Exercises: Chapter 7
129 129 137 146 152 155 161 166
Bibliography
169
Index
179
Preface This book is meant to be a succinct, self-contained, but thorough introduction to the state-of-the-art of the socalled ErdÅs-Falconer Distance Problem, also known as the finite field distance problem. This problem has the fortune of being new, easily understood, and yet it remains unsolved, making this topic ideal for study by advanced undergraduates, graduate students, and research mathematicians alike. The book is structured as follows. In Chapter 1 we start with background material such as a review of properties of finite fields and some basic ideas in combinatorics. In Chapter 2 we will outline the original distance problem by ErdÅs, a continuous version of the problem by Falconer, and a finite field distance problem which will be the core topic of the book. Chapter 3 will be spent proving the first explicit result on the ErdÅs-Falconer distance problem in great detail. Chapter 4 will introduce the topic of restriction theory from harmonic analysis, and we will discuss the role that it plays in proving results on the size of distance sets. Chapter 5 will include many of the generalizations of the distance problem including an extension of the problem to finite rings. Chapter 6 will outline a grouptheoretic approach to the distance problem pioneered by Elekes and Sharir ([43]) and ultimately used by Guth and Katz ([70]) to establish the ErdÅs distance problem in the plane. We will discuss what implications ix
x
Preface these ideas will have in regards to the finite field distance problem. The final chapter (Chapter 7) will be devoted to other topics in the area including incidence theory, sum-product phenomena, the Kakeya conjecture, the cap set problem, and Waringâs problem. The goal of this final chapter is to introduce the non-expert to a handful of research problems in the area, some of which are still active areas of research, while others only having been resolved very recently. We provide enough background in finite fields to allow the reader to understand the material we discuss here. Otherwise, we provide only a cursory introduction to each topic, leaving many resources and references for the motivated reader.
Acknowledgments This book would not have been possible without a long list of supportive colleagues, friends, and family. A special thanks is due to Alex Iosevichâmy mentor and friend who taught me much of what I know about combinatorics and analysis. Many thanks also to Jonathan Pakianathan, Bill Banks (for sharing both his mathematical expertise as well as his painting for the book cover), A. Prabhakar Rao, Ravindra Girivaru, Derrick Hart, Doowon Koh, Steve Senger, Krystal Taylor, Jeremy Chapman, YeÅim DemiroÄlu-Karabulat, and many others. I owe a debt of gratitude to the editors at the AMS Book Program for their encouragement and detailed feedback on earlier drafts of this book. Thanks to my supportive and encouraging family: to my parents for their unwavering support, to my brother who inspired my love of math when I was small and whose friendship I value highly, to my sister who has always stood by my side, and to my wife Laura and sons Patrick and Leo who bring much love, meaning, and entertainment to my life.
xi
10.1090/car/037/01
Chapter
1
Background Before we discuss anything related to the distance problem, we first discuss some background material. Particularly, we will lay out some of the basic ideas in combinatorics and abstract algebra. Readers familiar with notions like the pigeonhole principle, the integers modulo ð, and the construction of finite fields can safely skip to the next chapter.
1.1 Equivalence relations and the pigeonhole principle The pigeonhole principle states that if you were to place ð objects each into one of ð possible bins, then at least one bin must contain at least ð â ð â objects in it. Here âð¥â is the ceiling of ð¥ (meaning âð¥â is the smallest integer ð that is greater than or equal to ð¥). Thus, if I put 10 objects into 10 one of 3 possible bins, then at least one bin must contain at least â 3 â = 4 objects (see Figure 1.1). This obvious-sounding theorem is surprisingly useful and therefore surprisingly pervasive in combinatorics. This problem can be used both to prove the whimsical result that in the city of Saint Louis, Missouri, at least 3 people have exactly the same number of hairs on their head (humans having around 150, 000 hairs on their head, and there being around 318, 000 people living in the city limits of Saint Louis), and yet the pigeonhole principle can also be used to prove that if a real number can be well-approximated by many rational numbers, then that real number must be irrational (see Dirichletâs approximation theorem). Frequently we will use the pigeonhole principle to get quantitive results such as 1
2
Chapter 1. Background
Figure 1.1. Three bins containing a total of 10 objects. Note that if half of the objects were in 1 bin and the other half of the objects were in another bin (so that the third bin was empty), the statement would still be true: at least one bin contains at least 4 objects!
the following: If ð¥1 , ð¥2 , . . . , ð¥ð are positive real numbers, and if ð¥1 â
ð¥2 â
⯠â
ð¥ð ⥠ð¶, then max(ð¥ð ) ⥠ð¶ 1/ð where the maximum is taken over ð â {1, . . . , ð} (see Exercise 1.3). Another elementary notion that we will use is that of equivalence. An equivalence relation on a set ð is a subset ð
â ð à ð such that (ð¥, ð¥) â ð
for all ð¥ â ð, if (ð¥, ðŠ) â ð
, then (ðŠ, ð¥) â ð
, and if (ð¥, ðŠ) â ð
and (ðŠ, ð§) â ð
, then (ð¥, ð§) â ð
. These properties are called reflexivity, symmetry, and transitivity, respectively. We will examine some examples of equivalence relations below. Furthermore, if ð
is an equivalence relation on ð, then for each element ð¥ â ð, we may consider the equivalence class [ð¥] = {ðŠ â ð ⶠ(ð¥, ðŠ) â ð
}.
(1.1)
The set of all equivalence classes is called the quotient set of ð mod ð
, and it will be denoted ð/ð
. The first example of an equivalence relation we should discuss is that of the integers modulo ð. If ð and ð are integers, then we say that ð divides ð (written ð ⣠ð) if ð = ðð for some ð â â€. Thus, 4 ⣠12 as 12 = 4(3), but 5 †12. For a fixed integer ð ⥠2, we say that ð¥ is congruent to ðŠ modulo ð, written ð¥ â¡ ðŠ (mod ð), if ð ⣠(ð¥ â ðŠ). It can be checked (this is code for you should check) that ð
= {(ð¥, ðŠ) â †à †ⶠð¥ â¡ ðŠ
(mod ð)}
is an equivalence relation. Letâs work with ð = 5 for concreteness, so ð
= {(ð¥, ðŠ) â †à †ⶠð¥ â¡ ðŠ
(mod 5)}
= {(ð¥, ðŠ) â †à †ⶠ5 ⣠(ð¥ â ðŠ)} = {(ð¥, ðŠ) â †à †ⶠð¥ = ðŠ + 5ð for some integer ð}.
1.1. Equivalence relations and the pigeonhole principle
3
Now, letâs look at the equivalence class [1]. By definition (1.1) we have [1] = {ðŠ â †ⶠ1 â¡ ðŠ
(mod 5)}
= {ðŠ â †ⶠ5 ⣠(1 â ðŠ)} = {. . . , â9, â4, 1, 6, 11, . . . }. Notice that there are 5 possible equivalence classes of this equivalence relation: [0], [1], [2], [3], and [4], corresponding to the 5 possible remainders when dividing an integer by 5. Working mod 5 can also be thought of as doing arithmetic along a circle rather than along a line.
Addition by some integer ð corresponds to moving ð additional places along this circle. In this setting 4 + 2 = 1, since if we start at 4 and move 2 units in the clockwise direction, we land on the number 1. We could also write this as 4 + 2 â¡ 1 (mod 5), which follows by the definition. Multiplication of ð¥ by some integer ð then corresponds to moving ð¥ units ð times (or equivalently, moving ð units ð¥ times) starting at the position 0. For example, 4 â
2 = 3 since if we move four units 2 times (or if we move 2 units 4 times) we land on the number 3. It should now be clear why modular arithmetic is sometimes referred to as clock arithmetic. Changing the modulus from 5 to some other integer ð requires making a whole new clock. It is convenient to take the integers mod ð, and turn them into a set â€ð , where addition and multiplication are performed mod ð. In â€5 , we have 4+2 = 1 and 4â
2 = 3 as we saw above. The set â€ð is often called the
4
Chapter 1. Background
ring of integers mod ð, and this is language which we will justify shortly. Of special importance to us is the set of integers modulo a prime ð. When ð is prime, we write â€ð = ðœð . It can be shown (see Exercise 1.12) that the set ðœð has the very nice property that every nonzero element has a multiplicative inverse. Thus for all ð¥ â ðœð ⧵ {0}, there exists ðŠ â ðœð such that ð¥ðŠ = 1 in ðœð . Consequently ðœð has the zero product property: For ð, ð â ðœð , if ðð = 0, then either ð = 0 or ð = 0. These sets ðœð are called prime order finite fields, and fields will be described in detail below. That these sets are fields is why we write them as ðœð . Another important equivalence relation is the notion of a projective plane. For example let ðž = â3 ⧵{(0, 0, 0)}, and define the relation ð
â ðžÃ ðž, where ð¥, ðŠ â ðŒ are related if ð¥ and ðŠ lie on the same line through the origin. One can easily check (and you should!) that this is an equivalence relation (see Exercise 1.1). Its resulting quotient set is called the real projective plane ðâ2 . The same projective plane relation can be defined over other sets, such as ðœð for primes ð, though we need to be more careful with what we mean by a line. For a set ð, we let ð ð be the ð-fold Cartesian project of ð on itself. That is, ðð = ð Ãâµâ¯ Ãâ ð = {(ð¥1 , ð¥2 , . . . , ð¥ð ) ⶠð¥1 , ð¥2 , . . . , ð¥ð â ð}. ââµ ââµâµ ðâtimes
Much of the setting of this book will take place in settings like ðœðð . Formally, ðœðð is the ð-dimensional vector space over ðœð , but really we only need to think about ðœðð as the set of ð-tuples of elements in ðœð . For example, thinking of ðœ3 as {0, 1, 2}, we have ðœ23 = {(0, 0); (0, 1); (0, 2); (1, 0); (1, 1); (1, 2); (2, 0); (2, 1); (2, 2)}. It can help to visualize these sets as grids of points with integer coordinates (see Figure 1.2). At this point we are ready to discuss some geometry in ðœðð . A line in 2 ðœð is a set of the form {ð¡ð¥ + ðŠ ⶠð¡ â ðœð } where ð¥, ðŠ â ðœð2 and ð¥ â (0, 0). Notice that in ðœð2 , every line has exactly ð points. For example the line in ðœ23 between (0, 1) and (1, 2) is given by: â1 = {ð¡(1, 1) + (0, 1) ⶠð¡ â ðœ3 } = {(0, 1), (1, 2), (2, 0)}.
(1.2)
See Figure 1.3. Notice also that (1, 2) â (0, 1) = (1, 1) is the direction or slope of this line. This concept of direction is well defined since
1.1. Equivalence relations and the pigeonhole principle
5
Figure 1.2. Representations of ðœ23 with and without axes
Figure 1.3. The line â1 above
ð¡(1, 1)+(0, 1)â(ð (1, 1) + (0, 1)) = (ð¡âð )(1, 1) is a multiple of (1, 1). Thus, (1, 1) and (2, 2) give the same direction of a line since they are nonzero multiples of one another. There are 4 âdirectionsâ in which a line could travel ðœ23 . We will cover this idea of directions in more detail in Section 7.3. As another example, consider the unique line1 â2 in ðœ213 containing the points (0, 5) and (3, 3) (see Figure 1.4). A similar construction defines lines in higher dimensions, though they are more difficult to visualize. A line through ðŠ â ðœðð in direction 1 It is not immediately obvious that there is a unique line between two points, so try to prove it!
6
Chapter 1. Background
Figure 1.4. The line â2 in ðœ213 through (0, 5) and (3, 3)
ð¥ â (0, . . . , 0) is the set of points given by â = {ð¡ð¥ + ðŠ ⶠð¡ â ðœð }. Again, lines in ðœðð contain exactly ð points. For example, the unique line through the points (1, 0, 0) and (2, 2, 1) in ðœ35 (that is, the line through (1, 0, 0) in the direction (1, 2, 1)) is given by â3 = {ð¡(1, 2, 1) + (1, 0, 0) ⶠð¡ â ðœ5 } = {(1, 0, 0); (2, 2, 1); (3, 4, 2); (4, 1, 3); (0, 3, 4)}. In ðœðð , a sphere2 of âradiusâ ð¡ centered at the point (ð1 , . . . , ðð ) â ðœðð refers to the set ð ð¡ = {(ð¥1 , . . . , ð¥ð ) â ðœðð ⶠ(ð¥1 â ð1 )2 + . . . (ð¥ð â ðð )2 = ð¡} . 2 We use the term sphere to describe the set ð even in the case ð = 2, when it might ð¡ be more accurate to call the set a circle.
1.1. Equivalence relations and the pigeonhole principle
7
Note that we do not take square roots since not all elements in a finite field are squares. We will typically work with spheres centered at the origin. As an example, the sphere ð 1 in ðœ25 centered at the origin consists of the four points ð 1 = {(1, 0); (0, 1); (4, 0); (0, 4)} (see Figure 1.5).
Figure 1.5. The sphere ð 1 â ðœ25
Figure 1.6 shows the spheres ð 1 and ð 2 in ðœ211 (both centered at the origin), both of which containing 12 points. It turns out that all spheres with nonzero radius have the same number of points in ðœð2 .
Figure 1.6. The spheres ð 1 (left) and ð 2 (right) in ðœ211
There are some quirks here. In Euclidean space, a sphere of radius 0 is simply a point. In finite fields, we can have spheres with radius 0
8
Chapter 1. Background
containing many points. For example, the sphere ð 0 in ðœ25 centered at the origin contains the 9 points ð 0 = {(0, 0); (1, 2); (1, 3); (2, 1); (3, 1); (2, 4); (3, 4); (4, 2); (4, 3)}.
Figure 1.7. The sphere ð 0 in ðœ211 with and without the (Euclidean) circle (ð¥ â 5/2)2 + (ðŠ â 5/2)2 = 5/2 superimposed
In ðœ35 , the sphere ð 0 contains 30 points (see Figure 1.8). We will give an exact formula for the number of points in a sphere in Chapter 3. It is instructive to consider and illustrate your own examples of lines and sphere in the sets ðœðð . We will delve deeper into these types of geometrical objects later in the text, but at this point we can construct projective planes over some finite fields. Let ð be prime, suppose ðž = ðœð3 ⧵ {(0, 0, 0)}, and define ð
â ðž à ðž where ð¥ â ðž is related to ðŠ â ðž if ðŠ lies on the line âð¥ = {ð¡ð¥ ⶠð¡ â ðœð }. Notice that âð¥ is really just the line containing ð¥ and the origin (0, 0, 0) This relation is again an equivalence relation, and the resulting quotient set is called the finite projective plane ððº(2, ð). Perhaps the best known of these projective planes is the Fano plane ððº(2, 2) over ðœ2 (see Figure 1.9 below). There is much more we could say about these finite projective planes as they are interesting in their own right, but instead we refer the interested reader to [22].
1.2 Algebra and finite fields Given a set ð, a binary operation on ð is an operation acting on pairs of elements of ð. That is, a binary operation â is any function ð Ã ð â ð
1.2. Algebra and finite fields
Figure 1.8. The sphere ð 0 in ðœ35 superimposed on the (Euclidean) sphere (ð¥ â 5/2)2 + (ðŠ â 5/2)2 + (ð§ â 5/2)2 = 19/4.
Figure 1.9. The Fano plane. Image taken from [155].
9
10
Chapter 1. Background
for some set ð Thus ð¥ â ðŠ is well defined for every pair (ð¥, ðŠ) â ð à ð. Addition, multiplication, and subtraction are all binary operations on â, for example.
1.2.1 Rings A ring (ð
, +, â) is a set ð
together with two binary operations + and â satisfying a certain set of properties (which we list below). Rings typically are required to have a unit element 1 â ð
such that 1âð = ðâ1 = ð for all ð â ð
. Table 1.1. Addition and multiplication modulo 5
+ 0 1 2 3 4
0 0 1 2 3 4
1 1 2 3 4 0
2 2 3 4 0 1
3 3 4 0 1 2
4 4 0 1 2 3
â 0 1 2 3 4
0 0 0 0 0 0
1 0 1 2 3 4
2 0 2 4 1 3
3 0 3 1 1 2
4 0 4 3 2 1
Definition 1.2.1. A set (ð
, +, â) is a ring if it satisfies all of the following conditions: â¢
Closure of addition and multiplication: ð+ðâð
â¢
and
ðâðâð
Additive Associativity: ð + (ð + ð) = (ð + ð) + ð
â¢
for all ð, ð â ð
.
Additive Identity: There exists 0 â ð
such that ð+0=0+ð=ð
â¢
for all ð, ð, ð â ð
.
Additive Commutativity: ð+ð=ð+ð
â¢
for all ð, ð â ð
.
for all ð â ð
.
Additive Inverses: For every ð â ð
, there exists ð â ð
such that ð+ð=0
(we will write ð = âð).
1.2. Algebra and finite fields â¢
â¢
11
Left and Right Distributivity: ð â (ð + ð) = ð â ð + ð â ð
for all ð, ð, ð â ð
(ð + ð) â ð = ð â ð + ð â ð
for all ð, ð, ð â ð
.
Multiplicative Associativity: ð â (ð â ð) = (ð â ð) â ð
â¢
for all ð, ð, ð â ð
.
Multiplicative Identity: There exists 1 â ð
so that 1âð=ðâ1=ð
for all ð â ð
.
The canonical example of a ring (and one that will be particularly important in this text) is the set (â€ð , +, â) where â€ð = {0, 1, 2, . . . , ð â 1} is the set of integers mod ð, and where the addition and multiplication are performed mod ð. For the remainder of the text, we will frequently refer to the ring (ð
, +, â) as simply the set ð
if the operations of addition and multiplication are easily understood. Recall from the previous section that formally the elements of â€ð are the equivalence classes [0], [1], . . . , [ð â 1] of the equivalence relation {(ð¥, ðŠ) â †à †ⶠð¥ â¡ ðŠ
(mod ð)},
though it is more convenient to write ð¥ rather than [ð¥], so if we are working in â€5 we would write 2 â
3 = 1 rather than [2] â [3] = [1]. There are many other examples of rings including the integers â€, the rational numbers â, the real numbers â, and the complex numbers â all taken with the standard operations of addition and multiplication. Note that the set of natural numbers â is not a ring since elements of â do not have additive inverses. Other examples include sets such as ððÃð (â) which is the ring of ðÃð matrices with rational entries, where + denotes the usual sum of matrices and â stands for matrix multiplication. What are the additive and multiplicative identity in ððÃð (â)? Given any ring ð
, we define ð
[ð¥] to be the set of all polynomials in variable ð¥ where the coefficients are taken from some ring (ð
, +, â). Here, function addition is the operation (ð + ð)(ð¥) = ð(ð¥) + ð(ð¥) and function multiplication is (ð â
ð)(ð¥) = ð(ð¥) â ð(ð¥), where â is multiplication in the ring ð
. One can (and should) check that ð
[ð¥] is indeed a ring. For example, â[ð¥] is the ring of polynomials with real coefficients.
12
Chapter 1. Background
Many more examples of rings exist, and we even encounter some more examples throughout the text.
1.2.2 Groups Another fundamental concept in abstract algebra is that of a group. We say that a set ðº with a binary operation â is a group if (ðº, â) satisfies the following: â¢
Closure: ðâðâðº
â¢
for all ð, ð â ðº.
Associativity: ð â (ð â ð) = (ð â ð) â ð
â¢
Identity: there exists ð â ðº such that ðâð=ðâð=ð
â¢
for all ð, ð, ð â ðº
for all ð â ðº.
Existence of inverses: For every ð â ðº there exists ð â ðº such that ð â ð = ð â ð = ð.
We often write ð = ðâ1 as ð = âð if we are thinking of the group operation on ðº as being a type of addition. There are numerous examples of groups with which you are (probably) already familiar. For example (â€ð , +) the group of integers modulo ð under addition is a group. Another example is the set â€Ãð = {ð¥ â â€ð ⶠgcd(ð¥, ð) = 1}. â€Ãð is called the multiplicative subgroup of â€ð , and it consists of all the elements in â€ð which have inverses with respect to multiplication (Exercise 1.12 strikes again!). For example, â€Ã10 = {1, 3, 7, 9}, and for any prime ð we have ðœðà = {1, 2, . . . , ð â 1} = ðœð ⧵ {0}.
(1.3)
1.2. Algebra and finite fields
13
Other examples of groups include the group of invertible matrices in ððÃð (â) together with matrix multiplication (called the general linear group): ðºð¿ð (â) = {ð â ððÃð (â) ⶠdet(ð) â 0}, the orthogonal group: ðð (â) = {ð â ððÃð (â) ⶠð â1 = ð ð }, and the special orthogonal group: ðð (â) = {ð â ððÃð (â) ⶠð â1 = ð ð and det(ð) = 1}, where ð ð denote the matrix transpose of the matrix ð.
1.2.3 Fields A field is a ring (ð
, +, â) where every nonzero element in ð
has a multiplicative inverse under the operation â. Examples of fields include the rational numbers â, the real numbers â, and the complex numbers â Notice that †is not a field as most integers do not have multiplicative inverses in the integers. Furthermore by (1.3), the ring (â€ð , +, Ã) is a field if and only if ð is prime, and recall this is why we write â€ð = ðœð . There are numerous facts about fields which will be useful throughout the text. Proposition 1.2.2. If ðœ is a field and if ð¥, ðŠ â ðœ satisfy ð¥ðŠ = 0, then either ð¥ = 0 or ðŠ = 0. Proposition 1.2.3. Let ðœ be a field, and let ð(ð¥) â ðœ[ð¥] be a polynomial in ð¥ with coefficients from ðœ and with degree deg(ð) = ð. Then ð(ð¥) = 0 has at most ð solutions in ðœ. The proof of the first proposition is left as an exercise (see Exercise 1.6). We refer the reader to the excellent3 book [104] for proof of the last proposition and for more facts about fields (and finite fields in particular). 3 It really cannot be overstated: The more time you spend reading Finite Fields by Lidl and Niederetter ([104]), the better!
14
Chapter 1. Background
A field is said to have characteristic ð if ð is the smallest positive integer such that ðð¥ = 0 for all ð¥ â ð
, and where 0 â ð
is the additive identity in ð
. If no such positive ð exists (such as for â, â, and â), then we say these fields have characteristic zero. Furthermore, the field ðœð has characteristic ð. It turns out that all finite fields will have positive characteristic, and there also exist some fields of infinite order with positive characteristic (such as the ð-adic integers). It also turns out that any positive characteristic must be prime (see Exercise 1.7). The order of a finite field is simply the number of elements contained in that finite field. The finite fields of prime order are the easiest to comprehend and work with computationally, but they are far from the only finite fields. It turns out that there exist finite fields of order ðð for every prime ð and every positive integer ð. We will provide explicit methods of constructing such finite fields below. If ðŸ and ð¿ are two fields with the same operations, and if ðŸ â ð¿, then we say that ðŸ is a subfield of ð¿ (or equivalently that ð¿ is an extension field of ðŸ). If a field ðŸ â ð¿ is a proper subset, we call ðŸ a proper subfield of ðŸ (or ð¿ a proper extension field of ðŸ). We will frequently talk about polynomials with coefficients in a field ðœ. Recall that the set of such polynomials is denoted ðœ[ð¥]. Given ð â ðœ[ð¥], the splitting field of ð(ð¥) is the smallest extension field of ð¹ over which ð(ð¥) splits into linear factors. It can be shown that the splitting field of a polynomial always exists and is unique ([104]). Example 1.2.4. The function ð â â[ð¥] given by ð(ð¥) = ð¥2 + 1 has splitting field â. Example 1.2.5. The splitting field of ð â ðœ5 [ð¥] given by ð(ð¥) = ð¥2 + 1 is simply ðœ5 since ð(ð¥) = ð¥2 + 1 = (ð¥ + 3)(ð¥ + 2). Example 1.2.6. The polynomial ð â ðœ3 [ð¥] given by ð(ð¥) = ð¥2 + 1 has splitting field ðœ3 [ðŒ] â {ð¥ + ðŒðŠ ⶠð¥, ðŠ â ðœ3 and ðŒ2 = â1}. This set ðœ3 [ðŒ] has nine elements: ðœ3 [ðŒ] = {0, 1, 2, ðŒ, ðŒ + 1, ðŒ + 2, 2ðŒ, 2ðŒ + 1, 2ðŒ + 2}.
1.2. Algebra and finite fields
15
We will show that ðœ3 [ðŒ] is itself a field with characteristic 3, having ðœ3 as a subfield, and having order 9. If we can show this set is a field, it follows immediately that ðœ3 is a subfield, that the field has characteristic 3, and the field has order 9. Now the simplest way to show this set is a field is to notice that each nonzero element has a multiplicative inverse: 1â
1=1 2â
2=1 ðŒ â
2ðŒ = 1 (ðŒ + 1) â
(ðŒ + 2) = 1 (2ðŒ + 1) â
(2ðŒ + 2) = 1. It remains to show that ðœ3 [ðŒ] splits ð(ð¥) = ð¥2 + 1, but this follows immediately from ð¥2 + 1 = (ð¥ + ðŒ)(ð¥ + 2ðŒ). It turns out that the splitting field of the function ð â ðœð [ð¥] given by ð ð(ð¥) = ð¥ð â ð¥ is a field with ðð elements ([104]). In particular, there exists a finite field of every prime power order. Since any two fields of the same order are isomorphic ([104]), we can refer to the finite field ðœðð . From this we can clearly see some immediate consequences. Recalling that the characteristic of a finite field must be prime, it follows that if ð is not some power of a prime, then there is no field of order ð. If ð = ðð is a power of a prime, then there exists a field of order ðð and the field is unique up to isomorphism. Every field of order ðð has ðœð as a subfield. We can view the finite field ðœðð as an ð-dimensional vector space over its prime field ðœð . The field ðœðð is a subfield of ðœðð if and only if ð ⣠ð. In particular if ðœð is a finite field then its proper subfields must have order âð or smaller. This simple and seemingly insignificant point will be relevant throughout the text. If ðœðð is a subfield of ðœðð , then we can view ðœðð as a vector space over ðœðð . One final point bears repeating: The fields whose orders are prime powers but not primes such as ðœ4 , ðœ8 , and ðœ9 are not isomorphic to the rings â€4 , â€8 , and â€9 . The additive group (ðœð , +) is not cyclic unless ð is prime. However, for every finite field ðœð , the multiplicative subgroup ðœâð = ðœð ⧵ {0} is cyclic ([104]).
16
Chapter 1. Background
1.3 Basic inequalities Most of the inequalities we will use throughout this book will be straightforward, and we will discuss them as we go along. However, a few inequalities are so ubiquitous that is worth recording them here. We will only state these inequalities, and we leave it to the reader to verify them. The first inequality we will encounter with great regularity is the triangle inequality: Proposition 1.3.1. Suppose that ð(ð¡) is any real valued function. Then | | |â ð(ð¡)| †â |ð(ð¡)| . | | | ð¡ | ð¡ Perhaps the most useful inequality for us will be the Cauchy-Schwarz inequality (also called the Cauchy-Bunyakovsky-Schwarz inequality). This inequality takes many forms (for example, it is valid over any inner product space), but we will frequently used the Cauchy-Schwarz inequality for sums as follows. Proposition 1.3.2 (Cauchy-Schwarz). Suppose that ð(ð¡) and ð(ð¡) are nonnegative and real-valued functions. Then, 1/2 2
â ð(ð¡)ð(ð¡) †(â [ð(ð¡)] ) ð¡
ð¡
1/2 2
.
(â [ð(ð¡)] ) ð¡
The proof uses nothing more than the fact that 2ð¥ðŠ †ð¥2 + ðŠ2 for all real numbers ð¥ and ðŠ. A generalization of the Cauchy-Schwarz inequality is Hölderâs inequality: Proposition 1.3.3 (Hölderâs inequality). Suppose that ð(ð¡) and ð(ð¡) are nonnegative and real-valued functions. Suppose that ð and ð are both pos1 1 itive real numbers satisfying ð + ð = 1. Then, 1/ð ð
â ð(ð¡)ð(ð¡) †(â [ð(ð¡)] ) ð¡
ð¡
1/ð ð
(â [ð(ð¡)] ) ð¡
.
1.4. Notation
17
The values ð and ð are often called Hölder conjugates in this context. The proof of this inequality uses Youngâs inequality: If ð and ð are nonnegative real numbers, and if ð and ð are Hölder conjugates, then ðð ðð + ⥠ðð. ð ð Notice that the Cauchy-Schwarz inequality is simply Hölderâs inequality in the case ð = ð = 2. Finally, we have the following trivial but useful inequality which we can view as Hölderâs inequality in the case that ð = â and ð = 1: Proposition 1.3.4. Suppose that ð(ð¡) and ð(ð¡) are nonnegative real-valued functions. Then, â ð(ð¡)ð(ð¡) †(max ð(ð¡)) â ð(ð¡). ð¡
ð¡
ð¡
1.4 Notation We are almost ready to dive into the substance of the text. Before we do, we first provide a short list of some notation and conventions. â¢
ð will always denote a power of an odd prime, unless explicitly stated otherwise.
â¢
ðœð will denote the finite field with ð elements, where ð is always assumed to be odd. ðœðð is the ð-dimensional vector space over ðœð .
â¢
|ðž| will denote the cardinality of a finite set ðž. For ð¥ â âð , we may also use |ð¥| to denote the vectorâs length (i.e. Euclidean norm). For a complex number ð§ â â, we frequently rely on the trivial observation that |ð§|2 = ð§ð§, where ð§ is the complex conjugation of ð§. The notation | â
| should always be clear from the context.
â¢
The constant ð will often stand for a generic (positive) constant which may change from one occurrence to the next. We may add subscripts to particular constants to emphasize dependence on particular parameters.
â¢
The quantity log ð¥ always denotes the natural logarithm.
18
Chapter 1. Background
â¢
Given a ring ð
, ð
à denotes the set of units in ð
, and ð
â denotes the nonzero elements in ð
. Thus, ð
is a field if and only if ð
â = ð
à .
â¢
We use ð(ð¥) ⪠ð(ð¥) (or ð(ð¥) â« ð(ð¥)) to imply the existence of a constant ð > 0 such that ð(ð¥) †ðð(ð¥). Some authors use ð(ð¥) = ð(ð(ð¥)) interchangeably with ð(ð¥) ⪠ð(ð¥). If we wish to emphasize that the implicit constants depend on some parameter ð¡, then we may write ð(ð¥) âªð¡ ð(ð¥).
â¢
ð(ð¥) â ð(ð¥) signifies that ð(ð¥) ⪠ð(ð¥) ⪠ð(ð¥).
â¢
ð(ð¥) = ð(ð(ð¥)) means that limð¥ââ ð(ð¥)/ð(ð¥) = 0 .
â¢
ð(ð¥) ⌠ð(ð¥) means ð(ð¥) = ð(ð¥)(1 + ð(1)).
1.5 Exercises: Chapter 1 Exercise 1.1. Prove that the equivalence relation defining the projective plane ðâ2 is indeed an equivalence relation. Notice that a projective plane can be described in such a manner over any field ðœ. Exercise 1.2. Construct addition and multiplication tables for â€6 . Exercise 1.3. Prove that if ð¥1 + ⯠+ ð¥ð ⥠ð¶, there exists ð â {1, . . . , ð} such that ð¥ð ⥠ð¶/ð. Exercise 1.4. Write down the 9 à 9 multiplication table for ðœ9 , using the construction for ðœ3 [ðŒ] we discussed in Example 1.2.6. Exercise 1.5. Prove that if ðœðð is the ð-dimensional vector space over the finite field with ð elements, then ðœðð contains ðð vectors. Write out all nine vectors from the vector space ðœ23 and all twenty-five vectors in ðœ25 . Exercise 1.6. Prove that if ðœ is a field, and if ð¥, ðŠ â ðœ satisfy ð¥ðŠ = 0, then either ð¥ = 0 or ðŠ = 0. Exercise 1.7. Prove that if a field has positive characteristic, then that characteristic must be prime.
1.5. Exercises: Chapter 1
19
Exercise 1.8. Prove that ð¥2 â¡ 1 (mod ð) has at most two solutions for all primes ð, while ð¥2 â¡ 1 (mod 8) has 4 unique solutions (mod 8). Is this consistent with what we know about fields? Hint: Yes, it is! Exercise 1.9. Show that â€8 is not a field. Use the splitting field construction to construct a field of order 8. Exercise 1.10. Euclidâs Algorithm gives a method for computing gcd(ð, ð) for two positive integers ð and ð. Look up this algorithm, and compute gcd(2020, 7) using the algorithm. Exercise 1.11. Use Euclidâs Algorithm (see Exercise 1.10) to prove Bézoutâs Little Theorem: If ð and ð are relatively prime, then there exist integers ð¥ and ðŠ (called the Bézout coefficients of ð and ð) such that ðð¥ + ððŠ = 1. Exercise 1.12. Use Bézoutâs Little Theorem (see previous exercise) to prove that an element ð â â€ð has a multiplicative inverse if and only if gcd(ð, ð) = 1. That is, the congruence ðð¥ â¡ 1
(mod ð)
has a solution ð¥ â â€ð if and only if gcd(ð, ð) = 1. Hint: Write 1 = ðð¥+ððŠ since gcd(ð, ð) = 1, and recall that equivalence relations are reflexive. Exercise 1.13. Using your calculation from Exercise 1.10, find the Bézout coefficients of 2020 and 7. Then, use Exercise 1.12 to compute the multiplicative inverse of 7 (mod 2020). Exercise 1.14. Show that 5 â
2â1 â¡ 0 (mod 5) and 19 â
4â1 â¡ 1 (mod 5). Compare this to the circle and sphere superimposed in Figures 1.7 and 1.8.
10.1090/car/037/02
Chapter
2
The distance problem 2.1 Introduction to the distance problem Letâs play a game. Your goal is to draw three points on a blank sheet of paper, and then connect each pair of dots with a line, trying to minimize the number of distinct distances. How many distinct distances did you draw? If you picked three random points, then most likely you will have 3 distinct distances. If you drew an isosceles triangle, with two sides being the same length, then your drawing determined only two distances. Better still, if you drew an equilateral triangle so that all sides of the triangle are equal, then you win as you have three points determining only one distance, which is clearly the best you can do. Now, what happens if you have to select four points? What is the minimum number of distinct distances you achieve here? It turns out that a square will determine only two distances (the edge of the square and the diagonal), and this is best possible. Now what if you draw 5 points? 10 points? 100 points?
The complexity of the problem increases dramatically even for a small number of points (try this for 10 points for example). Therefore, rather than trying to find a function which would give the minimum of distances for exactly ð points, it is more productive to ask about the longterm behavior of such a quantity: Roughly how many distances do we 21
22
Chapter 2. The distance problem
achieve if we were to draw ð points on a sheet of a paper, when ð is very large? This question was originally posed by Paul ErdÅs in 1946 ([49]). Suppose ð(ð) denotes the minimum number of distances determined by a set of ð points in â2 , so that ð(3) = 1 as an equilateral triangle determines a single distinct distance, ð(4) = ð(5) = 2 as shown above with the square and pentagon, and so on. What is the best asymptotic lower bound for ð(ð), if we let ð go to infinity? More precisely, given a set ðž â â2 of cardinality |ðž| = ð, define Î(ðž) = {|ð¥ â ðŠ| ⶠð¥, ðŠ â ðž} â â,
(2.1)
where |ð¥| = âð¥12 + ⯠+ ð¥ð2 denotes the standard (Euclidean) distance. Then, the ErdÅs distance problem asks one for a lower bound on the quantity ð(ð) = min{|Î(ðž)| ⶠðž â â2 and |ðž| = ð}. In [49], ErdÅs showed that one always has the bound ð âð ⪠ð(ð) ⪠. âlog ð
(2.2)
The upper bound in (2.2) came by considering the set ðž = {(ð¥, ðŠ) â †à †ⶠ0 †ð¥, ðŠ †âð} when ð is a square. It is well known ([90]) that if ðŽ(ð¥) denotes the set of nonnegative integers less than or equal to ð¥ which are sums of two squares, then |ðŽ(ð¥)| ⌠ðŸð¥/âlog ð¥, where ðŸ = 0.7642 . . . is the LandauRamanjuan constant. It follows that |Î(ðž)| †2|ðŽ(ð)|, so that |Î(ðž)| â€
2ðŸð âlog 2ð
â
ð âlog ð
.
(2.3)
The lower bound was achieved by some clever combinatorial reasoning. Essentially, ErdÅsâ argument involved only elementary notions like the fact that two distinct circles intersect in at most 2 points, coupled with the pigeonhole principle. Details of this argument can be found in Section 6.2, and a thorough and intuitive description of these results and the progress thereafter are the subject of the excellent book The ErdÅs distance problem by Garibaldi, Iosevich, and Senger ([61]). ErdÅsâ bounds from (2.2) led him to conjecture ([15]) that ð(ð) â« ð1âð(1) for all sets ðž â â2 with cardinality ð. The lower bound was
2.1. Introduction to the distance problem
23
steadily improved ([18, 19, 98, 112, 143, 146, 151]) until more than sixty years after ErdÅs first wrote about this problem, a breakthrough was achieved by Guth and Katz ([70]). They showed that ð , ð(ð) â« (2.4) log ð a very near optimal bound which established the ErdÅs distance conjecture! Later (in Section 6.2) we will provide a brief outline of their proof which follows the so-called Elekes-Sharir framework ([43]), and we provide some related insights by applying analogues of their ideas to the finite field problem which we will describe below. The ErdÅs distinct distance problem has been considered in higher dimensions as well. Let ðð (ð) denote the minimum number of Euclidean distances determined by a set of ð points in âð . The integer lattice example giving the upper bound for ð = 2 shows that when ð ⥠3, we can hope for ðð (ð) ⪠ð2/ð , and no better. More precisely, let ðž = {(ð¥1 , . . . , ð¥ð ) â â€ð ⶠ1 †ð¥ð †ð1/ð for all ð = 1, . . . , ð} for some integer ð which is a perfect ðth power. Legendre showed that if ðµ(ð¥) denotes the set of positive integers which are the sum of three 5 squares, then ðµ(ð¥) satisfies ðµ(ð¥) ⌠6 ð¥ (see Exercise 2.2). Additionally, every integer is the sum of at most four squares by Lagrangeâs four-square theorem. Hence, ðð (ð) †ð ð ð2/ð ⪠ð2/ð for ð ⥠3. It is further conjectured that ðð (ð) â« ð2/ð with no logarithmic loss whenever ð ⥠3. ErdÅsâ bound ð(ð) â« ð1/2 can be adapted to higher dimensions to show that ðð (ð) â« ð1/ð . The best result in general higher dimensions belongs to Solymosi and Vu ([144]) who showed that 2
ðð (ð) â« ð ð
â
2 ð2 +2ð
.
2.1.1 ErdÅs unit-distance problem ErdÅs also considered a second distance problem. Given a set of ð points, how many times is it possible for a single distance to occur? Formally, we define ð(ð) = max |{âð¥ â ðŠâ = 1 ⶠð¥, ðŠ â ðž}| . |ðž|=ð
ðžââ2
24
Chapter 2. The distance problem
Note that given a point set ðž, we can simply scale the set so that the most frequent distance achieved is 1, and hence there is no harm in replacing âmost popular distanceâ with the distance 1. Now, ErdÅs showed that we must have the bound ð(ð) â« ð1+ð/ log log ð , for some ð > 0 by again considering the set {(ð¥, ðŠ) â †à †ⶠ1 †ð¥, ðŠ †âð} and using known results about the distribution of primes of the form ð = 4ð + 1 ([48]), since such primes can always be written as the sum of squares (see Exercise 2.4). He conjectured ([15]) that ð(ð) ⪠ð1+ð(1) , but the best bound established thus far is ð(ð) ⪠ð4/3 ,
(2.5)
due to the work of Spencer, Szemerédi, and Trotter ([140]), which relied on the work of Szemerédi and Trotter on incidences: Theorem 2.1.1 ([148]). Let ð â â2 be a set of points, and let ð¿ be a set of lines in â2 . An incidence is a pair (ð, â) such that ð â â. The number of incidences determined by the sets ð and ð¿ is then ðŒ(ð, ð¿) = |{(ð, â) â ð à ð¿ ⶠð â â}|. Finally we have ðŒ(ð, ð¿) ⪠(|ð|2/3 |ð¿|2/3 + |ð| + |ð¿|). See Section 7.1 for much more information on incidence theory and its applications to problems in additive combinatorial problems in particular. Really, the result of Spencer, Szemerédi, and Trotter relied on an adaptation of the work of Szemerédi and Trotter where lines were replaced by curves with no self-intersections. This was then applied to circles in the plane. This remarkable theorem of Szemerédi and Trotter has many applications to geometric combinatorics including a result on the sum-product problem ([41]) and the unit-distance problem ([140]). We explore incidence theory in more detail in Chapter 7.
2.2. Falconer distance problem
25
The unit-distance problem is only interesting in dimensions 3 and lower due to some arithmetic obstructions ([89]). In â3 , we have that ð(ð) ⪠ð2/3 , so that ð3 (ð) â« ð4/3 by the pigeonhole principle, where ð3 (ð) = max |{âð¥ â ðŠâ = 1 ⶠð¥, ðŠ â ðž}| . |ðž|=ð
ðžââ3
Until recently, the best known bound was ð3 (ð) ⪠ð3/2 , but now the world record belongs to Joshua Zahl ([160]), as he has demonstrated that 3
1
ð3 (ð) ⪠ð 2 â 394 +ð(1) . The unit distance problem remains one of the most important unsolved problems in combinatorics. See, for example, [1, 15, 93, 108, 117] and the references therein for current results on the problem and its variants.
2.2 Falconer distance problem In 1986, Falconer ([52]) considered a continuous version of the distance problem. While the Falconer distance problem is very similar to that of the ErdÅs distance problem, the Falconer distance problem requires a more technical setup, so these ideas should not be taken too seriously, especially if the notions of measure and fractal dimension are new to you. There are two main ideas necessary to understand the gist of the Falconer distance problem. The first idea is that of (Lebesgue) measure. The measure of a set ðž â âð is the quantity that mathematicians use which can loosely be thought of as length in â, area in â2 , volume in â3 , and so on. Also, a set ðž â âð has measure 0 if for any ð > 0, the set can be covered by a countable (or finite) collection of open balls whose union has volume ð. The only fact concerning measure which we will require here is that any countable set ðž â âð has measure 0 (see Exercise 2.5). The second idea to convey is that of Hausdorff dimension. We loosely describe dimension in the following intuitive way. If you were to take a line segment of length 1 and scale down the line segment by 1/2, you would unsurprisingly (and perhaps uninterestingly) get a line segment of length 1/2 = (1/2)1 . If you take a square of area 1, and scale down
26
Chapter 2. The distance problem
each side-length to half its size, the resulting square would be of area 1/4 = (1/2)2 . If you take a cube of area 1 and scale down each side to half its original size, you would make a cube with volume 1/8 = (1/2)3 . You can think of those powers in the terms (1/2)1 for the line segment, (1/2)2 for the square, and (1/2)3 for the cube as the reason why the line segment is 1-dimensional, the square is 2-dimensional, and the cube is 3-dimensional.
Now letâs consider the strange and wonderful fractal that is the Cantor set. The Cantor set ð is defined as follows: Start with the unit interval [0, 1] â â, and remove the middle third (1/3, 2/3). You are left with only the union of two intervals [0, 1/3] ⪠[2/3, 1]. Now from each of those intervals, remove the middle third, to be left with a union of 4 intervals, each of length 1/9. Taking a limit of this process gives a fractal called the Cantor set1 :
Notice that if you scale down the Cantor set by a factor of 3, you have split the Cantor set into two identical copies of itself, each 1/3 the size of the original. This means that the dimension ð of the Cantor set should satisfy (1/3)ð = 1/2. In this sense the Cantor set ð has fractal dimension ln(2)/ ln(3) â 0.63093 . . . . 1 Image
taken from [156].
2.2. Falconer distance problem
27
While Hausdorff dimension is the notion of fractal dimension used in the Falconer distance problem, there are many other notions of fractal dimension, particularly Minkowski dimension, box-counting dimension, and packing dimension. We will write dimð» (ðž) to denote the Hausdorff dimension of the set ðž. See Section 7.3 for an introduction to the Minkowski dimension, and the interested reader should see [110] for an excellent treatise on fractal geometric including rigorous definitions of measure, fractal dimension, and much more. Let Î(ðž) be the distance set defined as before. Falconer showed that if ðž â âð is compact and the set ðž has Hausdorff dimension dimð» (ðž) > ð+1 , then the set Î(ðž) would have positive Lebesgue measure. Further2 ð
more, he showed that for any ð < 2 , there exists a set ðž â âð such that dimð» (ðž) = ð , and yet the distance set Î(ðž) has Lebesgue measure zero. This led him to conjecture that if a set ðž â âð is compact with ð dimð» (ðž) > 2 , then the distance set Î(ðž) would have positive measure. The details of Falconerâs construction can be found in [53] or [110], for example. The Falconer distance problem can be thought of as a generalization of the Steinhaus Theorem ([53]), which states that if ðŽ â â is of positive (Lebesgue) measure, then the set ðŽ â ðŽ = {ð â ðⲠⶠð, ðâ² â ðŽ} contains some interval of the form [0, ð) for some ð > 0. The Falconer distance problem has received considerable attention, most notably from Wolff, Bourgain, and ErdoÄan. For a long while, the best known result in the plane was due to Wolff ([157]) as he demonstrated that Î(ðž) has positive Lebesgue measure when dimð» (ðž) > 4/3 for all compact sets ðž â â2 , hence why we refer to the results in Chapter 4 as âWolffâs Exponent.â Interestingly, Wolffâs bound was improved by Guth, Iosevich, Ou, and Wang ([69]) who demonstrated that if ðž â â2 is compact and has dimð» (ðž) > 5/4, then Î(ðž) has positive Lebesgue measure2 . If ðž â â2 is compact and also ð -Ahlfors-David regular for some ð ⥠1, then Orponen ([118]) has shown that Î(ðž) has packing dimension dimð (Î(ðž)) = 1. In higher dimensions the best known results currently belong to Du, Guth, Ou, Wang, Wilson, Zhang ([37]) for ð = 3 and Du and Zhang ([38] for ð ⥠4. They have shown that if ðž â â3 satisfies dimð» (ðž) > 9/5 and if ðž â âð 2 They actually proved a stronger so-called pinned-distance result, the likes of which we discuss in Section 5.4.
28
Chapter 2. The distance problem ð
1
1
(for ð ⥠4) satisfies dimð» (ðž) > 2 + 4 + 8ðâ4 , then Î(ðž) has positive Lebesgue measure in each case. See [47, 110, 157] for a more detailed treatise of the subject.
2.3 Finite field distance problem Finite fields have long been used as an uncomplicated setting in which one could play with Euclidean problems in an environment with fewer technical difficulties. To this end the distance problem was considered in this finite setting in order to gain some insight into the Euclidean distances problems discussed above. A finite field version of the distance problem will be described below. We will always require the parameter ð to be odd unless explicitly stated otherwise. Let ðœðð denote the ð-dimensional vector space over the finite field with ð elements. Recall that ð must be the power of an odd prime3 . For ð¥ = (ð¥1 , . . . , ð¥ð ) â ðœðð , we set âð¥â = ð¥ â
ð¥ = ð¥ð¡ ð¥ = ð¥12 + ⯠+ ð¥ð2 â ðœð , viewing ð¥ as a column vector. The quantity âð¥â is clearly not a norm in any analytic sense, though it should be observed that the function ð(ð¥) = âð¥â is preserved by rigid motions. More precisely, let ð ð (ðœð ) denote the set of ð à ð orthogonal matrices with entries in ðœð . Then, it is easy to check that âð¥â = âðð¥â for any ð â ð ð (ðœð ). As before, for ðž â ðœðð we define the distance set of ðž as Î(ðž) = {âð¥ â ðŠâ ⶠð¥, ðŠ â ðž} â ðœð . This notion of distance is also preserved by rigid motions in ðœðð . Example 2.3.1. Consider the set ðœ213 consisting of the five points ðž = {(4, 2); (4, 3); (5, 4); (7, 4); (3, 6)}. You can check that Î(ðž) = {0, 1, 2, 4, 5, 7, 8, 10} â ðœ13 (see Figure 2.1). The first result on distances in finite fields came from Bourgain, Katz, and Tao ([14]) who showed the following. 3 If finite fields are new mathematical objects to you, then it may be helpful to first consider the case when ð is prime, as ðœð is indeed a finite field when ð is prime.
2.3. Finite field distance problem
29
Figure 2.1. The set ðž from Example 2.3.1 represented by the set of integer points (ð¥, ðŠ) where 0 †ð¥, ðŠ †12
Theorem 2.3.2. Suppose that ðž â ðœð2 , where ð â¡ 3 (mod 4) is a prime. Suppose that ðž has cardinality |ðž| = ððŒ for some 0 < ðŒ < 2. Then, there exists a quantity ð = ð(ðŒ) such that 1
|Î(ðž)| â« |ðž| 2 +ð . Note that if ðŒ = 2, then |Î(ðž)| = |ðž|1/2 , and no better. Notice also that if ð â¡ 1 (mod 4), then there exists an element ð â ðœð such that ð2 = â1. Hence, we can construct a set ðž = {(ð¥, ðð¥) ⶠð¥ â ðœð } such that Î(ðž) = {0}, and yet |ðž| = ð. This finite field distance problem was rephrased and improved upon by Iosevich and Rudnev ([87]). Rather than showing an ErdÅs-style bound of |Î(ðž)| > |ðž|ðœ , it is more fruitful to ask how large a set ðž â ðœðð needs to be in order to ensure that Î(ðž) = ðœð , or more modestly that |Î(ðž)| â« ð. Therefore the finite field distance problem has flavors of both the ErdÅs distance problem and the Falconer distance problem, and is henceforth referred to as the ErdÅs-Falconer distance problem. Before we state the current conjecture, we first detail what progress has been made on this problem. Theorem 2.3.3 ([87]). Let ðž â ðœðð with |ðž| > 2ð
ð+1 2
. Then, Î(ðž) = ðœð .
30
Chapter 2. The distance problem We will prove this theorem in Chapter 3. In the same chapter, we will ð+1
give proofs of the weaker result |Î(ðž)| â« ð when |ðž| â« ð 2 , which turns out to have many applications to generalizations of the distance problem. Regarding counterexamples, notice that if ð = ð2 , and since ðœð contains a subfield isomorphic to ðœð , then we have found a set ðž â ðœðð with |ðž| = ðð = ðð/2 , and yet |Î(ðž)| = ð = âð, and no better. Even worse, adapting a counterexample from above, note that in any even dimension ð = 2ð, if â1 is a square in ðœð , then we could construct ðž = {(ð¥1 , ðð¥1 , ð¥2 , ðð¥2 , . . . , ð¥ð , ðð¥ð ) ⶠð¥ð â ðœð } which gives a set of size |ðž| = ðð/2 with Î(ðž) = {0}. Using these examples and the Falconer distance conjecture as a guide, it would be reasonable to assume that if |ðž| ⥠ð¶ðð/2 for a sufficiently large constant ð¶, then one achieves ðœð = Î(ðž). However, it was shown in [76] that the exponent
ð+1 2
is best possible at least in odd dimensions. That is, given ð+1
where ð ⥠3 is odd, we can construct ðž â ðœðð such that |ðž| â ð 2 âð and yet |Î(ðž)| ⪠ð1âð . In the Euclidean setting point-sets behave more or less the same regardless of the parity of the dimension of the ambient space. In finite fields, however, the parity of the dimension will play a much larger role. We will construct the above counterexample in Section 3.5. Finally, some partial improvement to Theorem 2.3.3 is known but only in the case ð = 2. Chapman, ErdoÄan, Hart, Iosevich, and Koh ([17]), showed that if ðž â ðœ2ð where ð â¡ 3 (mod 4) and |ðž| ⥠ð4/3 , then |Î(ðž)| ⥠ðð for some 0 < ð †1. A further paper ([7]) was able to extend this result to the case ð â¡ 1 (mod 4). Combining these results, we have: ðœðð ,
Theorem 2.3.4. Let ðž â ðœ2ð such that |ðž| ⥠ð4/3 for a sufficiently large constant ð¶. Then, there exists an absolute constant ð â (0, 1] such that |Î(ðž)| ⥠ðð. While this shows that the distance set is large when ðž is of sufficiently large cardinality (in fact we show that at the very least the distance set contains a positive proportion of the field ðœð ), we still do not know which elements of ðœâð are in the distance set. Thus, if we care whether or not 1 â Î(ðž), for example, then we must require that |ðž| > 2ð
ð+1 2
.
2.3. Finite field distance problem
31
In an interesting twist Murphy and Petridis ([114]) demonstrated that the 4/3 threshold is best possible in ðœ2ð , at least for arbitrary finite fields when you want to guarantee that all elements are in the distance set Î(ðž). Specifically they construct a family of sets ðž â ðœ2ð with |ðž| = ð4/3 such that |Î(ðž)| < ð. Other results are known to hold for certain sets ðž (such as Salem sets and subsets of spheres), and we collect such results in later chapters (see Subsection 3.3.1 and Theorem 5.3.6). Putting all of the results together, we now state two versions of the ErdÅs-Falconer distance conjecture. Conjecture 2.3.5 (Strong ErdÅs-Falconer Distance Conjecture). Let ðž â ðœðð , ð ⥠4 even, be such that |ðž| ⥠ð¶ðð/2 for a sufficiently large constant ð¶. Then, Î(ðž) = ðœð . Conjecture 2.3.6 (Weak ErdÅs-Falconer Distance Conjecture). Let ðž â ðœðð , ð ⥠2 even, be such that |ðž| ⥠ð¶ðð/2 for a sufficiently large constant ð¶. Then, |Î(ðž)| ⥠ðð for some 0 < ð †1. In the following chapters, we will recount the progress made on the strong and weak versions of the ErdÅs-Falconer conjectures and its many variants thus far. Before we prove any finite field distance result, I want to emphasize the importance and utility of studying finite field analogues of classical (Euclidean) problems. While we have not seen all of these problems just yet, I will mention a strange intertwining of results between both the finite field and Euclidean worlds. The Kakeya conjecture (see Section 7.3) asserts that any set containing a line in every direction must be appropriately large. The finite field analogue, long thought to be a difficult analytical problem, turned out to have an easy (one paragraph) algebraic solution using the so-called algebraic method (see Theorem 7.3.9). This algebraic method turned out not to be useful for the original Kakeya problem, though it did inspire G. Elekes and M. Sharir to recast the ErdÅs distance problem in terms of algebraic operations (see Section 6.2). This recasting of the problem allowed for the aforementioned solution of the ErdÅs distance problem in dimension ð = 2 by GuthKatz, but it did not yield any gains in higher dimensions due to some annoying algebraic obstructions there. The progress made on the ErdÅs
32
Chapter 2. The distance problem
distance problem further helped to improve the Falconer distance problem in dimension ð = 2, but also did not yield any new results in higher dimensions. The ErdÅs distance problem remains open for ð ⥠3, the ErdÅs-unit distance problem remains open in dimensions 2 and 3, the Falconer distance problem remains open in all dimensions ð ⥠2, and the finite field distance problem is open for all even dimensions ð ⥠2. At any step in this strange winding road of progress, it was not obvious that a solution to some problem was going to assist with the solution to another, but undoubtedly, finite fields have played a key role in better understanding some combinatorial problems, even in the Euclidean setting.
2.4 Exercises: Chapter 2 Exercise 2.1. Find a set ðž â ðœ23 such that |ðž| = 3 and 1 â Î(ðž). On the other hand, prove that any set ðž â ðœ23 with cardinality |ðž| ⥠4 satisfies Î(ðž) = ðœ3 . Moreover, find a set ðž â ðœ25 such that |ðž| = 10, and yet 1 â Î(ðž). Then, prove that if ðž â ðœ25 has cardinality |ðž| ⥠11, then Î(ðž) = ðœ5 . Exercise 2.2. Let ðµ = {ð â â€+ ⶠð = ð¥2 + ðŠ2 + ð§2 for some ð¥, ðŠ, ð§ â â€}, and consider its counting function ðµ(ð¥) = {ð â ðµ ⶠð †ð¥}. Provide a counting argument for: ðµ(ð¥) 5 lim = . 6 ð¥ââ ð¥ Hint: Use the following heuristic: Legendreâs 3 squares theorem states that an integer is not a sum of three squares if and only if it is of the form 4ð (8ð + 7) for nonnegative integers ð and ð. Finally, notice â
1 1 = . ð 6 8 â
4 ð=0 â
Exercise 2.3. Show that if you take ðž â ðœð2 to be the set of vertices which have an even number of zero components, then |ðž| = 2ðâ1 and yet Î(ðž) = {0}. However, using the pigeonhole-principle, show that if |ðž| > 2ðâ1 , then Î(ðž) = ðœ2 . This is why we emphasize that in the above theorems (and the text throughout), ð is assumed to be odd.
2.4. Exercises: Chapter 2
33
Exercise 2.4. Look up and write down any proof of Fermatâs Theorem on sums of two squares which states that an odd prime ð is a sum of two squares if and only if ð â¡ 1 (mod 4). Exercise 2.5. Prove that any countable set in âð has measure 0, using the definitions laid out in Section 2.2. Exercise 2.6. Prove that every element in ðœð is the sum of two squares using the pigeonhole principle. Conclude that the square lattice construction in (2.3) does not yield a logarithmic loss in ðœ2ð .
10.1090/car/037/03
Chapter
3
The Iosevich-Rudnev bound 3.1 Counting-method The goal of this chapter is to understand two different proofs of Theorem 2.3.3. The first proof will show that every element from ðœð is in the distance set of ðž so long as |ðž| ⥠4ð
ð+1 2
. The second proof will simply show
ð+1 2
that if |ðž| ⥠ð¶ð for a sufficiently large constant ð¶, then there exists a constant ð â (0, 1] such that the distance set satisfies |Î(ðž)| ⥠ðð. Both proofs involve studying the counting function ð(ð¡) = |{(ð¥, ðŠ) â ðž à ðž ⶠâð¥ â ðŠâ = ð¡}| which counts the number of pairs of points which are distance ð¡ apart from one another. Note that ð¡ â Î(ðž) if and only if ð(ð¡) > 0. Thus, we aim to show simply that ð(ð¡) > 0 for all ð¡ â ðœâð , since ð(0) > 0 trivially when ðž â â
as âð¥ â ð¥â = 0 for all ð¥ â ðž. Our first step is to convert ð(ð¡) into a sum ð(ð¡) =
â
1,
(3.1)
ð¥,ðŠâðž
âð¥âðŠâ=ð¡
and we will study this sum in detail. We next use a technique of analytic number theory to turn this sum into an exponential sum. Recall we are working in the finite field ðœð , where ð = ðð for some odd prime ð. A group character is defined on any abelian group ðº as a group homomorphism from that group ðº to the set of complex numbers of modulus 1. In addition to additive characters on the (additive) group 35
36
Chapter 3. The Iosevich-Rudnev bound
(ðœð , +), we will later encounter so-called multiplicative characters on the group (ðœâð , â
). The canonical additive character of ðœðð is the function ð ⶠðœð â â defined by ð(ð) = ð2ðð Tr(ð)/ð where Tr ⶠðœðð â ðœð is the linear function (called the field trace) given 2 ðâ1 by Tr(ð) = ð + ðð + ðð + ⯠+ ðð . Notice that when ð = 1 (that is, when ð itself is prime), Tr(ð) = ð, so that ð(ð) = ð2ððð/ð . Example 3.1.1. For the finite field ðœ5 , the canonical additive character is the function ð ⶠðœ5 â â given by ð(ð¥) = ð2ððð¥/5 . Notice that for ð¥ â ðœ5 , any integer congruent to ð¥ (mod 5) will yield the same value for ð(ð¥). Thus, ð(2) = ð2ððâ
2/5 = ð2ððâ
7/5 = ð(7) = ð4ðð/5 . Example 3.1.2. Consider the finite field ðœ9 as constructed in Example 1.2.6. We had ðœ9 = ðœ3 [ðŒ] = {ð¥ + ðŒðŠ ⶠð¥, ðŠ â ðœ3 and ðŒ2 = â1}. Now, the additive character on ðœ9 is ð(ð¥) = ð2ðð Tr(ð¥)/3 where Tr ⶠðœ9 â ðœ3 is defined as Tr(ð¥) = ð¥ + ð¥3 . One can compute Tr(0) = 0, Tr(1) = 2, Tr(2) = 1, and Tr(ðŒ) = ðŒ(ðŒ2 + 1) = 0. One can also compute (or show by linearity) that Tr(ðŒ + 1) = Tr(ðŒ) + Tr(1) = 2, for example. It then follows that ð(ðŒ + 1) = ð2ððâ
Tr(ðŒ+1)/3 = ð2ððâ
2/3 = ð4ðð/3 = ð(ðŒ)ð(1). The linearity of the field trace is what allows ð to be a homomorphism (see Exercise 3.1). All additive characters on ðœð are of the form ðð (ð) = ð(ðð) = ð2ððð Tr(ð)/ð for ð â ðœð . The character ð0 (ð) â¡ 1 is called the trivial character on ðœð , and all other characters are called nontrivial. All results mentioned in this book will be independent of which character is used (so long as we avoid the trivial character), so we will simply use the canonical additive character ð(â
) = ð1 (â
) throughout the text. For more information on characters in finite fields, see [104].
3.1. Counting-method
37
Characters satisfy the following orthogonality principle: ðâð â ð(ð¥ â
ð) = { ð¥âðœð ð
1 0
ð=0 ðâ 0
where for ð¥, ð â ðœðð , the quantity ð¥ â
ð = ð¥1 ð1 + . . . ð¥ð ðð denotes the standard inner product. In particular if ð = 1 we get â ð(ðð¡) = { ð¡âðœð
ð 0
ð=0 otherwise.
(3.2)
We denote the sphere of radius ð¡ by ð ð¡ = {ð¥ â ðœðð ⶠâð¥â = ð¡}. We will see shortly that the sphere ð ð¡ behaves like a Euclidean sphere in âð in many ways. For example two distinct spheres in ðœðð intersect in at most 2 points, the sphere ð ð¡ is (ð â 1)-dimensional, and the sphere ð ð¡ even satisfies the same so-called stationary phase estimates as Euclidean spheres. Spheres in ðœðð can exhibit some strange behaviors, such as the existence of nontrivial spheres of radius 0 (as we saw in Chapter 1). For a set ðž, we will write ðž(â
) to denote its characteristic function, so that we would write ðž(ð¥) = {
1 0
ð¥âðž ð¥ â ðž.
We will use this characteristic function notation frequently, especially for spheres ð ð¡ . Orthogonality will allow us to turn the sum (3.1) into the following sum: ð(ð¡) = â ðž(ð¥)ðž(ðŠ)ð ð¡ (ð¥ â ðŠ). (3.3) ð¥,ðŠ
We will study this sum by applying some techniques from harmonic analysis. For a function ð ⶠðœðð â â, we define the normalized Fourier transform of ð as Ë ð(ð) = ðâð â ð(ð¥)ð(âð¥ â
ð), ð¥âðœð ð
where ð is the canonical additive character of ðœð discussed above, and where ð¥ â
ð again denotes the standard dot-product for ð¥, ð â ðœðð . It turns that if ð is a nontrivial character on ðœ, then all the characters on
38
Chapter 3. The Iosevich-Rudnev bound
ðœðð are of the form ð(ð¥ â
ð) for some ð â ðœðð . This finite field Fourier transform is analogous to the Fourier transform often used in harmonic analysis, signal processing, modeling using PDEs, and engineering: Ë ð(ð) = â« ð(ð¥)ðâððð ð(ð¥ â
ð)ðð¥. âð
The orthogonality of characters in finite fields leads us to the following results: Proposition 3.1.3 (Inversion). Let ð ⶠðœðð â â. Then, Ë ð(ð¥) = â ð(ð)ð(ð¥ â
ð). ðâðœð ð
Proposition 3.1.4 (Plancherel). For a function ð ⶠðœðð â â, we have 2
Ë || = ðâð â |ð(ð¥)|2 . â ||ð(ð) ðâðœð ð
ð¥âðœð ð
We leave the proofs of these elementary Propositions to the reader (see Exercises 3.2 and 3.3). We are now ready to prove Theorem 2.3.3 for the first time. Applying inversion and Plancherel to the sum (3.3), and using that ð is a homomorphism, we see that ð(ð¡) = â ðž(ð¥)ðž(ðŠ)ð ð¡ (ð¥ â ðŠ) ð¥,ðŠâðœð ð
=
ðž(ð¥)ðž(ðŠ)ðËð¡ (ð)ð((ð¥ â ðŠ) â
ð)
â ð¥,ðŠ,ðâðœð ð
= â (â ðž(ðŠ)ð(âðŠ â
ð)) (â ðž(ð¥)ð(âð¥ â
ð))ðËð¡ (ð). ðâðœð ð
ðŠ
ð¥
Applying the definition of the Fourier transform, this becomes
Ë ðËð¡ (ð) Ë ð(ð¡) = â ðð ðž(ð) â
ðð ðž(ð) ð 2
Ë || ð ð¡Ì (ð). = ð2ð â ||ðž(ð) ð
3.1. Counting-method
39
In our first step, we applied inversion on the sphere, and then in the next step, we summed in ð¥ and ðŠ to get the Fourier transform of the characteristic function of the set ðž. This establishes 2
Ë || ð ð¡Ì (ð). ð(ð¡) = ð2ð â ||ðž(ð)
(3.4)
ð
Now, 2
Ë || ð ð¡Ì (0) + ð2ð ð(ð¡) = ð2ð ||ðž(0)
2
Ë || ð ð¡Ì (ð) â ð + ð
ð¡ , (3.5) ||ðž(ð)
â ðâ (0,. . .,0)
where ð is the term corresponding to ð = (0, . . . , 0) and where ð
ð¡ is the sum of the nonzero values of ð. Notice that by definition, we have Ë . . . , 0) = ðâð â ðž(ð¥) = ðâð |ðž|, ðž(0, ð¥âðœð ð
and likewise ðËð¡ (0, . . . , 0) = ðâð |ð ð¡ |. We next need to determine the cardinality of the sphere ð ð¡ â ðœðð . Proposition 3.1.5. Let ð ð¡ â ðœðð be as above. Then for ð ⥠2, we have |ð ð¡ | = {
ððâ1 + ð
ðâ1 2
ð ((â1)
ð¡)
ð is odd
ððâ1 + ð
ðâ2 2
ð ((â1)ð/2 ) (ð¿(ð¡)ð â 1)
ð is even.
ðâ1 2
Here we are using the notation ð(ð¥) = 1 if ð¥ â 0 is a square in ðœð , ð(ð¥) = â1 if ð¥ is not a square in ðœð , and ð(0) = 0. Also the quantity ð¿ ⶠðœð â {0, 1} is the 1-dimensional delta function ð¿(ð¡) = {
1 0
ð¡=0 ð¡ â 0.
In particular |ð ð¡ | †2ððâ1 independent of ð¡, and |ð ð¡ | = ððâ1 (1 + ð(1))
when either ð¡ â 0 or ð ⥠3.
In this sense, the sphere ð ð¡ â ðœðð is (ð â 1)-dimensional. Applying these bounds to the main term of our sum (3.5) we see that ð = ðâð |ðž|2 |ð ð¡ | = ðâð |ðž|2 ððâ1 (1 + ð(1)) =
|ðž|2 (1 + ð(1)), ð
(3.6)
by Proposition 3.1.5. Finally, we estimate the remainder, though first we need to more closely examine the quantity ð ð¡Ì (ð).
40
Chapter 3. The Iosevich-Rudnev bound
Proposition 3.1.6. Let ð ð¡ (â
) denote the characteristic function of the set ð ð¡ , and let ð¿(ð) = {
1 0
ð = (0, . . . , 0) ð â (0, . . . , 0).
Then, ð ð¡Ì (ð) =
ð+2 âðâ ð¿(ð) + ððð ðâ 2 â ð (âð¡ð â ) ð 4ð ð â 0
where for odd prime powers ð = ðâ , (â1)ââ1
ðâ¡1
(mod 4)
(â1)ââ1 ðâ
ðâ¡3
(mod 4).
ðð = { In particular for ð â (0, . . . , 0) we have |ð 0Ì (ð)| †ðâð/2 , while for ð¡ â 0 and ð â (0, . . . , 0): ||ðËð¡ (ð)|| †2ðâ
ð+1 2
.
We will delay the proofs Propositions 3.1.5 and 3.1.6 until Section 3.4 later in the chapter. For now, just implementing Proposition 3.1.6, we see that the remainder term from (3.5) satisfies |ð
ð¡ | †ð2ð
2
Ë || ||ðËð¡ (ð)|| ||ðž(ð)
â ðâ (0,. . .,0)
†max |ð ð¡Ì (ð)| ð2ð ð¡â 0,ðâ 0â
†2ð
â
ð+1 2
Ë || ||ðž(ð)
â
2
ðâ (0,. . .,0)
ð2ð
â
2
Ë || . ||ðž(ð)
ðâ (0,. . .,0)
We can bound the sum over all nonzero ð to that of all ð yielding |ð
ð¡ | †2ðâ
ð+1 2
2
Ë || . ð2ð â ||ðž(ð) ðâðœð ð
3.2. The ð¿2 -method
41
Applying Plancherelâs identity (Proposition 3.1.4), we see that |ð
ð¡ | †2ðâ
ð+1 2
2
ð2ð ðâð â |ðž(ð¥)| ð¥
= 2ð
ðâ1 2
|ðž|.
Putting together all of our estimates, we have seen that for ð¡ â 0, we have the estimate |ðž|2 (1 + ð(1)) + ð
ð¡ , ð
ð(ð¡) =
where |ð
ð¡ | †2ð
ðâ1 2
(3.7)
|ðž|. It follows that ð(ð¡) > 0 whenever ðâ1 |ðž|2 (1 + ð(1)) > 2ð 2 |ðž|, ð
which is certain to happen if |ðž| ⥠4ð Theorem 2.3.3.
ð+1 2
. This finishes our first proof of
Note 3.1.7. In the proof, we used only that |ð ð¡ | = ððâ1 (1 + ð(1)). One can go more carefully through the previous argument using the precise values for the size of the sphere, and it turns out that ð(ð¡) > 0 whenever |ðž| > 2ð
ð+1 2
(see Exercise 3.16).
3.2 The ð¿2 -method As before set ð(ð¡) = |{(ð¥, ðŠ) â ðž à ðž ⶠâð¥ â ðŠâ = ð¡}|, and we note that â ð(ð¡) = |ðž|2 . ð¡
42
Chapter 3. The Iosevich-Rudnev bound
Applying Cauchy-Schwarz to this sum, and letting 1â(ðž) (â
) denote the characteristic function of the distance set Î(ðž), we see that 2
2
4
|ðž| = ( â ð(ð¡)) = ( â 1â(ðž) (ð¡) â
ð(ð¡)) ð¡âðœð
ð¡âðœð
†â 12â(ðž) (ð¡) â
â ð2 (ð¡) ð¡âðœð
ð¡âðœð 2
= |Î(ðž)| â ð (ð¡) ð¡âðœð
which immediately yields |Î(ðž)| â¥
|ðž|4 . âð¡ ð2 (ð¡)
(3.8)
This method has some flexibility. For example we could also write: 2
2
4
|ðž| = (â ð(ð¡) + ð(0)) †2 (â ð(ð¡)) + 2ð(0)2 ð¡â 0
ð¡â 0
†2|Î(ðž)| â ð2 (ð¡) + 2ð2 (0) ð¡â 0
which implies that |Î(ðž)| â¥
1 |ðž|4 2
â ð2 (0)
âð¡â 0 ð2 (ð¡)
.
(3.9)
Note that we used the trivial inequality (ð + ð)2 †2(ð2 + ð2 ). We will choose whether we need (3.8) or (3.9) or other variants of this idea depending on each individual application of the method. Thus all of the work involved with showing that a set ðž has a large distance set is in finding an upper bound on the quantities âð¡ ð2 (ð¡) and âð¡â 0 ð2 (ð¡) (hence why we call this the âð¿2 -methodâ).
3.3 Finite field spherical averages We next turn to studying a finite field quantity involving so-called spherical averages.
3.3. Finite field spherical averages
43
Theorem 3.3.1. Let 2
2
Ë || = â ||ðž(ð) Ë || ð ð¡ (ð), ððž (ð¡) = â ||ðž(ð) âðâ=ð¡
ðâðœð ð
and define ððž (ð) =
ð3ð+1 â ð2 (ð¡). |ðž|4 ð¡âðœâ ðž ð
ð 2
Then for |ðž| ⥠4ð , we have |Î(ðž)| ⥠min(ðð,
ð ). ððž (ð)
In particular for sets ðž satisfying |ðž| ⥠4ðð/2 , we will have |Î(ðž)| â« ð whenever ððž (ð) â« 1. The quantity ððž (ð¡) is called a spherical average since the sum is taken over the sphere of radius ð¡. We will follow the main idea of the original proof from [87]. Recall that we already know (see Note 3.1.7) that if |ðž| > 2ð
ð+1 2
, then Î(ðž) = ðœð , so we may assume that the cardinality of ð
ðž satisfies 4ð 2 †|ðž| †2ð ð(0) =
ð+1 2
. Note that
2 |ðž|2 |ð 0 | Ë0 (ð) Ë || ð ||ðž(ð) â + ð2ð ðð ðâ (0,. . .,0)
so that |ð(0)| †2 since |ðž| †2ð
ð+1 2
ð ð |ðž|2 + ð 2 |ðž| †2ð 2 |ðž| ð
. Hence ð2 (0) †4ðð |ðž|2 â€
1 4 |ðž| 4
ð
when |ðž| ⥠4ð 2 . Note that the constant 4 can be improved though we make no attempt to do so here. Combing this with (3.9) we have established that |ðž|4 . |Î(ðž)| ⥠(3.10) 4 âð¡â 0 ð2 (ð¡) To prove Theorem 3.3.1, we start with some preliminary lemmas.
44
Chapter 3. The Iosevich-Rudnev bound
Proposition 3.3.2. For ð â ðœðð and ð â ðœð , define ðº(ð, ð) = â ð(ðâð¥â + ð â
ð¥). ð¥âðœð ð
Then ððð ðð/2 ð (âðâ(4ð)â1 )
ðâ 0
ð¿(ð)ðð
ð=0
ðº(ð, ð) = {
where ðð â {±1, ±ð} is independent of ð and ð, and where ð¿ is the delta function from above. Proof. The proof follows from the orthogonality of characters when ð = 0. When ð â 0, the result follows by completing the square and using the well known value of the so-called Gauss sum: ðð â â ð(ðð¥2 ) = ðð âð.
(3.11)
ð¥âðœð
Our first step is to break this sum into the sums of the components of the vector ð¥ = (ð¥1 , . . . , ð¥ð ): ðº(ð, ð) = â ð(ðâð¥â + ð â
ð¥) ð¥âðœð ð
= â ð(ðð¥12 + ⯠+ ðð¥ð2 + ð1 ð¥1 + ⯠+ ðð ð¥ð ) ð¥âðœð ð ð
= â â ð(ðð¥ð2 + ðð ð¥ð ). ð=1 ð¥ð âðœð
Completing the square is only possible in fields of odd order since we require 2 â ðœð to be invertible. To simplify notation we will write write ð to denote the quantity ððâ1 where ðâ1 is the multiplicative inverse of ð ð in the underlying field. For example in ðœ11 , we have
1 2
= 6. Recall that
3.3. Finite field spherical averages
45
we are assuming ð â 0. Now in any field of odd order we have ð ðð¥ð2 + ðð ð¥ð = ð (ð¥ð2 + ð ð¥ð ) ð ð2 ð2 ð = ð (ð¥ð2 + ð ð¥ð + ð2 â ð2 ) ð 4ð 4ð ðð 2 ð2ð . ) â ð 4ð Thus our expression for ðº(ð, ð) becomes = ð (ð¥ð +
ð
ðº(ð, ð) = â â ð (ð (ð¥ð + ð=1 ð¥ð âðœð ð
= â ð (â ð=1
ð2 ðð 2 ) ) ð (â ð ) ð 4ð
ð2ð ð 2 ) â ð (ð (ð¥ð + ð ) ) 4ð ð¥ âðœ ð ð
ð
ð
= ð(âðâ/4ð)(ðð ) . Applying (3.11), the expression for ðº(ð, ð) follows. To establish (3.11), we first write |ðð |2 = â ð(ð(ð¥2 â ðŠ2 )) ð¥,ðŠâðœð
= â ð(ðð¡)ð»(ð¡) ð¡âðœð
where ð»(ð¡) = |{(ð¥, ðŠ) â ðœð ⶠð¥2 â ðŠ2 = ð¡}|. A simple counting argument (see Exercise 3.4) shows that ð»(0) = 2ð â 1, while ð»(ð¡) = ð â 1 for ð¡ â 0. It follows that |ðð |2 = (2ð â 1) + (ð â 1) â ð(ðð¡) = ð.
(3.12)
ð¡â 0
Next, note that ðð = â ð(ðð¥2 ) = â ð(ð§)ð(ð§), ð¥âðœð
ð§âðœð
where ð(ð§) = |{ð¥ â ðœð ⶠðð¥2 = ð§}| = 1 + ð(ð§ðâ1 ) = 1 + ð(ð)ð(ð§), since ð(ðâ1 ) = ð(ð) for ð â 0. This shows that ðð = â ð(ð§)(1 + ð(ð)ð(ð§)) = ð(ð) â ð(ð¥)ð(ð¥). ð§âðœð
ð¥âðœð
(3.13)
46
Chapter 3. The Iosevich-Rudnev bound
Using (3.13), we obtain ðð = ðâð = ð(âð) â ð(ð¥)ð(ð¥) = ð(â1)ðð .
(3.14)
ð¥âðœð
Putting together (3.12) and (3.14), we see that ð2ð = ðð ð(â1)ðð = ð(â1)|ðð |2 = ð(â1)ð, and the result follows by taking square roots. It should be noted that precisely determining the signs of the quantity ðð is much more delicate than one first imagines, and such details can be found, for example, in [104]. Proposition 3.3.3. Fix ð¥ â ðœðð . Consider ð¢ð¥ ⶠðœð â â given by ð¢ð¥ (ð) = ðð (ð¥). Then, 1 ð¢ð¥Ì (ð) = ð(âðâð¥â). ð The proof is again straight-forward though it uses the orthogonality relation ð¢ð¥ (ð) = ðâ1 â ð(âð (âð¥â â ð)). ð
The details are left as an exercise (Exercise 3.11). Now, we put together these estimates. Recall (see (3.4)) that for ð¡ â ðœð we have 2
Ë || ð ð¡Ì (ð). ð(ð¡) = ð2ð â ||ðž(ð) ðâðœð ð
Taking the (1-dimensional) Fourier transform of ð and applying the above expression for ð gives ð(ð) Ì = ðâ1 â ð(ð)ð(âðð) ðâðœð 2
Ë || ððÌ (ð)ð(âðð) = ð2ðâ1 â â ||ðž(ð) ðâðœð ðâðœð ð
= ððâ1 â
2
Ë || ðð (ð¥)ð(âð â
ð¥)ð(âðð) â ||ðž(ð)
ðâðœð ð,ð¥âðœð ð
3.3. Finite field spherical averages
47
where in the last step we simply applied the definition of the (ð-dimensional) Fourier transform of the indicator function ð ð¡ . Applying orthogonality we have ð(ð) Ì = ððâ2 â
2
Ë || ð(âð â
ð¥)ð(ð (âð¥â â ð))ð(âðð) â ||ðž(ð)
ð ,ðâðœð ð,ð¥âðœð ð 2
Ë || ð(âð â
ð¥ + ð âð¥â). = ððâ2 â [ â ð(âð(ð + ð ))] â ||ðž(ð) ð âðœð
ðâðœð
ð,ð¥âðœð ð
Orthogonality implies that the sum inside the brackets is 0 unless ð = âð. Hence 2
Ë || ð(âð â
ð¥)ð(âðâð¥â) ð(ð) Ì = ððâ1 â ||ðž(ð) ð,ð¥âðœð ð 2
Ë || ðº(âð, âð) = ððâ1 â ||ðž(ð) ðâðœð ð
where ðº(âð, âð) is the sum appearing in Proposition 3.3.2. Applying our bound for ðº(âð â ð), we see that 2
3
Ë || ð (â ð(ð) = ððð ð 2 ðâ1 â ||ðž(ð) ðâðœð ð
âðâ ) 4ð
(3.15)
so long as ð = 0. Note that ð(0) Ì = |ðž|2 /ð. Applying the above expression for ð(ð), Ì we have that 2
â |ð(ð)| Ì = ð3ðâ2 ðâ 0
â
2
2
Ë || ||ðž(ð Ë â² )|| â ð ( ||ðž(ð) ðâ 0
ð,ðâ² âðœð ð
= ð3ðâ2
â ð,ðâ² âðœð ð
2
2
âðâ² â â âðâ ) 4ð
Ë || ||ðž(ð Ë â² )|| â ð (ð (âðâ² â â âðâ)) ||ðž(ð) ðâ 0
48
Chapter 3. The Iosevich-Rudnev bound
by the change of variables ð ⊠1/(4ð). Thus, 2
â |ð(ð)| Ì = ð3ðâ2 ðâ 0
2
2
Ë || ||ðž(ð Ë â² )|| â ð (ð (âðâ² â â âðâ)) ||ðž(ð)
â
ðâ 0
ð,ðâ² âðœð ð
= ð3ðâ2
2
2
2
2
Ë â² )|| â ð (ð (âðâ² â â âðâ)) Ë || ||ðž(ð ||ðž(ð)
â
ð
ð,ðâ² âðœð ð
â ð3ðâ2
Ë || ||ðž(ð Ë â² )|| . ||ðž(ð)
â ð,ðâ² âðœð ð
By Plancherel the second sum is simply ð3ðâ2
â
2
2
2
Ë || ||ðž(ð Ë â² )|| = ð3ðâ2 (|ðž|/ðð ) = |ðž|2 ððâ2 . ||ðž(ð)
ð,ðâ² âðœð ð
Applying orthogonality yet again, we see that â ð (ð (âðâ² â â âðâ)) = ð ðâðœð
so long as âðâ = âðâ² â, and the sum is zero otherwise. Thus, 2
â |ð(ð)| Ì = â|ðž|2 ððâ2 + ð3ðâ1
2
2
Ë â² )|| . Ë || ||ðž(ð ||ðž(ð)
â âðâ=âðâ² â
ðâ 0
Now, recall that the spherical average is given by 2
Ë || . ððž (ð¡) = â ||ðž(ð) âðâ=ð¡
In particular, â ððž2 (ð¡) = â
2
2
Ë || ||ðž(ð Ë â² )|| ||ðž(ð)
â
ð¡âðœð âðâ=ð¡,âðâ² â=ð¡
ð¡âðœð
=
â âðâ=âðâ² â
2
2
Ë || ||ðž(ð Ë â² )|| . ||ðž(ð)
(3.16)
3.3. Finite field spherical averages
49
Plugging this in yields 2
â |ð(ð)| Ì = ð3ðâ1 â ððž2 (ð¡) â ððâ2 |ðž|2 ðâ 0
ð¡
=ð
3ðâ1
â ððž2 (ð¡) â ððâ2 |ðž|2 + ð3ðâ1 ððž2 (0) ð¡â 0
|ðž|4 = 2 ððž (ð) â ððâ2 |ðž|2 + ð3ðâ1 ððž2 (0) ð where ððž (ð) is the quantity appearing in Theorem 3.3.1. Using Plancherel once again: 2
â ð2 (ð¡) = ð â |ð(ð)| Ì ð¡âðœð
ðâðœð
= ð( =
|ðž|4 2 + â |ð(ð)| Ì ) ð2 ðâ 0
|ðž|4 |ðž|4 + ððž (ð) â ððâ1 |ðž|2 + ð3ð ððž2 (0). ð ð
Peeling off the term ð(0), we see that â ð2 (ð) = ðâ 0
|ðž|4 |ðž|4 â |ðž|2 ððâ1 + ð3ð ððž2 (0) â ð2 (0) + ððž (ð). ð ð
Finally we need to establish that â|ðž|2 ððâ1 + ð3ð ððž2 (0) â ð2 (0) âª
|ðž|4 , ð
which follows immediately if ð3ð ððž2 (0) âª
|ðž|4 . ð
To this end, note that 2
Ë || ð 0 (ð) ððž (0) = â ||ðž(ð) ðâðœð ð
= ðâ2ð
â
ðž(ð¥)ðž(ðŠ)ð(âð¥ â
ð)ð(ðŠ â
ð)ð 0 (ð¥)
ð¥,ðŠ,ðâðœð ð
= ðð â ðž(ð¥)ðž(ðŠ)ð 0Ì (ð¥ â ðŠ). ð¥,ðŠâðœð ð
(3.17)
50
Chapter 3. The Iosevich-Rudnev bound Next, by Proposition 3.1.6, we have ð 0Ì (ð¥ â ðŠ) = ðâ1 ð¿(ð¥ â ðŠ) + ððð ðâ
ð+2 2
â ð (â ð â 0
ð+2 â 2
= ðâ1 ð¿(ð¥ â ðŠ) + ððð ð
âð¥ â ðŠâ ) 4ð
â ð (ð âð¥ â ðŠâ) ð â 0
= ðâ1 ð¿(ð¥ â ðŠ) + ððð ðâ
ð+2 2
â ð (ð âð¥ â ðŠâ) â ððð ðâ
ð+2 2
ð âðœð ð+2 â 2
= ðâ1 ð¿(ð¥ â ðŠ) + ððð ð
â
ðð¿(âð¥ â ðŠâ) â ððð ðâ
ð+2 2
by orthogonality. Note that ð¿(ð¥ â ðŠ) is the ð-dimensional delta function, while ð¿(âð¥ â ðŠâ) denotes the one-dimensional delta function. Our expression for ððž (0) is thus ððž (0) = ðŒ + ðŒðŒ + ðŒðŒðŒ, where ðŒ = ðâðâ1 â ð¿(ð¥ â ðŠ) = ðâðâ1 |ðž|, ð¥,ðŠâðœð ð
ðŒðŒ = ððð ðâ
3ð 2
â
3ð
ðž(ð¥)ðž(ðŠ) = ððð ðâ 2 ð(0), and
ð¥,ðŠâðœð ð
âð¥âðŠâ=0
ðŒðŒðŒ = ððð ðâ
3ð+2 2
â ðž(ð¥)ðž(ðŠ) = ððð ðâ
3ð+2 2
|ðž|2 .
ð¥,ðŠâðœð ð
Note that for each expression we have |ðŒ| = ðâðâ1 |ðž| †|ðž|2 ðâ 3ð
|ðŒðŒ| = ðâ 2 ð(0) †|ðž|2 ðâ |ðŒðŒðŒ| = ð
3ð+2 â 2
|ðž|2 †|ðž|2 ð
3ð+1 2 3ð+1 2
, and
3ð+1 â 2
,
using the trivial bound ð(0) †|ðž|2 and recalling that we are assuming 4ðð/2 †|ðž| †2ð
ð+1 2
. Finally it follows that
ð3ð ððž2 (0) ⪠ð3ð (|ðž|2 ðâ
3ð+1 2
2
) =
|ðž|4 , ð
3.3. Finite field spherical averages
51
which implies the following bound for (3.17): â ð2 (ð) ⪠ðâ 0
|ðž|4 |ðž|4 + ððž (ð). ð ð
By (3.10) we have achieved the lower bound |Î(ðž)| â«
|ðž|4 |ðž|4 ð
ððž (ð) +
|ðž|4 ð
â«
ð . ððž (ð) + 1
This completes the proof of Theorem 3.3.1 since if ððž (ð) ⪠1, then |Î(ðž)| â« ð.
3.3.1 Distance results for Salem sets In general, Salem sets are sets whose characteristic functions have maximal Fourier decay. Suppose that for ð â (0, . . . , 0) we have some bound of the form Ë || ⪠ððŒ |ðž|ðœ . ||ðž(ð) It follows that 2
Ë || ⪠ðð+2ðŒ |ðž|2ðœ + ðâð |ðž| = â ||ðž(ð) ð
|ðž|2 ð2ð
which implies that ðð+2ðŒ |ðž|2ðœ â«
|ðž| |ðž|2 |ðž| â 2ð ⥠ð . ðð ð ð
In general the best we can hope for is ðœ = 1/2 and ðŒ = âð. This leads us to the following definition. Definition 3.3.4. A set ðž â ðœðð is a Salem set if the bound Ë || ⪠ðâð â|ðž| ||ðž(ð) holds for all ð â ðœðð ⧵ {(0, . . . , 0)}. Theorem 3.3.5. If ðž â ðœðð is a Salem set with |ðž| ⥠ð¶ðð/2 for a sufficiently large constant ð¶, then |Î(ðž)| â« ð.
52
Chapter 3. The Iosevich-Rudnev bound
Proof. It is enough to show ððž (ð) ⪠1 by Theorem 3.3.1. Now, 2
Ë || ð ð¡ (ð) ððž (ð¡) = â ||ðž(ð) ðâðœð ð
⪠ðâ2ð |ðž| â ð ð¡ (ð) ðâðœð ð
= ðâ2ð |ðž||ð ð¡ |. Hence, ððž (ð) =
ð3ð+1 â ð2 (ð¡) |ðž|4 ð¡â 0 ðž
⪠ð3ð+1 |ðž|4 â ðâ4ð |ðž|2 |ð ð¡ |2 ð¡â 0 3ð+1
ð â
ðâ4ð |ðž|2 â
ð2ðâ1 |ðž|4 âª1 âª
when |ðž| â« ðð/2 as claimed.
3.4 Size and decay estimates for spheres We have proved Theorem 2.3.3 save for the proofs of Propositions 3.1.5 and 3.1.6, which we prove in detail now.
3.4.1 Proof of Proposition 3.1.5 We start with a few preliminaries. In what follows, we let ð(ð(ð¥) = ð¡) denote the number of solutions to the equation ð(ð¥) = ð¡. Recall that the function ð¿ ⶠðœð â {0, 1} is given by ð¿(ð¡) = {
1 0
ð¡=0 ð¡â 0
and ð ⶠðœð â {â1, 0, 1} is the quadratic character of ðœð : 1 ð(ð¡) = { 0 â1
ð¡ is a nonzero square ð¡=0 ð¡ is not a square.
3.4. Size and decay estimates for spheres
53
ð¡
If ð is prime, then ð(ð¡) = ( ð ) is called the Legendre symbol. The Legendre symbol ıis a multiplicative character on the (multiplicative) group ðœâð with the added condition that ð(0) = 0. Note that this implies that for ð¥, ðŠ â ðœâð , we have ð(ð¥ðŠ) = ð(ð¥)ð(ðŠ) and that â ð(ð) = â ð(ð) = 0. ðâðœð
ðâðœð â
See (3.19) below. We further define ð(ð¡) = ð¿(ð¡)ð â 1 = {
ðâ1 â1
ð¡=0 ð¡ â 0.
In a few instances, we will complete the square, so recall that we must assume that ð ⥠3 is odd. We next lay out a series of simple but useful equalities involving the above functions. Proposition 3.4.1. Let ð, ð¡, ð â ðœð . Then the following hold: 2 ð(ð¥ = ð) = 1 + ð(ð) = { 1 0 2
ð is a nonzero square ð=0 ð is not a square
â ð(ð) = 0
(3.18)
(3.19)
ðâðœð
â ð(ð¡) = 0
(3.20)
ð¡âðœð
â ð¿(ð) = 1
(3.21)
ðâðœð
(ð â ð)(ð¡) = â ð(ð)ð(ð¡ â ð) = ð â
ð(ð¡)
(3.22)
ðâðœð
â ð(ð2 + 4ðð) = ð(ð)
(3.23)
ðâðœð
â ð(ð¡ð â ð2 ) = ð(â1)ð(ð¡)
(3.24)
ðâðœð
â ð(ð)ð(ð¡ â ð) = ðð(ð¡). ðâðœð
(3.25)
54
Chapter 3. The Iosevich-Rudnev bound
Proof. We will leave (3.18), (3.20), (3.21), and (3.22) as exercises. Note that (3.24) follows from (3.23). Now, (3.19) follows by orthogonality as ð is a character over the group ðœâð . Alternatively, â ð(ð) = â (1 + ð(ð) â 1) = â (ð(ð¥2 = ð) â 1) = ð â ð = 0. ðâðœð
ðâðœð
ðâðœð
The equality â ð(ð¥2 = ð) = ð ðâðœð
follows from the fact that if ð is a nonzero square, then ð¥2 = ð has two distinct solutions, say ð and âð. Thus, each nonzero element in ðœâð is a solution to some equation ð¥2 = ð for a unique ð. It follows that there ðâ1 ðâ1 are 2 values ð such that ð¥2 = ð has two solutions. In other words, 2 elements in ðœð are nonzero squares. The rest now follows easily. Finally, we prove (3.23). Now, â ð(ð2 + 4ð¡ð) = â ð((ð + 2ð¡)2 â 4ð¡2 ) ðâðœð
ðâðœð
= â ð(ð2 â 4ð¡2 ) ðâðœð
= â [1 + ð(ð2 â 4ð¡2 ) â 1] ðâðœð
= ð(ð¥2 = ð2 â 4ð¡2 ) â ð. In Exercise 3.4, you are asked to show that ð(ð¥2 â ð2 = ð¡) = ð + ð(ð¡). This completes the proof. We are finally ready to prove Proposition 3.1.5. When ð = 2, by (3.18) we have |ð ð¡ | = â ð(ð¥12 = ð)ð(ð¥22 = ð) ð+ð=ð¡
= â ð(ð¥12 = ð)ð(ð¥22 = ð¡ â ð) ðâðœð
= â [1 + ð(ð)][1 + ð(ð¡ â ð)]. ðâðœð
3.4. Size and decay estimates for spheres
55
Now because ð is multiplicative we have [1 + ð(ð)][1 + ð(ð¡ â ð)] = 1 + ð(ð) + ð(ð¡ â ð) + ð(ð¡ð â ð2 ). Note that by (3.19), the sum over ð(ð) and ð(ð¡ â ð) is zero. Hence |ð ð¡ | = â [1 + ð(ð¡ð â ð2 )] ðâðœð
= ð + â ð(ð¡ð â ð2 ) ðâðœð
= ð + ð(â1)ð(ð¡), where the last line followed from (3.24). For even values ð ⥠2 we apply induction with the case ð = 2 above being our base case. For ð ⥠4 we have 2 2 |ð ð¡ | = â ð(ð¥12 + ⯠+ ð¥ðâ2 = ð)ð(ð¥ðâ1 + ð¥ð2 = ð¡ â ð) ðâðœð
= â (ððâ3 + ð(ð)ð ((â1)
ðâ2 2
)ð
ðâ4 2
) â
(ð + ð(â1)ð(ð¡ â ð)) ,
ðâðœð
since 2 ð(ð¥12 + ⯠+ ð¥ðâ2 = ð) = ððâ3 + ð(ð)ð ((â1)
ðâ2 2
)ð
ðâ4 2
by the inductive hypothesis and 2 ð(ð¥ðâ1 + ð¥ð2 = ð¡ â ð) = ð + ð(â1)ð(ð¡ â ð)
by the base case. It follows that |ð ð¡ | = â (ððâ2 + ð(ð)ð ((â1)
ðâ2 2
)ð
ðâ4 2
ð(â1)ð(ð¡ â ð))
ðâðœð ð
= ððâ1 + ð ((â1) 2 ) ð
ðâ4 2
â ð(ð)ð(ð¡ â ð) ðâðœð
ð
= ððâ1 + ð ((â1) 2 ) ð
ðâ2 2
ð(ð¡)
56
Chapter 3. The Iosevich-Rudnev bound
where the last line follows from (3.19). This established Proposition 3.1.5 when ð ⥠2 is even. When ð ⥠3 is odd, 2 |ð ð¡ | = â ð(ð¥12 + . . . ð¥ðâ1 = ð)ð(ð¥ð2 = ð¡ â ð) ðâðœð
= â [ððâ2 + ð ((â1)
ðâ1 2
)ð
ðâ3 2
ð(ð)] (1 + ð(ð¡ â ð)) .
ðâðœð
Distributing the quantity in the sum, and applying (3.20) and (3.19) (and a change of variables), we see that |ð ð¡ | = ððâ1 + ð
ðâ3 2
ð ((â1)
ðâ1 2
ð ((â1)
ðâ1 2
) â ð(ð)ð(ð¡ â ð) ðâðœð
= ððâ1 + ð
ðâ1 2
ð¡) ,
by (3.25). This completes the proof of Proposition 3.1.5.
3.4.2 Proof of Proposition 3.1.6 Letâs jump right in. First, write ð ð¡Ì (ð) = ðâð â ð ð¡ (ð¥)ð(âð¥ â
ð) ð¥âðœð ð
= ðâðâ1 â â ð(ð âð¥â â ð ð¡)ð(âð¥ â
ð) ð âðœð ð¥âðœð ð
=
ð¿(ð) + â â ð(ð âð¥â â ð ð¡)ð(âð¥ â
ð). ð ð ð â 0 ð¥âðœð
Breaking apart the sum in ð¥ = (ð¥1 , . . . , ð¥ð ) into the sum of its component pieces, we see that ð
ð ð¡Ì (ð) =
ð¿(ð) + ðâðâ1 â ð(âð ð¡) â â ð(ð ð¥ð2 â ð¥ð ðð ) ð ð â 0 ð=1 ð¥ âðœ ð
ð
ð
=
ð 2 ð2 ð¿(ð) + ðâðâ1 â ð(âð ð¡) â â ð (ð (ð¥ð â ð ) â ð ) ð 2ð 4ð ð â 0 ð=1 ð¥ âðœ ð
ð
ð
=
ð2 ð¿(ð) + ðâðâ1 â ð(âð ð¡) â â ð(ð ð¥ð2 )ð (â ð ) . ð 4ð ð â 0 ð=1 ð¥ âðœ ð
ð
3.5. Finite field counterexample
57
Applying (3.11) yields ð ð¡Ì (ð) = =
ð21 + ⯠+ ð2ð ð¿(ð) + ðâðâ1 ððð ðð/2 â ð (âð ð¡ â ) ð 4ð ð â 0 ð+2 âðâ ð¿(ð) + ððð ðâ 2 â ð (âð ð¡ â ). ð 4ð ð â 0
ð
Notice that for ð, ð â ðœâð we are using the notation ð to mean ððâ1 . In the sum above, the exact value of ðð is surprisingly difficult to compute, but it turns out (see [90, 104]) that if ð = ðâ for an odd prime ð, then (â1)ââ1
ðâ¡1
(mod 4)
(â1)ââ1 ðâ
ðâ¡3
(mod 4).
ðð = { The two inequalities in Proposition 3.1.6 will follow from the following deep result of Weil ([154]) on so-called Kloosterman sums, finishing our proof of the proposition. Theorem 3.4.2. Define ðŸ(ð, ð) = â ð (ðð + ðð â1 ) . ð â 0
Then, |ðŸ(ð, ð)| †2âð if ðð â 0. Note that ðŸ(0, 0) = ð â 1, while ðŸ(0, ð¡) = ðŸ(ð¡, 0) = â1 for ð¡ â 0. The proof of Theorem 3.4.2 is beyond the scope of this book. For an elementary1 proof, see [136].
3.5 Finite field counterexample In this section we prove the following counterexample to the ErdÅsFalconer distance conjecture constructed in [76]. ð+1
Theorem 3.5.1. Let ð ⥠3 be odd. Then the exponent 2 in Theorem 2.3.3 is best possible. In particular for any ð > 0, there exists a set ðž â ðœðð such that |ðž| â ð
ð+1 âð 2
and yet the distance set satisfies |Î(ðž)| ⪠ð1âð .
1 An elementary proof is a proof which only relies on basic techniques. It does not mean that the proof is easy to follow! Schmidtâs approach to the proof (following work of Stepanov and Thue) is both elegant and intricate.
58
Chapter 3. The Iosevich-Rudnev bound
We will call ð£ â ðœðð a null-vector if ð£ â (0, . . . , 0) and yet ð£ â
ð£ = 0. We start with a preliminary lemma. Proposition 3.5.2. Let ð ⥠4 be even. Then, there exist at least ð/2 mutually orthogonal null vectors in ðœðð . Proof. If ð â¡ 1 (mod 4), then there exists an element ð â ðœð such that ð2 = â1. Then, let ð£ð = (0, . . . , 0, 1, ð, . . . , 0), where there is a 1 in the (2ðâ1)-st coordinate, and an ð in the 2ð-th coordinate. This produces ð/2 mutually orthogonal null vectors in ðœðð . If ð â¡ 3 (mod 4), we will need to be more creative. Suppose that ð â¡ 0 (mod 4). Then we will take ð£ 1 = (ð, ð, ð, 0, . . . , 0) and ð£ 2 = (0, âð, ð, ð, 0, . . . , 0). Note that ð£ 1 â
ð£ 2 = 0 for all ð, ð, ð â ðœð . Next, we take ð£ 3 = (0, 0, 0, 0, ð, ð, ð, 0, . . . , 0) and ð£ 4 = (0, 0, 0, 0, 0, âð, âð, ð, 0 . . . , 0), and so on. Note that there always exists a triple (ð, ð, ð) â ðœ3ð such that ð2 + ð2 + ð2 = 0 and ððð â 0 (see Exercise 3.13). This provides ð/2 mutually orthogonal null vectors when ð â¡ 0 (mod 4). It remains to treat the case ð â¡ 2 (mod 4). We first handle ð = 6 by simply taking ð£ 1 = (ð, ð, ð, 0, 0, 0) ð£ 2 = (âð, ð, 0, ð, 0, 0) ð£ 3 = (0, âð, ð, ð, 0, 0). All higher dimensions ð â¡ 2 (mod 4) can be handled similarly. For example when ð = 10, we may take ð£ 1 = (ð, ð, ð, 0, 0, 0, 0, 0, 0, 0) ð£ 2 = (âð, ð, 0, ð, 0, 0, 0, 0, 0, 0) ð£ 3 = (0, âð, ð, ð, 0, 0, 0, 0, 0, 0) ð£ 4 = (0, 0, 0, 0, ð, ð, ð, 0, 0, 0) ð£ 5 = (0, 0, 0, 0, 0, âð, ð, 0, 0, 0). This completes the proof of Proposition 3.5.2. Now let ð ⥠3 be odd, so that ð = 2ð + 1 for some ð ⥠1. We have shown that there exist at least ð mutually orthogonal null-vectors whose
3.6. Relations to the Falconer problem
59
last coordinate is 0. Suppose we take ðŽ â ðœð to be any arithmetic proð+1 gression of length ð †2 , and let ð ð = (0, . . . , 0, 1). Finally, put ðž = {ð¡ ð ð£ ð + ðð ð ⶠð â ðŽ, and ð = 1, . . . , ð, where ð¡ ð runs through ðœð }. First, note that |ðž| = ðð â
ð, since for each of the ð vectors ð£ ð , the quantity ð¡ ð runs through ðœð . Next, note that since ð£ ð â
ð£ð = 0 for all 1 †ð †ð †ð, it follows that for ð¥, ðŠ â ðž we have âð¥ â ðŠâ = âð¡1 ð£ 1 + ðð ð â ð¡2 ð£ 2 â ðâ² ð ð â = (ð â ðâ² )2 ,
(3.26)
by a direct calculation (see Exercise 3.10), and thus |Î(ðž)| †2ð â 1. We leave the remainder of the proof of Theorem 3.5.1 as an exercise (Exercise 3.14).
3.6 Relations to the Falconer problem Finally we should mention that the approaches to the ErdÅs-Falconer distance problem have been heavily influenced by some of the approaches to the Falconer distance problem, and it will be instructive to (very) briefly outline the strategies employed in the Falconer distance problem, though one should not take the following integrals and inequalities too seriously. Kenneth Falconerâs original proof of his result concerning the Falconer distance problem centered around the observation that ð à ð {(ð¥, ðŠ) â âð à âð ⶠ1 †|ð¥ â ðŠ| †1 + ð} ⪠ð, for an appropriate probability measure2 ð, and this estimate in turn utilized the stationary phase estimate |1Ìð (ð)| ⪠|ð|â
ð+1 2
,
(3.27)
where 1ð denotes the characteristic equation of the sphere ð ðâ1 â âð . This idea essentially reduces to computing the number of times a distance is repeated, and this was our first line of attack (from Section 3.1). 2 Loosely, a probability measure on a set ðž is a measure such that ð(ðž) = 1. These particular measures used by Falconer are called Frostman measures.
60
Chapter 3. The Iosevich-Rudnev bound
Furthermore, Mattila showed that in order to prove Falconerâs conjecture it is enough to get a bound of the form â
2
2
â« (â« ||ð Ë(ð¡ððð )|| ðð) ð¡ ðð¡ < â
(3.28)
1
where ð is again the Frostman measure ([109]). The integral (3.28) will be referred to simply as ð(ð). Again, do not worry at all about understanding this integral, perhaps other than to notice that an ð¿2 -bound on this integral gives a distance set result. This was the motivation for having examined the finite field Mattila sum ððž (ð) from Section 3.3. The ubiquity of the connections between continuous problems and finite field analogues can be surprising, and they will continue to be highlighted throughout the remainder of the book.
3.7 Exercises: Chapter 3 Exercise 3.1. Suppose ð = ðð , and let ðð ⶠðœð â ðœð be the field trace, i.e., the function defined by 2
Tr(ð¥) = ð¥ + ð¥ð + ð¥ð + ⯠+ ð¥ð
ðâ1
.
Prove that ð(ð¥) = ð2ðð Tr(ð¥)/ð is a group homomorphism from ðœð to the set ââ = â ⧵ {0}, thereby establishing that ð(ð¥) is a character on ðœð . Exercise 3.2. Prove Proposition 3.1.3. That is, prove that we have Ë â
ð) ð(ð¥) = â ð(ð)ð(ð¥ ðâðœð ð
for any function ð ⶠðœðð â â. Exercise 3.3. Prove that we have Ë ð(ð) â ð(ð) Ì = ðâð â ð(ð¥)ð(ð¥) ðâðœð ð
ð¥âðœð ð
for all functions ð, ð ⶠðœðð â â. This establishes Proposition 3.1.4. Exercise 3.4. Let ð»(ð¡) = |{(ð¥, ðŠ) â ðœð ⶠð¥2 â ðŠ2 = ð¡}|. Show that ð»(ð¡) = {
2ð â 1 ðâ1
ð¡=0 ð¡ â 0.
3.7. Exercises: Chapter 3
61
Exercise 3.5. Prove (3.18) by showing that the number of solutions to the equation ð¥2 = ð is given by 1 + ð(ð), where ð is the quadratic character of ðœð . Exercise 3.6. Prove equation (3.20). Exercise 3.7. Verify that â ð¿(ð + ð) = 1 ðâðœð
holds for any ð â ðœð by a direct calculation. This generalizes (3.21). Exercise 3.8. Prove that (ð â ð)(ð¡) = âð ð(ð)ð(ð¡ â ð) = ð â
ð(ð¡), verifying (3.22). Exercise 3.9. Prove (3.25). That is, show that â ð(ð)ð(ð¡ â ð) = ðð(ð¡). ðâðœð
Exercise 3.10. Verify equation (3.26) by a direct calculation. Exercise 3.11. Prove Proposition 3.3.3. More precisely, for fixed ð¥ â ðœðð , define ð¢ð¥ ⶠðœð â â given by ð¢ð¥ (ð) = ðð (ð¥). Show that we have ð¢ð¥Ì (ð) =
1 ð(âðâð¥â). ð
Exercise 3.12. Prove the following generalization of (3.7). Show for functions ð, ð ⶠðœðð â â, that â ð(ð¥)ð(ðŠ)ð ð¡ (ð¥ â ðŠ) = ð¡âðœð
where |ð
(ð, ð)| †2ð
ðâ1 2
|ð ð¡ | âðâ1 âðâ1 + ð
(ð, ð), ðð
âðâ2 âðâ2 . Here, we are writing 1/ð
âðâð = ( â |ð(ð¥)|ð )
.
ð¥âðœð
Exercise 3.13. Show that when ð â¡ 3 (mod 4), the equation ð2 + ð2 + ð2 = 0 has a nontrivial solution (ð, ð, ð) â ðœâð à ðœâð à ðœâð .
62
Chapter 3. The Iosevich-Rudnev bound
Exercise 3.14. Finish the proof of Theorem 3.5.1. More precisely, given any ð â (0, 1), construct a set ðž such that |ðž| â ð ð1âð .
ð+1 âð 2
and yet |Î(ðž)| âª
Exercise 3.15. Show that two distinct spheres in ðœ2ð intersect in at most two points. What can we say about the intersection of spheres in higher dimensions? Exercise 3.16. Strengthen our proof of Theorem 2.3.3. Show that if ðž â ðœðð satisfies |ðž| > 2ð for |ð ð¡ | and ð ð¡Ì (ð).
ð+1 2
, then Î(ðž) = ðœð by using more precise estimates
10.1090/car/037/04
Chapter
4
Wolffâs exponent 4.1 Introduction We previously mentioned that Tom Wolff gave a â4/3-resultâ toward the Falconer distance problem (see Section 2.2). In this chapter, we prove a 4/3-result toward the finite field distance problem. Namely we will discuss the following improvement to the distance problem in two dimensions, first proved in [17]. Theorem 4.1.1. Suppose that ðž â ðœ2ð with |ðž| ⥠ð4/3 , where ð is sufficiently large. Then there exists a constant ð > 0 such that |Î(ðž)| ⥠ðð. Proof of Theorem 4.1.1. We will use the ð¿2 -methods outlined in Section 3.2. Whether we use (3.8) or (3.9) will depend on whether ð â¡ 1 (mod 4) or ð â¡ 3 (mod 4). When ð â¡ 3 (mod 4), we will utilize (3.8): |Î(ðž)| â¥
|ðž|4 , âð¡ ð(ð¡)2
since in this case there do not exist nontrivial circles of radius zero in two dimensions. When ð â¡ 1 (mod 4), we will need to account for such isotropic spheres. Theorem 4.1.1 will follow from the following upper bound, which we achieve when ð â¡ 3 (mod 4) and |ðž| ⥠ð4/3 : â ð(ð¡)2 †(1 + â3)ðâ1 |ðž|3 .
(4.1)
ð¡
Thus it remains to prove (4.1). First we apply the following result. 63
64
Chapter 4. Wolffâs exponent
Theorem 4.1.2. For all odd values ð and for ðž â ðœ2ð , we have the equality 2
â ð(ð¡)2 = ð6 â (ððž (ð¡)) + Î(ð, ðž), ð¡âðœð
ð¡âðœð
where Î(ð, ðž) =
|ðž|4 â ð|ðž|2 , ð
and where
2
Ë || ððž (ð¡) = â ||ðž(ð) ðâðð¡ 1
is the circular average of ðž which we encountered in the previous chapter. We will prove Theorem 4.1.2 shortly. The bound (4.1) will follow from combining Theorem 4.1.2 with the following remarkable bound derived by Iosevich and Koh ([83]). Theorem 4.1.3. If ðž â ðœ2ð , then we have that 2
Ë || †â3ðâ3 |ðž|3/2 . max â ||ðž(ð) ð¡â 0
âðâ=ð¡
The proof of Theorem 4.1.3 lies just beyond the scope of the book. The motivated reader is encouraged to look up the original source. However, we will next show that the combination of Theorems 4.1.2 and 4.1.3 is sufficient to achieve (4.1). We start by pulling out the term ð = 0,â and then used Proposition 1.3.4. Thus, 4
2
Ë 0)|| + ð6 (max ððž (ð¡)) â ||ðž(ð) Ë || + Î(ð, ðž) â ð(ð¡)2 †ð6 ||ðž(0, ð¡â 0
ð¡âðœð
ðâ (0,0)
†ð6 ðâ8 |ðž|4 + ð6 â3ðâ3 |ðž|3/2 â
(ðâ2 |ðž| â ðâ4 |ðž|2 ) + Î(ð, ðž) = ðâ2 |ðž|4 + â3ð|ðž|5/2 â â3ðâ1 |ðž|5/2 + ðâ1 |ðž|4 â ð|ðž|2 †ðâ1 |ðž|4 + â3ð|ðž|5/2 †(1 + â3)ðâ1 |ðž|4 . 1 We previously called this quantity the spherical average, though we use circles to refer to 2-dimensional spheres.
4.2. Proof of ð¿2 estimate for ð(ð¡)
65
In the final steps we grouped everything together and used that |ðž| ⥠ð4/3 . Putting everything together, we have shown that when ð â¡ 3 (mod 4) and ðž â ðœ2ð satisfies |ðž| ⥠ð4/3 , then |Î(ðž)| â¥
1 1 + â3
ð.
When ð â¡ 1 (mod 4), the proof of Theorem 4.1.1 is similar, though we must use the inequality (3.9) since in this case, there exist nontrivial circles of radius 0. We leave it to the motivated reader to complete the proof in the case ð â¡ 1 (mod 4), using the original paper [17] as a guide. When ð â¡ 1 (mod 4), the authors ([17]) actually showed that if |ðž| ⥠ð4/3 , then |Î(ðž)| ⥠ðð â
ð, where 2
ðð =
(1 â 2/ð)
1 + â3 â â3/ð2/3
Notice that ðð is increasing if ð ⥠2, ðð â ðð â¥
1 3
. 1
1+â3
as ð â â, and that
if ð ⥠25.
4.2 Proof of ð¿2 estimate for ð(ð¡) We still need to prove Theorem 4.1.2. Recall that ð(ð¡) is our counting function: ð(ð¡) = |{(ð¥, ðŠ) â ðž à ðž ⶠâð¥ â ðŠ|| = ð¡}|. We start by applying our bound (3.4): 2
Ë || ð ð¡Ì (ð) ð(ð¡) = ð4 â ||ðž(ð) ðâðœ2ð
so that ð2 (ð¡) = ð8
â
2
2
Ë || ||ðž(ð Ë â² )|| ð ð¡Ì (ð)ð ð¡Ì (ðâ² ) ||ðž(ð)
ð,ðâ² âðœ2ð
= ðŽ(ð¡) + ðµ(ð¡) + ð¶(ð¡)
66
Chapter 4. Wolffâs exponent
where ðŽ(ð¡) = ðâ4 |ðž|4 |ð ð¡ |2 ðµ(ð¡) = 2ð8
â
2
2
Ë || ||ðž(0, Ë 0)|| ð ð¡Ì (ð)ð ð¡Ì (0, 0), and ||ðž(ð)
ðâðœ2ð ⧵{(0,0)}
ð¶(ð¡) = ð8
2
2
Ë â² )|| ð ð¡Ì (ð)ð ð¡Ì (ðâ² ). Ë || ||ðž(ð ||ðž(ð)
â ð,ðâ² âðœ2ð ⧵{(0,0)}
Note that ðµ(ð¡) contains both the sums when ð = (0, 0) and ðâ² â (0, 0) as well as ðâ² = (0, 0) and ð â (0, 0), so we can combine the sums by symmetry. Hence, â ð(ð¡)2 = â [ðŽ(ð¡) + ðµ(ð¡) + ð¶(ð¡)] . ð¡âðœð
ð¡âðœð
The easiest term to handle is â ðŽ(ð¡) = ðâ4 |ðž|4 â |ð ð¡ |2 . ð¡âðœð
ð¡âðœð
We will need the following averaging results for the sphere, whose proofs we delay until the end of the chapter. Proposition 4.2.1. Let ð ð¡ â ðœðð . Then we have â |ð ð¡ |2 = ð2ðâ1 + ðð â ððâ1 . ð¡âðœð
Moreover, if ð, ðâ² â 0,â then we have âðâ1 â ð ð¡Ì (ð)ð ð¡Ì (ðâ² ) = ð2ð [â1 + ðð¿(âðâ â âðâ² â)] ð ð ð¡
where ð¿(â
) is the 1-dimensional delta function. If ð ⥠2 is even and ð â 0,â then â |ð ð¡ |ð ð¡Ì (ð) = ð¿(âðâ) â ðâ1 . ð¡âðœð
Note 4.2.2. For reference recall that (â1)ââ1
ð = ðâ , ð â¡ 1
(mod 4)
(â1)ââ1 ðâ
ð = ðâ , ð â¡ 3
(mod 4).
ðð = {
4.2. Proof of ð¿2 estimate for ð(ð¡)
67
In particular notice that ð4ð = 1 for all ð, and 1
ðâ¡1
(mod 4)
â1
ðâ¡3
(mod 4).
ð2ð = { From Proposition 4.2.1 it follows that â ðŽ(ð¡) = |ðž|4 (ðâ1 + ðâ2 â ðâ3 ) .
(4.2)
ð¡
By Proposition 4.2.1 if ð = 2 and ð â (0, 0), then we have â |ð ð¡ |ð ð¡Ì (ð) = ð¿(âðâ) â ðâ1 . ð¡âðœð ð
We are ready to handle the second sum. By the results above we have 2
Ë || ð ð¡Ì (ð) â ðµ(ð¡) = 2ð8 ðâ6 |ðž|2 â |ð ð¡ | â ||ðž(ð) ð¡
ð¡âðœð
ðâ (0,0) 2
Ë || â |ð ð¡ |ð ð¡Ì (ð) â ||ðž(ð)
= 2ð2 |ðž|2
ð¡
ðâ (0,0)
= 2ð2 |ðž|2
2
Ë || â 2ð|ðž|2 (ðâ2 |ðž|4 â ðâ4 ) ||ðž(ð)
â ðâ (0,0),âðâ=0
2
Ë || â 2ðâ2 |ðž|4 + 2ðâ3 |ðž|4 â 2ðâ1 |ðž|3 . = 2ð2 |ðž|2 â ||ðž(ð) âðâ=0
Finally, we examine â ð¶(ð¡) = ð8 â
â
2
2
Ë â² )|| ð ð¡Ì (ð)ð ð¡Ì (ðâ² ). Ë || ||ðž(ð ||ðž(ð)
ð¡âðœð ð,ðâ² âðœ2ð ⧵{(0,0)}
ð¡âðœð
From Proposition 4.2.1 and Note 4.2.2, we have â ð ð¡Ì (ð)ð ð¡Ì (ðâ² ) = ðâ2 ð¿(âðâ â âðâ² â) â ðâ3 . ð¡
68
Chapter 4. Wolffâs exponent
Hence, â ð¶(ð¡) = ð5 ð¡âðœð
2
2
Ë || ||ðž(ð Ë â² )|| (â1 + â ð (ð (âðâ â âðâ² â))) â ||ðž(ð) ð
ð,ðâ² â 0â
2
= ð6
2
2
2
Ë || ||ðž(ð Ë â² )|| â ð5 ( â ||ðž(ð) Ë || ) . ||ðž(ð)
â ð,ðâ² â 0â âðâ=âðâ² â
ðâ 0â
For the first sum, we use the same expression we derived in (3.16). For the second sum, we add and subtract in the term corresponding to ð = 0â and apply Plancherel to see that 2
â â 2â Ë â ð¶(ð¡) = ð6 â â â |ðž(ð)| â ð5 (ðâ2 |ðž| â ðâ4 |ðž|2 )2 â ð¡âðœð ð¡âðœð ââðâ=ð¡ â ðâ 0â â 2
= ð6 (
2
2 Ë |ðž(ð)| ) + ð6 â (ððž (ð¡)) + Î(ð, ðž)
â
ð¡âðœâð
ðâð0 ⧵{(0,0)} 2
2
= ð6 (ððž (0) â ðâ4 |ðž|2 ) + ð6 â (ððž (ð¡)) + Î(ð, ðž) ð¡âðœâð 2
= ð6 â (ððž (ð¡)) â 2ð2 |ðž|2 ððž (0) + ðâ2 |ðž|4 + Î(ð, ðž), ð¡âðœð
where Î(ð, ðž) = âð|ðž|2 +2ðâ1 |ðž|3 âðâ3 |ðž|4 . Putting together the above estimates, finally gives us 2 2
Ë || ) + â ð(ð¡) = ð â ( â ||ðž(ð) 2
ð¡âðœð
6
ð¡âðœð
âðâ=ð¡
|ðž|4 â ð|ðž|2 . ð
This completes the proof of Theorem 4.1.2.
4.2.1 Proof of Proposition 4.2.1 We will prove Proposition 4.2.1 only in the case ð = 2. We leave the general case as an exercise. Let ð ð¡ â ðœðð . Recall that ð(ð¥) = 1 if ð¥ â ðœâð is a nonzero square, ð(ð¥) = â1 if ð¥ â ðœâð is not a square, and ð(0) = 0.
4.2. Proof of ð¿2 estimate for ð(ð¡)
69
Moreover, recall that ð(0) = ð â 1 and ð(ð¡) = â1 for ð¡ â 0. In particular, we will use that â ð(ð¡) = 0. ð¡âðœð
Now we have that â |ð ð¡ |2 = â(ð + ð(â1)ð(ð¡))2 ð¡
ð¡
= â(ð2 + ðð(â1)ð(ð¡) + ð(ð¡)2 ) ð¡
= ð3 + ðð(â1) â ð(ð¡) + (ð â 1)2 + ð â 1 ð¡
= ð3 + ð2 â ð. Next, we look at â ð ð¡Ì (ð)ð ð¡Ì (ðâ² ). ð¡
Recall from Section 3.4.2 that if ð = 2 and ð â (0, 0), we have ð ð¡Ì (ð) = ð2ð ðâ2 â ð (ð ð¡ + ð â 0
âðâ ). 4ð
Note that ðð â {±1, ±ð}, so that ð4ð = 1. Hence for ð = 2 and ð, ðâ² â (0, 0), we have â ð ð¡Ì (ð)ð ð¡Ì (ðâ² ) = ð4ð ðâ4 â ð (ð ð¡ + ð¡
ð â 0
âðâ âðâ² â ) â ð (âð â² ð¡ â ) 4ð ð â² â 0 4ð â²
= ðâ4 â â ð(ð¡(ð â ð â² ))ð ( ð ,ð â² â 0 ð¡
= ðâ3 â ð ( ð â 0
âðâ âðâ² â â ) 4ð 4ð â²
âðâ âðâ² â â ). 4ð 4ð
Again applying a change of variables, we have â ð ð¡Ì (ð)ð ð¡Ì (ðâ² ) = ðâ3 â ð(ð (âðâ â âðâ² â)) ð¡
ð â 0
= ðâ3 [â1 + â ð(ð (âðâ â âðâ² â))] ð
= âð
â3
â2
+ ð ð¿(âðâ â âðâ² â),
70
Chapter 4. Wolffâs exponent
where ð¿(â
) is the (1-dimensional) delta function. Finally we examine â |ð ð¡ |ð ð¡Ì (ð). ð¡âðœð
For ð â (0, 0) we have â |ð ð¡ |ð ð¡Ì (ð) = ð2ð ðâ2 â ð (â ð¡
ð â 0
âðâ ) â
â (ð + ð(â1)ð(ð¡)) ð(âð ð¡) 4ð ð¡
= ð2ð ðâ2 ð(â1) â ð (â ð â 0
âðâ ) â (ð(ð¡)ð(âð ð¡)) . 4ð ð¡
Applying the definition of ð and using orthogonality multiple times (along with a change of variables), we see that â |ð ð¡ |ð ð¡Ì (ð) = ðâ2 â ð (â ð¡
ð â 0
= ðâ2 â ð (â ð â 0
= ðâ1 â ð (â ð â 0
âðâ ) [ð â 1 â â ð(âð ð¡)] 4ð ð¡â 0 âðâ ) [ð â â (ð(âð ð¡))] 4ð ð¡ âðâ ) 4ð
= ðâ1 â ð (ð âðâ) ð â 0
1 = ð¿(âðâ) â . ð
4.3 Restriction and extension theory As before it is instructive to try to understand the connections between the continuous restriction problem and the finite field analogue. This section will introduce us to these ideas, but the standard disclaimer applies: Do not take this section too seriously, and it is okay to skip this section if you would like. Before we delve into restriction theory, we first want to establish the notation used in restriction and extension theory.
4.3. Restriction and extension theory
71
4.3.1 Fourier transform notation for finite fields It will be helpful to discuss some notation at this time. For ð ⶠðœðð â â, Ë ð, Ë and ðâš for the normalized Fourier transform, the nonwe write ð, normalized Fourier transform, and the inverse Fourier transform, respectively. More precisely, if ð is the canonical additive character over ðœð , then we define Ë ð(ð) = ðâð â ð(ð¥)ð(âð â
ð¥), ð¥âðœð ð
Ë Ë ð(ð) = â ð(ð¥)ð(âð â
ð¥) = ðð ð(ð), ð¥âðœð ð
and Ë Ë ðâš (ð) = â ð(ð¥)ð(ð â
ð¥) = ð(âð) = ðð ð(ð). ð¥âðœð ð
We will write
1/ð ð
âðâð¿ð (ðœðð ,ðð) = ( â |ð(ð)| ) ðâðœð ð
while
1/ð
âðâð¿ð (ðœðð ,ðð¥) = (ðâð â |ð(ð¥)|ð ) ð¥âðœð ð
and âðâð¿ð (ð,ðð)
1 â |ð(ð¥)|ð ) =( |ð| ð¥âð
1/ð
where ð â ðœðð is any subset sphere of some given radius, and where ðð denotes the surface measure of ð which we can think of as the function ðð ðð(ð¥) = |ð| ð(ð¥), again using ð(ð¥) as the characteristic function of the set ð. Finally for a set ðž â ðœðð , we will write âðâð¿â (ðž,ðð¥) = ðâð max |ð(ð¥)|. ð¥âðž
This notation is economical as we can now write the Plancherel identity as âðËâ = âðâð¿2 (ðœ2ð ,ðð¥) . â â 2 ð¿2 (ðœð ,ðð)
72
Chapter 4. Wolffâs exponent
We further define the (additive) convolution of ð and ð as ð â ð(ð¥) = â ð(ðŠ)ð(ð¥ â ðŠ). ðŠâðœð ð
It follows easily that Ë Ë ð(ð) ð â ð(ð) = ð(ð) Ì
(4.3)
Ë = ðË â ð.Ì ðð
(4.4)
and We leave these last two equalities as an exercise (see Exercise 4.2). The interested reader can consult the Proof of Lemma 4.4 in [17] for a more Ë ð, Ë and ðâš along with some of the in-depth discussion of the objects ð, underlying mathematics.
4.3.2 Restriction theory The finite field restriction problem refers to establishing the following inequality. We use the notations developed in Section 4.3.1. Problem 4.3.1 (Finite Field Restriction Problem). Determine the values 1 †ð, ð †â such that there exists a constant ð¶ð,ð,ð so that for all functions ð ⶠðœðð â â, we have Ë ð¿ð (ð,ðð) âªð,ð,ð âðâ ð ð âðâ ð¿ (ðœð ,ðð) .
(4.5)
By duality this is equivalent to asking for â(ððð)âš âð¿ðâ² (ðœðð ,ðð) âªð,ð,ð âðâð¿ðâ² (ð,ðð) .
(4.6)
We will use the notation ð
(ð â ð) and ð
â (ðâ² â ðâ² ) to denote the best constants appearing in (4.5) and (4.6), respectively. The restriction problem has been intricately and intimately tied to many problems in geometric combinatorics including the distance problem. We next note some straight-forward estimates. Proposition 4.3.2. We have ð
â (ð1 â ð) †ð
â (ð2 â ð)
for
ð2 †ð1
ð
â (ð â ð1 ) †ð
â (ð â ð2 )
for
ð2 †ð1 .
and
4.4. Exercises: Chapter 4
73
Furthermore, we have ð
â (ð â â) = 1
for
1 †ð †â,
(4.7)
ð
â (ð â 2) ⌠âð
for
2 †ð †â.
(4.8)
and
We will leave the proof of this proposition as an exercise (see Exercise 4.1). There are many results in restriction theory which have been proven in the setting of finite fields ([84, 85, 105â107, 111]). Theorem 4.1.3 can be and should be properly viewed as a restriction inequality in the case ð(ð¥) = ðž(ð¥) for some set ðž. It is instructive to compare the finite field restriction problem with its Euclidean analogue, and we encourage the reader to do so. Indeed the Euclidean restriction problem has a rich history, but we leave it to the interested reader to pursue the Euclidean restriction problem on their own time (see, for example, [102, 149]).
4.4 Exercises: Chapter 4 Exercise 4.1. Prove the four parts of Proposition 4.3.2. Exercise 4.2. Verify equations (4.3) and (4.4). Exercise 4.3. Find a formula for ð(0) when ð â¡ 3 (mod 4). Exercise 4.4. Prove Proposition 4.2.1 for general dimensions ð. Exercise 4.5. Prove that â |ð ð¡Ì (ð)|2 = ðâð â ðâðâ1 + ð¿(ð)ðâ1 , ð¡âðœð
where ð¿(â
) is the ð-dimensional delta function. Exercise 4.6. Let ð and ð be conjugate exponents with 1 < ð, ð < â. 1 1 That is, for 1 < ð, ð < â, assume that ð + ð = 1. Show that |Î(ðž)| â¥
|ðž|2ð ð/ð
(âð¡ ð(ð¡)ð )
.
74
Chapter 4. Wolffâs exponent
Exercise 4.7. Prove the following generalization of Theorem 4.1.2. For even dimensions ð ⥠2 show that: 2 2
â1
4
â ð(ð¡) = ð |ðž| + ð ð¡
3ð
2
Ë || ) â ððâ1 |ðž|2 . â ( â ||ðž(ð) ð¡
âðâ=ð¡
Exercise 4.8. Find an explicit value for âð¡ ð(ð¡)2 for odd dimensions ð ⥠3.
10.1090/car/037/05
Chapter
5
Rings and generalized distances Next we turn our attention to studying generalizations of the distance problem. We will examine some methods by which to generalize the problem: we can either vary the ambient space of the underlying problem (which brought finite field models to light in the first place), or we can generalize our notion of a distance. We could also think of distance set problems in the more general category of finite point configurations, though we take up this topic in Chapter 6.
5.1 Distances in finite rings The first generalization we will discuss involves moving from the finite field ðœð to the cyclic ring â€ð â â€/ðâ€. We use the notation â€ð to denote this ring for maximal clarity. For a finite field ðœð to be defined, ð must be a power of a prime. In the case of the ring of integers modulo ð, â€ð is meaningfully defined for all integers ð ⥠2. Recall that â€Ãð is the set of units in â€ð , so that â€Ãð = {ð¥ â †ⶠ1 †ð¥ †ð and gcd(ð¥, ð) = 1}1 . In this setting of integers mod ð, the distance problem can be worded similarly to the finite field case. First, â€ðð = â€ð Ãâ¯Ãâ€ð is the free module 1 Again we are conflating integers with the set of equivalence classes mod ð. Technically, the set of units modulo ð refers to the set of equivalence classes defined by the equivalence relation which we used to define the integers mod ð, such that the representatives of those equivalence classes are relatively prime to ð. To enumerate this technically correct definition is much too burdensome to be helpful, hence why we will continue to conflate the set of integers and their equivalence classes.
75
76
Chapter 5. Rings and generalized distances
of rank ð over the set of integers modulo ð. For ð¥ = (ð¥1 , . . . , ð¥ð ) â â€ðð , we define âð¥â = ð¥12 + ⯠+ ð¥ð2 â â€ð , and informally we refer to âð¥â as the âlengthâ or ânormâ of ð¥. Again, the function ð(ð¥) = âð¥â is not a norm in any (analytic) sense, though it behaves like a norm in an algebraic sense. Indeed, âðð¥â = âð¥â for any ð¥ â â€ðð and ð â ð ð (â€ð ). Here ð ð (â€ð ) = {ð â ððð¡ ð (â€ð ) ⶠðð ð = ðŒ} is the set of ð à ð orthogonal matrices with entries in â€ð . The distance set of ðž â â€ðð is defined analogously: Î(ðž) = {âð¥ â ðŠâ ⶠð¥, ðŠ â ðž}. The ErdÅs-Falconer distance problem in â€ðð is as follows. Problem 5.1.1. Let ðž â â€ðð . Find the minimal value of ðŒ such that there exist positive constants ð¶, ð (independent of ð) so that if |ðž| ⥠ð¶ððŒ then |Î(ðž)| ⥠ðð. The first such result was achieved in [26]. Theorem 5.1.2. Suppose that ð = ðâ is a power of an odd prime. If ðž â â€ðð has cardinality |ðž| â« â(â + 1)ð
(2ââ1)ð+1 2â
, then Î(ðž) â â€Ãð ⪠{0}.
Remark 5.1.3. Note that when â = 1, â€ð = ðœð is a field and the results in Theorem 5.1.2 match those of Theorem 2.3.3. Also, note |â€Ãð | = 2 ðâ â ðââ1 . Thus, we have shown that |Î(ðž)| ⥠3 ð for sets ðž â â€ð of sufficiently large cardinality. Other than Theorem 5.1.2, only one other such result is known ([24]). Theorem 5.1.4. Suppose that ð ⥠3, and ð has prime factorization ð = ðŒ ðŒ ð1 1 . . . ðð ð , where 2 < ð1 < ⯠< ð ð . Then if ðž â â€ðð has cardinality â
|ðž| â« ð(ð)ðð ð1 divisors of ð.
ðâ2 2
, then Î(ðž) = â€ð . Here, ð(ð) is the number of positive
Corollary 5.1.5. Let ð ⥠3, and assume ð = ðâ is the power of an odd prime. If ðž â â€ðð has cardinality |ðž| â« âð
(2ââ1)ð+2 2â
, then Î(ðž) = â€ð .
5.1. Distances in finite rings
77
The proof of Theorem 5.1.2 will mimic the proof of Theorem 2.3.3, though we will have more arithmetic difficulties with which to contend such as the presence of zero divisors. In turn, the proof of Theorem 5.1.4 will follow the outline to Theorem 5.1.2, though much more care is needed in evaluating the corresponding exponential sums.
5.1.1 Fourier analysis over â€ðð Just as in the finite field setting, for a function ð ⶠâ€ðð â â, we define the Fourier transform as Ë ð(ð) = ðâð â ð(âð¥ â
ð)ð(ð¥). ð¥ââ€ð ð
Here, ð(ð¥) = ð2ððð¥/ð is the canonical additive character in ðœð . Note that ð is well defined since ð is periodic with period ð, and hence evaluating ð at any two elements in some equivalence class will yield the same output. Just as was the case in finite fields, we have orthogonality, Plancherel and inversion identities: Proposition 5.1.6. We have ðâð â ð(ð¥ â
ð) = { ð¥ââ€ð ð
1 0
ð â (0, . . . , 0) ð = (0, . . . , 0).
Moreover, let ð and ð be functions mapping â€ðð to the complex numbers â. Then, Ë ð(ð) â ð(ð) Ì = ðâð â ð(ð¥)ð(ð¥) ðââ€ð ð
ð¥ââ€ð ð
Ë ð(ð¥) = â ð(ð)ð(ð¥ â
ð). ð¥ââ€ð ð
Lemma 5.1.7. Let ð ð¡ = {ð¥ â â€ðð ⶠâð¥â = ð¡} denote the sphere of radius ð¡ in â€ðð . Suppose that ð is odd. If ð¡ â â€Ãð is a unit and ð = 2, or if ð ⥠3 and ð¡ â â€ð is arbitrary, then we have |ð ð¡ | = ððâ1 (1 + ð(1)). For all dimensions ð ⥠2 and all radii ð¡ â â€ð , we have |ð ð¡ | ⪠ððâ1 .
78
Chapter 5. Rings and generalized distances
Finally, we will need the following estimates on the Fourier transform of the characteristics function of the sphere: Lemma 5.1.8. Let ð ð¡ denote the indicator function of the sphere of radius ð¡. First, suppose that ð = ðâ is a power of an odd prime. Then, for ð ⥠2, ð¡ â â€Ãð , and ð â (0, . . . , 0) we have |ð ð¡Ì (ð)| †â(â + 1)ðâ
ð+2ââ1 2â
ðŒ
. ðŒ
If ð > 2, and ð has the prime factorization ð = ð1 1 . . . ðð ð where 2 < ð1 < ⯠< ð ð , then for ð â (0, . . . , 0), we have |ð ð¡Ì (ð)| â€
ð(ð) â ðâ2 ð 2 ð 1
for all ð¡ â â€ð . We are now primed to prove Theorems 5.1.2 and 5.1.4. As usual for ðž â â€ðð we define ð(ð¡) = |{(ð¥, ðŠ) â ðž à ðž ⶠâð¥ â ðŠ| = ð¡}|. For each unit ð¡ â â€Ãð , our goal will be to show that ð(ð¡) > 0 which implies the existence of a pair of elements ð¥ and ðŠ in ðž such that âð¥ â ðŠâ = ð¡, and thus ð¡ â Î(ðž). First, suppose that ð = ðâ is the power of an odd prime. We again use the set to denote its own characteristic (or indicator) function. For example: 1
ð¥âðž
0
ð¥ â ðž.
ðž(ð¥) = { Thus, we can write ð(ð¡) as: ð(ð¡) = â ðž(ð¥)ðž(ðŠ)ð ð¡ (ð¥ â ðŠ) ð¥,ðŠ
= â ðž(ð¥)ðž(ðŠ)ðËð¡ (ð)ð((ð¥ â ðŠ) â
ð) ð¥,ðŠ,ð
= â (â ðž(ðŠ)ð(âðŠ â
ð)) (â ðž(ð¥)ð(âð¥ â
ð))ðËð¡ (ð). ð
ðŠ
ð¥
5.1. Distances in finite rings
79
Applying the definition of the Fourier transform, this becomes
Ë Ë ðËð¡ (ð) â
ðð ðž(ð) ð(ð¡) = â ðð ðž(ð) ð 2
Ë || ð ð¡Ì (ð) = ð2ð â ||ðž(ð) ð
Just as before, if we separate the zero vector term ð = (0, . . . , 0) from the rest of the sum, we see that
ð(ð¡) = ðâð |ðž|2 |ð ð¡ | + ð2ð
2
Ë || ð ð¡Ì (ð). ||ðž(ð)
â ðâ (0,. . .,0)
â ð + ð
ð¡ . By Lemma 5.1.7, since ð¡ â â€Ãð , we have |ð ð¡ | = ððâ1 (1 + ð(1)). Hence the main term satisfies ð = ðâð |ðž|2 |ð ð¡ | = ðâ1 |ðž|2 (1 + ð(1)). By Lemma 5.1.8, since ð â (0, . . . , 0) and again using the fact that ð¡ â â€Ãð , we see that |ð ð¡Ì (ð)| †â(â + 1)ðâ
ð+2ââ1 2â
.
Thus, the modulus of the remainder term satisfies |ð
ð¡ | †ð2ð
max
ðâ (0,. . .,0)
|ð ð¡Ì (ð)|
†ð2ð â
â(â + 1)ð
â
2
Ë || ||ðž(ð)
ðâ (0,. . .,0) ð+2ââ1 â 2â
2
Ë || â ||ðž(ð) ðâðœð ð
80
Chapter 5. Rings and generalized distances
where we put back in the term ð = (0, . . . , 0). We can finish our estimate by applying Plancherel: |ð
ð¡ | †ð2ð â
â(â + 1)ðâ
ð+2ââ1 2â
ðâð â ðž(ð¥)2 ð¥ââ€ð ð
= ð2ð â
â(â + 1)ðâ = â(â + 1)ðâ
ð+2ââ1 2â
ð+2ââ1 2â
ðâð |ðž|
ðð |ðž|.
We have demonstrated that ð(ð¡) =
|ðž|2 (1 + ð(1)) + ð
ð¡ ð
where the Remainder term ð
ð¡ satisfies |ð
ð¡ | †â(â + 1)ðâ
ð+2ââ1 2â
ðð |ðž|,
for any ð¡ â â€Ãð . It follows that if ð+2ââ1 |ðž|2 (1 + ð(1)) > â(â + 1)ðâ 2â ðð |ðž|, ð
(5.1) (2ââ1)ð+1
then ð(ð¡) > 0, and this inequality (5.1) holds if |ðž| ⥠ð¶â(â + 1)ð 2â for a sufficiently large constant ð¶. This establishes Theorem 5.1.2. ðŒ ðŒ Now, assume that ð has prime decomposition ð = ð1 1 . . . ðð ð for ðâ1 primes 2 < ð1 < ⯠< ð ð . We again have |ð ð¡ | = ð (1 + ð(1)), but the decay estimate is changed. For ð â (0, . . . , 0), we have the bound â
ðâ2
|ð ð¡Ì (ð)| †ðâ1 ð(ð)ð1 2 . Following the same outline as the proof of the previous theorem, write ð(ð¡) = ðâ1 |ðž|2 (1 + ð(1)) + ð
ð¡ , â
where |ð
ð¡ | †ð(ð)ððâ1 |ðž|ð1
ðâ2 2
. Hence, ð(ð¡) > 0 whenever â
|ðž| ⥠ð¶ð(ð)ðð ð1
ðâ2 2
for a sufficiently large constant ð¶, which establishes Theorem 5.1.4. It remains to prove the Lemmas about the sphere ð ð¡ .
5.1. Distances in finite rings
81
5.1.2 Gauss sums over â€ð Before we prove Lemmas 5.1.7 and 5.1.8, we first provide background in character theory over â€ð as well as some well-known results on quadratic Gauss sums over â€ð . First we consider the additive group â€ð . The canonical (additive) character on this group is the function ð(ð¥) = ð2ððð¥/ð . This function is periodic with period ð, so we can view this function as taking inputs from â€, though it is also well-defined on â€ð . All other characters on â€ð are of the form ðð (ð¥) â ð(ðð¥) = ð2ðððð¥/ð for ð â â€ð . Thus if ð = 0, then we achieve the trivial character: ð0 (ð¥) â¡ 1. Proposition 5.1.6 yields an orthogonality relation for the canonical additive character: ðâ1 â ð(ðð¥) = { ð¥ââ€ð
1 0
ðâ¡0 ðâ¢0
(mod ð) (mod ð)
It is important to note that all characters on â€ðð are of the form ð(ðŒ â
ð¥), where ð(ð¥) = ð2ððð¥/ð , and where ðŒ, ð¥ â â€ðð with ðŒ â
.ð¥ denoting the usual dot product: If ðŒ = (ðŒ1 , . . . , ðŒð ) â â€ðð and ð¥ = (ð¥1 , . . . , ð¥ð ) â â€ðð , then ðŒ â
ð¥ = ðŒ1 ð¥1 + ⯠+ ðŒð ð¥ð . Next, we consider the multiplicative subgroup â€Ãð . Recall that ð¥ â â€ð is a unit if and only if gcd(ð¥, ð) = 1. The multiplicative characters on â€ð (i.e., the characters on the multiplicative subgroup â€Ãð ) are functions ð ⶠâ€Ãð â â such that ð is not identically 0, ð is totally multiplicative (meaning that ð(ð¥ðŠ) = ð(ð¥)ð(ðŠ)), and ð(ð¥ + ð) = ð(ð¥) for all ð¥, so that ð is periodic with period ð. We can extend these functions to be defined on all of â€ð (and thus on all of â€) by adding the condition that ð(ð¥) = 0 if gcd(ð¥, ð) â 1. These characters are called Dirichlet characters mod ð. While there is much to say and learn about Dirichlet characters, we focus only on the Jacobi symbol, which is the one and only Dirichlet character that I know that is relevant to our study of distance sets2 . Recall that we 2 I cannot say for sure whether other Dirichlet characters are irrelevant to the study of distance sets since we cannot fully solve the problem!
82
Chapter 5. Rings and generalized distances
have already seen the Legendre symbol modulo an odd prime ð: 1 ð¥ ( )={ 0 ð â1
ð¥ is congruent to a nonzero square mod ð ð¥ â¡ 0 (mod ð) ð¥ is not congruent to a nonzero square mod ð.
The Jacobi symbol is defined as a product of Legendre symbols, and it is defined only for odd integers ð ⥠1. Definition 5.1.9. Let ð ⥠3 be an odd integer with prime factorization ð ð ð = ð11 . . . ððð . Then the Jacobi symbol is the function ð
ð
ð¥ 1 ð¥ ð¥ ð ( ) = ( ) ... ( ) . ð ð1 ðð ð¥
The Jacobi symbol also follows the convention that ( 1 ) = 1 for all integers ð¥. We leave it to the reader to verify that this truly defines a Dirichlet character. We are now ready to discuss quadratic Gauss sums over the integers modulo ð. Definition 5.1.10 (Quadratic Gauss sums). For a fixed integer ð ⥠2 and for integers ð and ð, we denote by ðº(ð, ð, ð) the following sum: ðº(ð, ð, ð) â â ð(ðð¥2 + ðð¥) ð¥ââ€ð
where ð(ð¥) = ð2ððð¥/ð . For convenience, we set ðº(ð, 0, ð) = ðº(ð, ð). Since ð is fixed, the quantity ðº(ð, ð, ð) can be viewed as a function in the variables ð and ð, but keeping track of the value ð will turn out to be important for us, so we include this value in the definition above. Note also that ðº(ð, ð, ð) = ðº(ðâ² , ðâ² , ð) if ð â¡ ðâ² (mod ð) and ð â¡ ðâ² (mod ð). This is because we can view ð as a function over â€ð , as well as a function over â€. This slight abuse of notation should not cause any confusion. For the same reason, we may think of ðº(ð, ð, ð) as taking values from â€ð or â€, since only the equivalence class of the integers ð and ð is relevant for this sum. Proposition 5.1.11 ([90]). Fix an odd positive integer ð ⥠3, and let ðº(ð, ð, ð) denote the Gauss sum above, keeping in mind that we write
5.1. Distances in finite rings
83
ðº(ð, ð) = ðº(ð, 0, ð). For any positive integer ð with gcd(ð, ð) = 1, we have ð ðº(ð, ð) = ðð ( ) âð ð â
where ( ð ) denotes the Jacobi symbol mod ð, and ðð = {
1 ð
ð â¡ 1 (mod 4) ð â¡ 3 (mod 4).
Furthermore, for general values of ð â â€ð , we have ð
ðº(ð, ð, ð) = {
ð
ð
gcd(ð, ð) â
ðº ( gcd(ð,ð) , gcd(ð,ð) , gcd(ð,ð) ) 0
gcd(ð, ð) ⣠ð ðð¡âððð€ðð ð.
Definition 5.1.12 (Generalized Gauss Sum). For a fixed odd integer ð ⥠3, we let ð(ð¥) denote a Dirichlet character modulo ð, and we let ðð (ð¥) = ð2ðððð¥/ð be an additive character on â€ð . Then, we set ð(ð, ðð ) = â ð(ð¥)ðð (ð¥). ð¥ââ€ð
When ð = 1, it is customary to write ð(ð, ð1 ) = ð(ð). Proposition 5.1.13 ([90]). If ð(ð¥) denotes the Jacobi symbol mod ð, then we have the bound |ð(ð, ðð )| †âð. ð¥
In particular, if ð(ð¥) = ( ð ) is the Legendre symbol mod ð, and if ðð (ð¥) = ð2ðððð¥/ð , then we have the same bound: |ð(ð, ðð )| †âð.
5.1.3 Proof of Lemma 5.1.7 We are ready to prove the estimates on the cardinality of the sphere as well as the bound on the Fourier transforms of the characteristic function of the sphere. We first prove the Lemma in the case that ð = ðâ is a power of an odd prime. As before, put ð(ð¥) = ð2ððð¥/ð , and suppose ð is a fixed unit in â€ð .
84
Chapter 5. Rings and generalized distances
Then by orthogonality (see Proposition 5.1.6), we have |ðð | = â ðð (ð¥) = ðâ1 â â ð(ð (âð¥â â ð)) ð ââ€ð ð¥ââ€ð ð
ð¥ââ€ð ð
= ðâ1 â â ð (ð (ð¥12 + ⯠+ ð¥ð2 â ð)) ð ââ€ð ð¥ââ€ð ð
= ðâ1 â â ð (ð ð¥12 ) . . . ð (ð ð¥ð2 ) ð(âð ð) ð ââ€ð ð¥ââ€ð ð
= ðâ1 (ðâ + ð0 + ⯠+ ðââ1 ) , where â
ðð =
â ð(ð ð¥12 ) . . . ð(ð ð¥ð2 )ð(âð ð)
0â€ð â€ðâ1 ð¥ââ€ð ð ð£ððð (ð )=ð ð
â
=
0â€ð â€ðâ1 ð£ððð (ð )=ð
â
=
2
( â ð(ð ð¥ )) ð(âð ð) ð¥ââ€ð
(ðº(ð , ð))ð ð(âð ð).
0â€ð â€ðâ1 ð£ððð (ð )=ð
Recall that for an integer ð¥, we write ð£ððð (ð¥) = ð if ðð is the largest power of the prime ð that divides ð¥, and ð£ððð (0) = â. The term ðâ refers to the sum when ð = 0, so that ðâ = ðð = ðâð . For 0 †ð †â â 1, note that if 0 †ð †ð â 1 and ð£ððð (ð ) = ð, then ð can be written in the form ð = ðð ð â² , where gcd(ð â² , ð) = 1 and 0 †ð Ⲡ†ðââð is uniquely determined3 . Using this fact, along with Proposition 5.1.11, we see that ðð = ððð
â (ðº(ð , ðââð ))ð ð(âð ð) ð ââ€Ã ðââð ð
= ððð ðððââð (ðââð ) 2
â ð (ð )
ð(ââð)
ð(âð ð)
ð ââ€Ã ðââð
ð
where ð(ð ) = ( ð ) is the Legendre symbol mod ð. 3 Note that the condition that ð is finite precludes the possibility of having ð = 0, but the statement is still true including 0 in the range of ð .
5.1. Distances in finite rings
85
If ð(â â ð) is even, we see that ð
ð
ðð = ðâ 2 +ð 2 ðððââð
ð
â ð(âð ð) ð ââ€Ã ðââð
ð
= ðâ 2 +ð 2 ðððââð ( â ð(âð ð) â ð ââ€ðââð ð
ð
= âðâ 2 +ð 2 ðððââð
â
â
ð(âð ð))
ð âðâ€ðââðâ1
ð(âð ð),
ð âðâ€ðââðâ1 ð
so that |ðð | †ð(â+ð) 2 (ðââðâ1 ) = ð using the notation
â(
ð+2 ðâ2 )+ð( 2 )â1 2
. Note that we are
ðâ€ððŒ = {ðð¥ ⶠð¥ â †and 0 †ð¥ †ððŒ â 1} . We are tacitly identifying the ring â€ð with the set of integers {0, . . . , ðâ1} which represent its equivalence classes. In doing so, we can think of ðâ€ððŒ as the set of non-units in the ring â€ððŒ+1 . In other words, we identify ðâ€ððŒ = â€ððŒ+1 ⧵ â€ðÃðŒ+1 , which is equivalent to viewing â€ððŒ+1 = â€ðÃðŒ+1 ⪠ðâ€ððŒ . Now if ð(â â ð) is odd, then, ð
ðð = ð(â+ð) 2 ðððââð
â ð(ð )ð(âð ð) ð ââ€Ã ðââð
ð
= ð(â+ð) 2 ðððââð ( â ð(ð )ð(âð ð) â ð ââ€ðââð
â
ð(ð )ð(âð ð)) .
ð âðâ€ðââðâ1
ââµâµâµâµâµââµâµâµâµâµâ
ââµâµâµâµâµâµââµâµâµâµâµâµâ
ð(ð,ðâð )
ð
Proposition 5.1.13 gives |ð(ð, ðâð )| †âðââð , while the triangle inequality yields the crude bound | | â ð(ð )ð(âð ð)|| †|ð(ð )ð(âð ð)| †ðââðâ1 . |ð
| = || â ð âð†ð âð†| | ðââðâ1 ðââðâ1
86
Chapter 5. Rings and generalized distances
When ð(â â ð) is odd, it follows that ð
1
|ðð | †ð(â+ð) 2 (ð(ââð) 2 + ðââðâ1 ) . ââð 2 ð+2 ðâ2 â 2 +ð 2 â1
Noting that
†â â ð â 1 for ð †â â 2, we have shown that |ðð | â€
2ð when 0 †ð †â â 2, and |ðââ1 | †2ðâðâ our estimates show: ð
ðâðâ 2
|ðð | †|ðââ1 | †{
âðâ
2ð
ðâ1 2
. Altogether,
ð(â â 1) is even ðâ1 2
ð(â â 1) is odd.
Thus, we have |ðð | = ððâ1 + ðâ1 (ð0 + ⯠+ ðââ1 ), where ð
ââ1
|ð0 + ⯠+ ðââ1 | †â |ðð | †{ ð=0
âðâðâ 2 2âð
âðâ
ð(â â 1) is even ðâ1 2
ð(â â 1) is odd.
(5.2)
Putting everything together, and recalling that we set ð = ðâ , we have ââ1
|ðð | = ðâ(ðâ1) + ð (ðâ1 â |ðð |)
(5.3)
ð=0 ð
=ð
ðâ1
+ ð ({
âðâ(ðâ1)â 2 âð
â(ðâ1)â
ð(â â 1) is even
ðâ1 2
ð(â â 1) is odd
})
= ðâ(ðâ1) (1 + ð(1)).
(5.4)
The general case follows from the Chinese Remainder Theorem. Recall â â that if ð = ð11 . . . ððð , then â€ð â
â€ðâ1 à ⯠à †1
â
ððð
and
â€Ãð â
â€Ãâ1 à ⯠à â€Ãâð . ð1
ðð
(5.5)
See Exercise 5.6 for more details. Write ð¡ â â€Ãð as ð¡ = (ð¡1 , . . . , ð¡ ð ), where ð¡ð â â€Ãâð . To find the number of solutions to the equation âð¥â = ð¡ in ðð â€ð , one must solve the equation âð¥â = ð¡ð in each component â€ðâð . The ð number of solutions in â€ð is then the product of the number of solutions in â€ðâð . Hence: ð
ð
|ð ð¡ | = â |ð ð¡ð | = ððâ1 (1 + ð(1)), ð=1
which completes the proof of Lemma 5.1.7.
5.1. Distances in finite rings
87
5.1.4 Fourier decay of the sphere Next we prove Lemma 5.1.8. We first require ð = ðâ to be a power of an odd prime. For ð â â€ðð ⧵ {(0, . . . , 0)}, by applying the definition of the Fourier transform, and using orthogonality, we see that ððÌ (ð) = ðâð â ðð (ð¥)ð(âð¥ â
ð) ð¥ââ€ð ð
= ðâðâ1 â â ð((ð¥12 + ⯠+ ð¥ð2 â ð)ð¡)ð(âð â
ð¥) ð¡ââ€ð ð¥ââ€ð ð
= ðâðâ1 â
â
ð¥ââ€ð ð
ð((ð¥12 + ⯠+ ð¥ð2 â ð)ð¡)ð(âð â
ð¥)
ð¡ââ€ð ⧵{0}
since the term corresponding to ð¡ = 0 is 0 by Proposition 5.1.6 as ð â (0, . . . , 0). Thus, ð
ððÌ (ð) = ðâðâ1 â ð(âðð¡) â ( â ð(ð¥ð2 ð¡ â ðð ð¥ð )) ð¡â 0
ð=1
ð¥ð ââ€ð
ð
= ðâðâ1 â ð(âðð¡) â ðº(ð¡, âðð , ð) ð¡â 0
(5.6)
ð=1
where we regrouped our terms and recognized the last sums as quadratic Gauss sums. By Proposition 5.1.11, ðº(ð¡, âðð , ð) = 0, unless ðð â¡ 0 (mod gcd(ð¡, ð)). Note also that ð£ððð (ð¡) = ð implies that gcd(ð¡, ð) = ðð . Now, we partition ââ1
ððÌ (ð) = â ðððÌ (ð), ð=0
where applying the last equality in Proposition 5.1.11 yields ðððÌ (ð) â ððÌ (ð)|ð£ðð
ð (ð¡)=ð
ð
= ðâðâ1
â ð£ððð (ð¡)=ð
ð(âðð¡)ððð â ðº ( ð=1
ð¡ âðð ââð , ,ð ). ðð ðð
For convenience, we put ð¢ = ð¡/ðð and ðð = ðð /ðð . We note that ð¢ is a unit and ðð are integers, since ðð â¡ 0 (mod ðð ) (as otherwise the quadratic Gauss sum vanishes). Again, we write ð = (ð1 , . . . , ðð ) and
88
Chapter 5. Rings and generalized distances ð
âðâ = âð=1
ð2ð , ð2ð
ðððÌ (ð) = ðâðâ1
â
and ( ðð ) denotes the Jacobi symbol. Hence, ð
ððð â
ð 2 (ââð) ðððââð â
ð (âðð¢ â
â ð£ððð (ð¡)=ð ââ1
ð
= ðâðâ1 â ð 2 (â+ð) ðððââð ð=0
â
ð (â
áµââ€Ã ðââð
ð âðâ ð¢ ) â
( ââð ) 4ð¢ ð
(ââð)ð âðâ ð¢ â ðð¢) ( ) . 4ð¢ ð
To finish the argument, we claim that we have the bound | ðŸð | ðŸ | | ð¢ | â ð (ðð¢â1 + ðð¢) ( ) | †(ðŸ + 1)ð 2 , ð | |áµââ€Ã | ððŸ |
(5.7)
where ð â â€ððŸ is arbitrary, ð â â€ððŸ is a unit, and ðŸ is a positive integer. Accepting the claim for the moment, we see that ââ1
ð
|ððÌ (ð)| †ðâðâ1 â (â â ð + 1)ð 2 (â+ð) ð
ââð 2
ð=0
†(â + 1)ðâðâ1 ð
(ð+1)â 2
†(â + 1)ðâðâ1 ð
(ð+1)â 2â
ââ1 (
âð
ðâ1 )ð 2
ð=0
†â(â + 1)ðâ
1
âð â
((
ðâ1 )(ââ1)) 2
ð+2ââ1 2â
from which the result follows. It remains to justify (5.7). For convenience, we define the Salié sum (or twisted Kloosterman sum) as ð¥ ð(ð, ð, ð) = â ð (ðð¥â1 + ðð¥) ( ) , ð ð¥ââ€Ã ð
and we define the Kloosterman sum as ðŸ(ð, ð, ð) = â ð (ðð¥â1 + ðð¥) . ð¥ââ€Ã ð
H. Salié ([132]) gave the bound |ð(ð, ð, ð)| †2âð for gcd(ð, ð, ð) = 1 and ð an odd prime. Note that when ð = ððŸ and ðŸ is even, we have ð(ð, ð, ð) = ðŸ(ð, ð, ð). A. Weil ([154]) provided the well known bound |ðŸ(ð, ð, ð)| †ð(ð) gcd(ð, ð, ð)1/2 ð1/2 , where ð(ð) denotes the number of
5.1. Distances in finite rings
89
positive divisors of ð. If ðŸð is even, then the sum (5.7) is reduced to the Kloosterman sum ðŸ(ð, ð, ð) which has size |ðŸ(ð, ð, ð)| †(ðŸ + 1)ððŸ/2 , since gcd(ð, ð, ð) = 1 because ð is a unit (mod ð). Finally, when ðŸ and ð are odd we invoke the bound | | | â ð (ðð¥â1 + ðð¥) ( ð¥ )| †(ðŸ + 1)ððŸ/2 | ð | |ð¥ââ€ððŸ |
where ðŸ is odd ([134]). Next we turn our attention to the case when ð has prime factorization ðŒ ðŒ ð = ð1 1 . . . ðð ð . We start with the equation (5.6):
ð
ððÌ (ð) = ðâðâ1 â ð(âðð¡) â ðº(ð¡, âðð , ð)ð . ð¡â 0
ðœ
ð=1
ðœ
Our first step is to write ð¡ = ð1 1 . . . ððð ð¢ where ð¢ â â€Ãðâ² is uniquely deðŒ âðœ
ðŒ âðœ
termined for modulus ðâ² = ð1 1 1 . . . ðð ð ð and ðœð ⥠0. Note that ðœð < ðŒð for some ð since ð¡ â 0. We will use the notation âðœ to denote the sum over all such (ðœ1 , . . . , ðœ ð ). For ð = (ð1 , . . . , ðð ) and ðœ = (ðœ1 , . . . , ðœ ð ), we ðœ ðœ define ðð,ðœ to be 1 if ð1 1 . . . ððð ⣠ðð for all ð, and zero otherwise. When ðð ðð,ðœ = 1, we put ðð = ðœ1 ðœð . For notational convenience, we will write ððœ =
ðœ ð1 1
ðœ . . . ððð .
ð1 . . .ðð
Applying Proposition 5.1.11,
ð
ð ð¡Ì (ð) = ðâðâ1 â â ð(âððœ ð¢ð¡) â ðº (ððœ ð¢, âðð , ð) ðœ áµââ€Ãâ² ð
ð=1 ð
= ðð,ðœ ðâðâ1 â â ððœð ð(âððœ ð¢ð¡) â ðº(ð¢, âðð , ðâ² ). ðœ áµââ€Ãâ² ð
ð=1
90
Chapter 5. Rings and generalized distances
Noting that by completing the square we have ðº(ð¢, âðð , ðâ² ) = â ð (ð¢ð¥2 â ðð ð¥) ð¥ââ€ðâ²
= â ð (ð¢ (ð¥ â ð¥ââ€ðâ²
ðð 2 âð2 ) )ð( ð) 2ð¢ 4ð¢
= â ð (ð¢ð¥2 ) ð ( ð¥ââ€ðâ²
âð2ð ) 4ð¢
where we used a change of variables in the last line. Taking the product of these terms and applying Proposition 5.1.11 yields ð
ð ð¡Ì (ð) = ðð,ðœ ðâðâ1 â â ððœð ð(âððœ ð¢ð¡) â ðº(ð¢, âðð , ðâ² ) ðœ áµââ€Ãâ² ð
ð=1 ð
= ðð,ðœ ðâðâ1 â ððœð (ðâ² ) 2 ðððâ² â ð (âððœ ð¢ð¡ â áµââ€Ã ðâ²
ðœ
âðâ ð¢ ð )( â²) . 4ð¢ ð
We will apply the trivial bound to the sum in ð¢ â â€Ãðâ² : | | | âðâ ð¢ ð | ðœ ðŒ âðœ ðœ ðŒ âðœ | â ð (âð1 1 . . . ððð ð¢ð¡ â ) ( â² ) | †ðâ² = ð1 1 1 . . . ðð ð ð |áµââ€Ã 4ð¢ ð | | | ðâ² to see that (
|ð ð¡Ì (ð)| †ðâðâ1 â ð1
ðŒ1 +ðœ1 2
)ð
(
. . . ðð
ðŒð +ðœð 2
)ð ðŒ âðœ ð1 1 1
ðœ ð
(
= ðâðâ1 â â ðð ðœ ð=1
ðŒð +ðœð 2
)ð+ðŒð âðœð
.
ðŒ âðœð
. . . ðð ð
5.2. Distances between two sets
91
Writing ðœð = ðŒð â ðð , we have ð
(
|ð ð¡Ì (ð)| †ðâðâ1 â â ðð
2ðŒð âðð 2
)ð+ðð
ðð 2
)ð+ðð
ðœ ð=1 ð
(ðŒð â
= ðâðâ1 â â ðð ðœ ð=1 ð
â
= ðâðâ1 â ðð â ðð ðœ
ððð â2 2
ð=1 ðâ2 â 2
†ðâ1 ð(ð)ðð
for some ðð since ðð > 0 for at least one value ð as ð¡ â 0. The largest value obtained by the quantity occurs when ð1 = 1 and ðð = 0 for ð > 1. Hence, â
|ð ð¡Ì (ð)| †ðâ1 ð(ð)ð1
ðâ2 2
.
This completes the proof of Proposition 5.1.8.
5.2 Distances between two sets Our next generalization consists of examining distances determined by pairs of sets. We put Î(ðž, ð¹) = {âð¥ â ðŠâ ⶠð¥ â ðž, ðŠ â ð¹}, where ðž, ð¹ â ðœðð or ðž, ð¹ â â€ðð . The techniques used in Section 3.1 show that Î(ðž, ð¹) â ðœâð whenever |ðž||ð¹| ⥠ð¶ðð+1 for a sufficiently large constant ð¶ (see Exercise 5.9). Note that 0 â Î(ðž, ð¹) if ðž â© ð¹ â â
, though this need not be guaranteed by the size condition on |ðž||ð¹|. Notice that the exponent ð + 1 cannot be improved unconditionally because of the previously discussed counterexamples in the finite field setting. However, some progress has been made on the problem. The first result in this direction was given by Koh and Shen ([99]). Theorem 5.2.1. Suppose that ðž, ð¹ â ðœ2ð are such that |ðž||ð¹| â« ð8/3 . Then, there exists a constant 0 < ð †1 such that |Î(ðž, ð¹)| ⥠ðð.
92
Chapter 5. Rings and generalized distances
We will delve into some of these ideas later in the section, though we leave it to the interested reader to fill in any missing details. This is the only result (out of Theorems 5.2.1, 5.2.2, and 5.2.3) that allows ðž and ð¹ to be arbitrary sets, and hence it is the only result that is a true generalization of the ErdÅs-Falconer distance problem for arbitrary sets. Up to logarithms, the problem was settled by Dietmann [36] in all dimensions, albeit for sets ðž and ð¹ of different sizes. Because of this restriction, these results will not push forward any progress on the finite field distance problem. Theorem 5.2.2. Suppose that ðž, ð¹ â ðœðð are such that |ðž||ð¹| â« ðð log ð where we have |ð¹| â« ð such that
ð+1 2
log ð. Then there exists a constant 0 < ð †1 |Î(ðž, ð¹)| ⥠ðð.
The logarithmic factors were removed in [100], thus completely settling the conjecture in the general case: Theorem 5.2.3. Suppose that ðž, ð¹ â ðœðð such that |ðž||ð¹| â« ðð where we have |ð¹| â« ð
ð+1 2
. Then there exists a constant 0 < ð †1 such that |Î(ðž, ð¹)| ⥠ðð.
Proof. We will sketch the proof of Theorem 5.2.3 as we have already developed the necessary machinery to do so. We need to recall a few facts from earlier chapters. First and foremost we will be using ð(ð¡) = {(ð¥, ðŠ) â ðž à ð¹ ⶠâð¥ â ðŠâ = ð¡}. There is no harm in using the same notation as in the single set case (ðž = ð¹) since no confusion should occur. Proposition 5.2.4. Let ðž, ð¹ â ðœðð . Then, |Î(ðž, ð¹)| â¥
|ðž|2 |ð¹|2 âð¡ ð(ð¡)2
(5.8)
and 2
|Î(ðž, ð¹)| â¥
(|ðž||ð¹| â ð(0)) . âð¡â 0 ð(ð¡)2
(5.9)
5.2. Distances between two sets
93
You have probably already guessed that we will use equation (5.8) when ð â¡ 3 (mod 4) and thus we need (5.9) in the case ð â¡ 1 (mod 4). Also, recall that ððž (ð¡) denotes the spherical average: 2
Ë || . ððž (ð¡) = â ||ðž(ð) âðâ=ð¡
Proposition 5.2.5. Let ðž, ð¹ â ðœðð . Then â ð(ð¡)2 †ð¡
|ðž|2 |ð¹|2 + ð2ð |ð¹| max(ððž (ð¡)) ð ð¡âðœð
(5.10)
and 2
| |ðž|2 |ð¹|2 3ð || Ë ð¹(ð) Ë || +ð2ð |ð¹| max(ððž (ð¡)). (5.11) â ð(ð¡) †+ð | â ðž(ð) ð ð¡â 0 |âðâ=0 | ð¡ 2
Proposition 5.2.6. For ð ⥠2 even and ðž, ð¹ â ðœðð with |ðž||ð¹| ⥠16ðð we have 1 2 |ðž||ð¹|, (5.12) (|ðž||ð¹| â ð(0)) ⥠36 and 2 2 2 | | 3ð | | Ë ð¹(ð) Ë | †|ðž| |ð¹| + ð(0)2 . (5.13) ð | â ðž(ð) ð | |âðâ=0 Finally we have estimates on the maximal spherical average, which we have already seen play a major role in Theorem 4.1.1. Proposition 5.2.7. Let ðž â ðœðð . If ð ⥠3 is odd then max (ððž (ð¡)) †min(ðâð |ðž|, 2ðâðâ1 |ðž| + 2ðâ
3ð+1 2
ð¡âðœð
|ðž|2 ) .
(5.14)
If ð ⥠4 is even, then we have the same bound away from the 0-sphere: max(ððž (ð¡)) †min(ðâð |ðž|, 2ðâðâ1 |ðž| + 2ðâ ð¡â 0
3ð+1 2
|ðž|2 ) .
When ð = 2 we have a superior bound (see Theorem 4.1.3): max (ððž (ð¡)) †3ðâ3 |ðž|3/2 . ð¡â 0
(5.15)
94
Chapter 5. Rings and generalized distances
Putting together Propositions 5.2.4, 5.2.5, 5.2.6, and 5.2.7, we see that if |ðž||ð¹| â« ðð and |ðž| †|ð¹|, then
|Î(ðž, ð¹)| â«
⧠min(ð, |ð¹| ðâ1 ) ⪠ð 2
ðâ¥2
⚠⪠|ðž|1/2 |ð¹| â© min(ð, ð )
ð = 2.
Theorem 5.2.3 immediately follows from this inequality.
5.3 Generalized distances Instead of using two different sets, we could simply change what we mean by a distance. For example Koh and Iosevich ([82]) studied the set Îð (ðž) â {(ð¥1 â ðŠ1 )ð + ⯠+ (ð¥ð â ðŠ ð )ð ⶠð¥, ðŠ â ðž} for ðž â ðœðð . When ð = 2, we get the familiar finite field distance set. Notice that when ð = 3, rather than utilizing the well-known Gauss sums, the so-called Kummer sums come into play. They demonstrated the following result. Theorem 5.3.1. Let ðž â ðœðð be such that |ðž| â« ð (mod 3) is prime. Then,
ð+1 2
, where ð â¡ 1
Î3 (ðž) = ðœð . If ð = 2, then under the same conditions on ðž and ð we have Îð (ðž) = ðœð for all ð ⥠2. Furthermore, if ðž â ðœðð and ð¹ â ðœðð satisfy |ðž||ð¹| â« ðð+1 , then Î3 (ðž, ð¹) â {(ð¥1 â ðŠ1 )3 + ⯠+ (ð¥ð â ðŠ ð )3 ⶠð¥ â ðž, ðŠ â ð¹} satisfies Î3 (ðž, ð¹) â ðœâð . An alternative way to generalize the distance set was undertaken in [27]. The authors studied the so-called ð-resultant set Îð (ðž) â {âð¥1â â ⯠â ð¥ðâ â ⶠð¥1â , . . . , ð¥ðâ â ðž}.
(5.16)
5.3. Generalized distances
95
Recall that for ð¥â = (ð¥1 , . . . , ð¥ð ) â ðœðð , we have âð¥â = ð¥12 + ⯠+ ð¥ð2 . The results obtained were independent of the sign ± between the vectors ð¥ðâ , so the same conclusion could apply to sets of the form {âð¥1â ± ⯠± ð¥ðâ â ⶠð¥1â , . . . , ð¥ðâ â ðž}, though we will stick with our original definition (5.16). Note that the usual finite-field distance set is then Î2 (ðž) = Î(ðž). The results are as follows. Theorem 5.3.2. Let ðž â ðœðð . Then there exists a constant ð â (0, 1] such that |Î3 (ðž)| ⥠ðð when |ðž| â« ð |Î4 (ðž)| ⥠ðð when |ðž| â« ð
ð+1 1 â 6ð+2 2
1 ð+1 â 6ð+2 2
for ð = 4 and ð = 6, and
for even ð ⥠8.
Proof. We will provide a sketch of a proof of Theorem 5.3.2, though the details get somewhat technical, so we instruct the interested reader to comb through the original source as much as possible. We let ð ð (ð¡) be our counting function: ð ð (ð¡) = |{(ð¥1â , . . . , ð¥ðâ ) â ðž ð ⶠâð¥1â â ⯠â ð¥ðâ â = ð¡}|. Then we have |Îð (ðž)| â¥
(|ðž|ð â ð ð (0))2 âð¡âðœâ ð2ð (ð¡)
(5.17)
ð
so as in the previous chapters we need a good upper bound on â ð2ð (ð¡).
ð¡âðœâð
The authors were able to obtain the estimate 2
2
â1
â ð ð (ð¡) †ð |ðž|
2ð
ð¡âðœð
+ð
ð(2ðâ1)
| ð| | . Ë â || â (ðž(ð£)) | | ðâðœð |âð£â=ð
Additionally, we use some restriction estimates. Proposition 5.3.3. If ð ⥠4 is even, then for ð > 4 â ð
Ë || ) ⪠ðâðð+ðâ1 |ðž| max ( â ||ðž(ð) ð¡â 0
âðâ=ð¡
24 , 3ð+4
we have
3ððâ3ð+4ð+2 3ð+4
.
96
Chapter 5. Rings and generalized distances
Note that this proposition is meaningful for ð = 3 if ð = 4 or ð = 6, and it is meaningful for ð ⥠4 and in even dimensions ð ⥠8. We have one further proposition before we assemble the pieces. Proposition 5.3.4. If ð ⥠2 is even, ð ⥠2, and ðž â ðœðð satisfies |ðž| ⥠3ðð/2 , then |ðž|ð |ðž|ð â ð ð (0)2 ⥠. 3 Furthermore, we have 2
ð
2ððâð
| ð| | â ð (0)2 †4ðâ1 |ðž|2ð | â (ðž(ð)) Ë ð | | | |âðâ=0
and ð
Ë || ) . â ð ð (ð¡)2 ⪠ðâ1 |ðž|2ð + ððð |ðž|ðâ1 max ( â ||ðž(ð) ðâ 0
ð¡â 0
âðâ=ð
Putting everything together and using (5.17) we have |Îð (ðž)| â«
(|ðž|ð â ð ð (0))2 ð
Ë || ) ðâ1 |ðž|2ð + ððð |ðž|ðâ1 maxðâ 0 (ââðâ=ð ||ðž(ð) |ðž|2ð
â«
ð
.
Ë || ) ðâ1 |ðž|2ð + ððð |ðž|ðâ1 maxðâ 0 (ââðâ=ð ||ðž(ð) Depending on which term in the denominator is the dominant term, and using Proposition 5.3.3, we see that â â |ðž|ð+1 |Îð (ðž)| â« min âð, â ð â ðð max Ë || ) â ||ðž(ð) ð (â ðâ 0 âðâ=ð â â â« min(ð,
|ðž|ð+1 ððâ1 |ðž|
3ððâ3ð+4ð+2 3ð+4
)
â«ð whenever |ðž| â« ð
ð+1 1 â 6ð+2 2
.
We have a weaker result in even dimensions ð ⥠8 concerning Î3 (ðž).
5.3. Generalized distances
97
Theorem 5.3.5 ([29]). Let ð ⥠8 be even, and assume ðž â ðœðð . Then we are guaranteed that |Î3 (ðž)| ⥠ðð, so long as |ðž| â« ð
ð+1 1 â 9ð+18 2
.
The above results are intriguing in that for the first time in even dimensions ð ⥠4, we are able to dip below the exponent threshold of ð+1 implying that there is not some algebraic obstruction as in the odd2 ð+1
dimensional case. This gives some hope that the exponent 2 in higher even dimensions can be improved in the ErdÅs-Falconer distance probð+1 lem. It should be noted that we are able to dip below the previous 2 threshold because, for example, one can think of the 3-resultant set Î3 (ðž) = {âð¥ â ðŠ â ð§â ⶠð¥, ðŠ, ð§ â ðž} = {âð¥ â ð€â ⶠð¥ â ðž, ð€ â ðž + ðž} = Î(ðž, ðž + ðž). Now for some sets ðž â ðœðð , it may be that ðž+ðž is sufficiently close to ðž (if ðž has a lattice structure, for example), but we were able to overcome this obstacle by directly appealing to the applications of restriction theory, rather than viewing the 3-resultant distance set as a distance set result between two different sets. There has also been progress on the ð-resultant modulus problem for subsets of algebraic varieties. For a polynomial ð â ðœð [ð¥1 , . . . , ð¥ð ], we call ð = {ð¥ â ðœðð ⶠð(ð¥) = 0} a regular variety if |ð| â ððâ1 and ð+1
Ë ||ð(ð) || ⪠ðâ 2 when ð â (0, . . . , 0). A regular variety ð â ðœ2ð will be called nondegenerate if ð(ð¥) does not contain any linear factor. Theorem 5.3.6 ([28]). Let ð â ðœðð be a regular variety, ðž â ð, and ð ⥠3. ðâ1
1
If |ðž| â« ð 2 + ðâ1 then Îð (ðž) â ðœâð when ð ⥠2 is even and Îð (ðž) = ðœð when ð ⥠3 is odd. Many of these results can also be extended to include some of the generalizations previously discussed in this chapter. For example we considered the multiset case: Theorem 5.3.7. For ðž1 , ðž2 , ðž3 â ðœðð , define Î3 (ðž1 , ðž2 , ðž3 ) = {âð¥1â + ð¥2â + ð¥3â â ⶠð¥ðâ â ðžð , for ð = 1, 2, 3}.
98
Chapter 5. Rings and generalized distances
Then, |Î3 (ðž)| â« ð whenever |ðž1 ||ðž2 ||ðž3 | â« ð ð = 6.
3(
ð+1 1 â 6ð+2 ) 2
for ð = 4 and
Similar results hold for ð ⥠8, and for the 3-resultant set when ð = 8. Further generalizations of Theorem 5.3.6 were achieved by Hieu-Pham ([79]). Theorem 5.3.8. Let ð be a non-degenerate quadratic form on ðœðð , and assume that ð â ðœðð is a regular variety, with ðž â ð, and ð ⥠3. If |ðž| â« ð
ðâ1 1 + ð+1 2
, then for ð¡ â 0, we have
|{(ð¥1 , . . . , ð¥ð ) ⶠð(ð¥1 , . . . , ð¥ð ) = ð¡}| = (1 â ð(1))
|ðž|ð . ð
5.4 Pinned distances Instead of looking at the entire set of distances, sometimes it is sufficient to look at the set of distances where one endpoint is fixed or pinned. One of the first results to this end was a continuous result by Peres and Schlag ([121]), where they prove a pinned version of Falconerâs result. Specifically they proved: Theorem 5.4.1. Let ðž â âð have dimð» (ðž) > point ðŠ â ðž such that the pinned distance set
ð+1 . 2
Then, there exists a
{|ð¥ â ðŠ| ⶠð¥ â ðž} has positive 1-dimensional Lebesgue measure. Recall that we have intuitively (and informally) defined the notions of Hausdorff dimension and Lebesgue measure in Section 2.2. Guth, Iosevich, Ou, and Wang ([69]) have improved this bound in two dimensions, thereby improving the best known bound for the âunpinnedâ Falconer distance problem in â2 as well. Theorem 5.4.2. Let ðž â â2 be compact with dimð» (ðž) > 5/4. Then, there exists a point ðŠ â ðž such that the pinned distance set {|ð¥ â ðŠ| ⶠð¥ â ðž} has positive 1-dimensional Lebesgue measure.
5.4. Pinned distances
99
Naturally, we consider the situation in finite fields. For ðŠ â ðž, define the set of pinned distances to be ÎðŠ (ðž) = {âð¥ â ðŠâ ⶠð¥ â ðž}. It is not necessary to specify that ðŠ â ðž in order for the pinned distance set ÎðŠ (ðž) to be studied, but notice that if ðŠ â ðž, then ÎðŠ (ðž) â Î(ðž), so requiring ðŠ â ðž allows us to turned pinned distance results into generalizations of classical distance problem results. We first state and prove a simple pinned distance result ([17]). Theorem 5.4.3. Let ðž â ðœðð be such that |ðž| â« ð 1 ðŠ â ðž such that |ÎðŠ (ðž)| ⥠2 ð.
ð+1 2
. Then there exists
Proof of Theorem 5.4.3. This proof will be similar to the proof discussed in âð¿2 -methodâ discussed in Section 3.2. Just like in the proof of Theorem 2.3.3, we let ððŠ (ð¡) = |{ð¥ â ðž ⶠâð¥ â ðŠâ = ð¡}|, so that ððŠ (ð¡) denotes the number of points ð¥ â ðž which are at distance ð¡ from the pinned value ðŠ. We first note that by changing the order of summation, we obtain: â â ððŠ (ð¡) = |ðž|2 . ðŠâðž ð¡âðœð ð
Applying Cauchy-Schwarz in ðŠ, and using 1âðŠ (ðž) (â
) to denote the characteristic function of ÎðŠ (ðž), we see that 2
|ðž|4 = ( â â ððŠ (ð¡)) ðŠâðž ð¡âðœð 2
= ( â â 1âðŠ (ðž) (ð¡)ððŠ (ð¡)) ðŠâðž ð¡âðœð
†â â 1âðŠ (ðž) (ð¡)2 â ððŠ (ð¡)2 ð¡âðœð ðŠâðž
ðŠâðž
= â |ÎðŠ (ðž)| â
â â ððŠ (ð¡)2 . ðŠâðž
ð¡âðœð ðŠâðž
100
Chapter 5. Rings and generalized distances
We can conclude that â |ÎðŠ (ðž)| ⥠ðŠâðž
âð¡âðœ
ð
|ðž|4 , âðŠâðž ððŠ (ð¡)2
so we need only to find a suitable upper bound for âð¡,ðŠ ððŠ (ð¡)2 . To this end notice that ððŠ (ð¡) = â ðž(ð¥). ð¥âðœð ð
âð¥âðŠâ=ð¡
Squaring gives us ððŠ (ð¡)2 =
ðž(ð¥)ðž(ð¥â² ).
â ð¥,ð¥â² âðœð ð âð¥âðŠâ=âð¥â² âðŠâ=ð¡
Summing over ð¡ â ðœð and ðŠ â ðž then leads us to â â ððŠ (ð¡)2 = â ð¡âðœð ðŠâðž
â
ð¡âðœð
ðž(ð¥)ðž(ð¥â² )ðž(ðŠ)
ð¥,ð¥â² ,ðŠâðœð ð
âð¥âðŠâ=âð¥â² âðŠâ=ð¡
â
=
ðž(ð¥)ðž(ð¥â² )ðž(ðŠ).
ð¥,ð¥â² ,ðŠâðœð ð âð¥âðŠâ=âð¥â² âðŠâ
Applying orthogonality in dimension ð = 1, we see that âð¡âðœ âðŠâðž ððŠ (ð¡)2 ð equals the quantity ðâ1 â
â
ð(ð (âð¥ â ðŠâ â âð¥â² â ðŠâ))ðž(ð¥)ðž(ð¥â² )ðž(ðŠ).
(5.18)
ð âðœð ð¥,ð¥â² ,ðŠâðœð ð
A quick calculation (see Exercise 5.1) shows that (5.18) becomes â â ððŠ (ð¡)2 = ðâ1 â
â
ð(ð (âð¥â â 2ð¥ â
ðŠ))ð(âð (âð¥â² â â 2ð¥â² â
ðŠ))
ð âðœð ð¥,ð¥â² ,ðŠâðž
ð¡âðœð ðŠâðž
2
=ð
â1
| | â â || â ð(ð (âð¥â â 2ð¥ â
ðŠ))|| | ð âðœð ðŠâðž |ð¥âðž
= ðŒð =0 + ðŒðŒð â 0 , where 2
| | ðŒð =0 = ðâ1 â â || â ð(ð (âð¥â â 2ð¥ â
ðŠ))|| = ðâ1 |ðž|3 | ð =0 ðŠâðž |ð¥âðž
5.4. Pinned distances
101
and
2
ðŒðŒð â 0 = ð
â1
| | â â || â ð(ð (âð¥â â 2ð¥ â
ðŠ))|| . | ð â 0 ðŠâðž |ð¥âðž
We finally need only to bound the last sum ðŒðŒð â 0 . We first extend the sum in ðŠ â ðž to a sum in ðŠ â ðœðð : 2
| | ðŒðŒð â 0 = ðâ1 â â || â ð(ð (âð¥â â 2ð¥ â
ðŠ))|| | ð â 0 ðŠâðž |ð¥âðž
2
â€ð
â1
| | â â || â ð(ð (âð¥â â 2ð¥ â
ðŠ))|| | ð â 0 ðŠâðœð |ð¥âðž ð
=ð
â1
â ð(ð (âð¥â â âð¥â² â))ð(2ð ðŠ â
(ð¥ â ð¥â² )).
â â
ð â 0 ðŠâðœð ð¥,ð¥â² âðž ð
Summing in ðŠ â ðœðð and applying orthogonality yields ðŒðŒð â 0 †ððâ1 â ð â 0
â
ð(ð (âð¥â â âð¥â² â))
ð¥,ð¥â² âðž âð¥â=âð¥â² â
= ððâ1 (ð â 1)|ðž| †ðð |ðž|. Notice that ðŒð =0 ⥠ðŒðŒð â 0 whenever ðâ1 |ðž|3 ⥠ðð |ðž|. It follows that â â ððŠ (ð¡)2 = ðŒð =0 + ðŒðŒð â 0 †2ðâ1 |ðž|3 ð¡âðœð ðŠâðž
whenever |ðž| ⥠ð
ð+1 2
. Putting together all of the pieces, we see that
â |ÎðŠ (ðž)| ⥠ðŠâðž
so long as |ðž| ⥠ð
âð¡âðœ
ð
ð+1 2
|ðž|4 |ðž|4 ⥠2ðâ1 |ðž|3 âðŠâðž ððŠ (ð¡)2
. Thus, in this range of ðž, we have 1 1 â |Î (ðž)| ⥠ð. |ðž| ðŠâðž ðŠ 2
Theorem 5.4.3 follows from the pigeonhole principle. That is, there must 1 exist a ðŠ â ðž such that |ÎðŠ (ðž)| ⥠2 ð.
102
Chapter 5. Rings and generalized distances
Note that the proof of the above result was from [17], and if you more carefully analyze the final bound: 1 1 â |Î (ðž)| ⥠ð, |ðž| ðŠâðž ðŠ 2 you achieve the following. ð+1
Theorem 5.4.4. Let ðž â ðœðð , and suppose |ðž| â« ð 2 . Then there exists a subset ðž â² â ðž such that |ðž â² | â« |ðž|, and for all ðŠ â ðž â² , we have |ÎðŠ (ðž)| â¥
1 ð. 2
When ð = 2, this result was improved in [71]. Theorem 5.4.5. Let ðž â ðœ2ð , and suppose |ðž| â« ð4/3 . Then there exists a subset ðž â² â ðž such that |ðž â² | â« |ðž|, and for all ðŠ â ðž â² , we have |ÎðŠ (ðž)| â« ð. Interestingly, Theorem 5.4.5 was proved used a geometric argument involving perpendicular bisectors and point-line incidences in ðœðð . It should be noted that Theorem 5.4.5 is a strengthening of Theorem 4.1.1. Finally the following was obtained in [122] for product sets in ðœð2 . Theorem 5.4.6. Let ðž = ðŽ à ðŽ â ðœð2 , where ð is prime. Then, there exists ðŠ â ðž such that |ÎðŠ (ðž)| â« min{ð, |ðž|3/4 }. Note that Theorems 5.4.5 and 5.4.6 both give the same bound: |ðž| â« ð4/3 â¹ |ÎðŠ (ðž)| â« ð. Furthermore Theorem 5.4.5 holds for all sets ðž â ðœ2ð for a positive proportion of ð¥ â ðž, while Theorem 5.4.6 holds only for product sets in ðœð2 for ð a prime, and for at least one ðŠ â ðž. However, Theorem 5.4.6 does give a quantitative upper bound for |ÎðŠ (ðž)| when ðž = ðŽ à ðŽ has small cardinality.
5.5. Exercises: Chapter 5
103
5.5 Exercises: Chapter 5 Exercise 5.1. Verify that for ð¥, ðŠ, ð§ â ðœðð , we have âð¥ â ðŠâ â âð§ â ðŠâ = (âð¥â â 2ð¥ â
ðŠ) â (âð§â â 2ð§ â
ðŠ). Exercise 5.2. Complete the square in â€ð . That is if gcd(2ð, ð), then find all ð¥ â â€ð such that ðð¥2 + ðð¥ + ð = 0. Exercise 5.3. Suppose that ð â â€Ãð , where ð ⥠3 is odd. Prove that ðº(ð, ð, ð) = ðº(ð, ð)ð(âð2 /4ð). Exercise 5.4. Prove that |ðº(ð, ð)|2 †ð for all ð â â€ð . Exercise 5.5. Suppose ð is a Dirichlet character mod ð and gcd(ð, ð) = 1. Show that ð(ð, ðð ) = ð(ð)ð(ð). Exercise 5.6 (Chinese Remainder Theorem). Suppose that ð and ð are positive integers, each greater than or equal to 2. Prove that if gcd(ð, ð) = 1, then we have the following ring isomorphisms: â€ðð â
â€ð à â€ð and â€Ãðð â
â€Ãð à â€Ãð . Exercise 5.7. Prove Theorem 5.2.1. That is, using the techniques laid out in the proof of Theorem 4.1.1, prove that if ðž, ð¹ â ðœ2ð and |ðž||ð¹| â« ð8/3 , then |Î(ðž, ð¹)| â« ð. Exercise 5.8. Prove the following elementary interpolation theorem4 : Let (ðð ) be a sequence of positive real numbers, and suppose that ð
â ðð †ðŽ ð=1 4 First shown to me by Alex Iosevich, and it appears in his fantastic book A View From the Top ([81]).
104
Chapter 5. Rings and generalized distances
while
1/2
ð
(â
ðð2 )
†ðµ.
ð=1
Then we have ð
(â
1/ð ð ðð )
2
†ðŽð
2
â1 2â ð
ðµ
ð=1
for all 1 †ð †2. Exercise 5.9. Show that if the sets ðž, ð¹ â ðœðð satisfy |ðž||ð¹| â« ðð+1 , then Î(ðž, ð¹) â ðœâð . When must 0 â Î(ðž, ð¹)?
10.1090/car/037/06
Chapter
6
Configurations and group actions 6.1 Finite configurations In the previous chapters we were working with the distance set Î(ðž) = {âð¥ â ðŠâ ⶠð¥, ðŠ â ðž} and its vertex set ð·ð¡ (ðž) = {(ð¥, ðŠ) â ðž à ðž ⶠâð¥ â ðŠâ = ð¡}, where ðž â ðœðð . It is instructive to think of distances as 2-point configurations. In this chapter, we study ð-point configurations, but some extra care is required.1 All of these problems can be extended in the same ways as the distance problem (working over different spaces, using different notions of distance, etc.), but we will concentrate on configurations over finite fields as that setting is the simplest and cleanest. As a first glimpse letâs consider an example. Let ðž â ðœ2ð , let â1 , â2 , â3 â â ðœð , and consider the set {(ð¥, ðŠ, ð§) ⶠâð¥ â ðŠâ = â1 , âðŠ â ð§â = â2 , âð¥ â ð§â = â3 } of all 3-point configurations with side-lengths â1 , â2 , and â3 , respectively. It is sometimes convenient to assume that âð â 0 so that we can avoid 1 At this point in the text, we cease to provide as elementary an overview as possible, opting instead to deliver a succinct history of the development of this area as it is expansive. We invite the reader to read the original sources whenever possible as we only provide the gist of each proof where appropriate.
105
106
Chapter 6. Configurations and group actions
triangles containing isotropic lines (i.e., nontrivial lines with length 0), specifically when ð â¡ 1 (mod 4).
Figure 6.1. The nondegenerate triangle (ðŽ, ðµ, ð¶) with distances â1 , â2 , â3 ; the degenerate triangle (ð·, ðž, ð¹).
We will refer to a 3-point configuration as a âtriangleâ though it should be clear we are only ever discussing the vertex set of a triangle and not the edges between the vertices (see Figure 6.1). We will say that two triangles are equivalent if one is a rotated, shifted copy of the other2 . We say that a triangle is nondegenerate if there is no line containing all three vertices of a triangle. It turns out that this set of equivalence classes of triangles is 3-dimensional3 . Already a natural question should spring to mind: Question 6.1.1. Let ðž â ðœ2ð . What is the minimum power ðœ such that there exists a constant ð > 0 so that |{(âð¥ â ðŠâ, âðŠ â ð§â, âð¥ â ðŠâ) ⶠð¥, ðŠ, ð§ â ðž}| ⥠ðð3 , whenever |ðž| â« ððœ ? Before we can answer this question, we must now rigorously define finite point configurations in ðœðð . A ð-simplex will refer to a set of (ð+1)points in ðœðð , so that a triangle (or more precisely, the vertices of a triangle) forms a 2-simplex, and a distance is a 1-simplex. A priori there 2 This 3 This
language will be justified shortly. is not as obvious as one may think!
6.1. Finite configurations
107
is no relationship between ð and ð. We say that two ð-simplices Î1 = {ð¥0 , . . . , ð¥ð } â ðœðð and Î2 = {ðŠ0 , . . . , ðŠ ð } â ðœðð are equivalent (Î1 ⌠Î2 ) if there exists a translation vector ð â ðœðð and an orthogonal matrix ð â ð ð (ðœð ) such that ð¥ð = ððŠ ð + ð, for all ð = 0, . . . , ð, where ð¥ð , ðŠ ð , and ð are viewed as column vectors. Just like in the Euclidean setting, a matrix ð is orthogonal if ðð = ðâ1 . It is easy to check that this defines an equivalence relation (see Exercise 6.3), so we can now genuinely say that Î1 and Î2 are equivalent. We frequently abuse notation and refer to the linear transformation determined by ð as simply ð itself. Now, we say a ð-simplex Î is nondegenerate if the vertices ð¥0 , . . . , ð¥ð are affinely independent (i.e., the set of points ð¥1 â ð¥0 , . . . , ð¥ð â ð¥0 are linearly independent). Note that a ðsimplex in ðœðð must be degenerate if ð > ð, so we will only concentrate on the case ð †ð with the case ð = ð (such as triangles in ðœ2ð ) perhaps being the most compelling. Let ⌠denote the equivalence of simplices in ðœðð under translations and rotations defined above, and consider ðð (ðž) = ðž ð+1 / âŒ, the resulting quotient set of equivalence classes. The following proposition will give the dimensionality of the set of simplices. Proposition 6.1.2. Suppose that Î1 = {ð¥0 , . . . , ð¥ð } and Î2 = {ðŠ0 , . . . , ðŠ ð } are two nondegenerate simplices in ðœðð . Then Î1 ⌠Î2 if and only if you can order the simplices Î1 and Î2 such that âð¥ð â ð¥ð â = âðŠ ð â ðŠð â for all pairs 0 †ð < ð †ð. Proof. Exercise. (See Exercise 6.4.) Note that we can encode any nondegenerate ð-simplex as a (ð + 1) à (ð + 1) symmetric matrix. For example, the set of triangles {ð¥, ðŠ, ð§} with sides âð¥ â ðŠâ = ð, âð¥ â ð§â = ð, âðŠ â ð§â = ð can be encoded as a 3 à 3 matrix 0 ð ð [ð 0 ð ] . ð ð 0
108
Chapter 6. Configurations and group actions
Different values of ð, ð, and ð will lead to different triangles. A straightforward count shows that the set of all (possibly degenerate) triangles is 3-dimensional. Generalizing this, we see that the set of nondegenerate ð-simplices is (ð+1 )-dimensional. When ð †ð, the set of degenerate 2 simplices is much smaller than the nondegenerate simplices (see Exercise 6.6). We are now ready to state some results. We will start with the earliest results and work our way forward chronologically. In [25] the authors studied ð-simplices in ðœðð . The main result was the following: Theorem 6.1.3. Let ðž â ðœðð satisfy |ðž| ⥠ððð , where ðâ1/2 ⪠ð †1. Then the set of ð-simplices satisfies |ðð (ðž)| â« ððâ1 ð(
ð+1 ) 2
.
Remark 6.1.4. The proof of Theorem 6.1.3 essentially amounts to estimating |{(ð¥, ð¥1 , . . . , ð¥ð ) â ðž ð+1 ⶠâð¥ â ð¥ð â = 1}| , and then employing some combinatorial reasoning. The original bound on that quantity led to the range ðâ1/2 ⪠ð †1. The range was later ðâ1
improved to ðâ 2 ⪠ð †1 in a subsequent paper ([6]), though better results are now known. In [73] Hart and Iosevich gave the first bound on general ð-simplices in ðœðð albeit for ð ⪠âð: Theorem 6.1.5. Fix ð, and let ð > (ð+1 ). Let ðž â ðœðð . Then 2 ð+1 ) 2
|ðð (ðž)| â« ð( ðð
ð
so long as |ðž| â« ð ð+1 + 2 . The main thrust of the argument was an induction argument used to âbootstrapâ from (ð â 1)-simplices to ð-simplices. Theorem 6.1.6. We define a ð-simplex recursively: Let ð = (ð¡ ð,ð ) denote a ð à ð symmetric matrix with entries in ðœâð . Let ð ð = {ð¡ ð,1 , . . . , ð¡ ð,ð }
6.1. Finite configurations
109
denote the first ð values in the ðth row, and finally let âð denote the set of values on or under the main diagonal in the first ð rows of the matrix, so âð = {ð¡1,1 ; ð¡2,1 ; ð¡2,2 ; . . . ; ð¡ ð,1 ; . . . ; ð¡ ð,ð } = ð1 ⪠⯠⪠ð ð . Now, we write ð¯â1 = {(ð¥0 , ð¥1 ) â ðžÃðž ⶠâð¥0 âð¥1 â = ð¡1,1 }, so that is the pair of vertices in ðž who are distance ð¡1,1 apart. Then, ð¯â2 is the set of triples in ðž whose vertices form triangles with side lengths ð¡1,1 ; ð¡2,1 ; ð¡2,2 . ðâ3 is the set of 6 = (3+1 ) vertices in ðž determining distances ð¡1,1 ; ð¡2,1 ; ð¡2,2 ; ð¡3,1 ; ð¡3,2 ; ð¡3,3 , 2 and so on for larger simplices. Formally, we have ð¯âð ={(ð¥0 , . . . , ð¥ðâ1 , ð¥ð ) â ð¯âðâ1 à ðž ⶠâð¥ð â ð¥ð â â ð ð }. Note that varying the values in the matrix ð will change the vertex set, though just like the sphere of varying radii having the same cardinality, ð+1 the values in ð will not change our estimates. Then for all âð â (ðœâð )( 2 ) , we have |ðž|ð+1 |ð¯âð | = ð+1 (1 + ð(1)) ð( 2 ) ðð
ð
when |ðž| â« ð ð+1 + 2 . In particular |ð¯âð | > 0 for ðž in the same range. The proof that |ð¯â1 | =
|ðž|2 (1 + ð(1)) ð
ð+1
when |ðž| â« ð 2 is the estimate we derived in (3.6). For the inductive step, the authors write |ðâð | =
â
â
ð¯âðâ1 (ð¥0 , . . . , ð¥ðâ1 )ðž(ð¥ð ),
ð¥0 ,. . .,ð¥ð âð¥0 âð¥ð â=ð¡1,ð ;. . .;âð¥ðâ1 âð¥ð â=ð¡ð,ð
and they apply the same Fourier Analytic techniques that were applied in the distance problem. Here ð¯âð (â
) is the characteristic function of the set ðâð . An improvement to Theorem 6.1.5 was given in [17]. Theorem 6.1.7. Fix ð †ð â 1, and suppose that ðž â ðœðð satisfies |ðž| â« ð+ð
ð 2 . Then ðž determines a positive proportion of the ð-simplices. That is, for sets ðž of sufficiently large cardinality we have |ðð (ðž)| â« ð(
ð+1 ) 2
.
110
Chapter 6. Configurations and group actions
The proof of Theorem 6.1.7 involves aspects of the previous proofs: both bootstrapping and estimating the number of so-called ð-stars: |{(ð¥, ð¥1 , . . . , ð¥ð ) â ðž ð+1 ⶠâð¥ â ð¥ð â = ð¡ ð }| where ð¡ ð â ðœâð for ð = 1, . . . , ð. The authors were essentially able to prove that |{(ð¥, ð¥1 , . . . , ð¥ð ) â ðž ð+1 ⶠâð¥ â ð¥ð â = ð¡ ð }| > 0 (6.1) ð+ð
for all choices ð¡ ð â ðœâð for ð = 1, . . . , ð, whenever |ðž| â« ð 2 . The pigeonhole principle then allows us to move from ð-stars to ð-simplices. Despite not being able to improve the ð-simplex bound |ðž| â« ð inequality (6.1) was extended to sets of smaller cardinality ([6]):
ð+ð 2
, the
Proposition 6.1.8. Let ðž â ðœðð , and let ð¡1 , . . . , ð¡ ð â ðœâð . Then |{(ð¥, ð¥1 , . . . , ð¥ð ) â ðž ð+1 ⶠâð¥ â ð¥ð â = ð¡ ð }| > 0 for all ð
0, then max(ð, ð) ⥠âðð. Exercise 6.2. Let ðº denote the 1-distance graph of the entire finite field ðœðð . Prove that the chromatic number of the distance graph satisfies 10 ð(ðº) â« ð
ðâ1 2
.
Exercise 6.3. Fix ð¡ â ðœâð . Prove that ðº ð¡ is simple, undirected, connected, and regular. What other properties does ðº ð¡ have? Exercise 6.4. Let Î1 = {ð£ 0 , . . . ð£ ð } and Î2 = {ð€ 0 , . . . , ð€ ð } be two nondegenerate simplices in ðœðð . Prove that if âð£ ð â ð£ð â = âð€ ð â ð€ð â for all 0 †ð < ð †ð, then Î1 ⌠Î2 . Exercise 6.5. If Î1 is nondegenerate and Î1 ⌠Î2 , then Î2 is nondegenerate. Exercise 6.6. Fix ð and ð such that ð †ð. Let ðŽ â ðð (ð) be the set of ð+1 nondegenerate simplices. Prove that ðŽ = ð (ð( 2 ) ) . 10 I had once thought that ð(ðº) should be uniformly bounded independent of ð, which turned out to be absurdly false!
6.6. Exercises: Chapter 6
127
Exercise 6.7. Let ðºð¿2 (ðœð ) denote the set of invertible 2 à 2-matrices with entries in ðœð . Show that |ðºð¿2 (ðœð )| = (ð2 â 1)(ð2 â ð). What is the cardinality of ðºð¿ð (ðœð )? Exercise 6.8. Let ð(2; ð) denote the set of 2 à 2-orthogonal matrices with entries in ðœð . That is, ð(2; ð) = {ðŽ â ð2 (ðœð ) ⶠðŽðŽð = ðŒ} . Prove that |ð(2; ð)| = {
2ð â 2 2ð + 2
ð â¡ 1 (mod 4) ð â¡ 3 (mod 4).
Hint: Start by showing that |ðð(2; ð)| = |ð 1 |, where ðð(2; ð) = {ðŽ â ð(2; ð) ⶠdet(ðŽ) = 1}. Exercise 6.9. Let Graff(ð, ð) denote the set of ð-dimensional affine planes in ðœðð . Let (ð¥, â) â ðœðð à Graff(ð â 1, ð). We say that (ð¥, â) is equivalent to (ð¥â² , ââ² ) if after translating ð¥ to ð¥â² , there exist an orthogonal matrix ð â ð ð (ðœð ) that maps â to ââ² . Prove that this is indeed an equivalence relation. Let Î(ðž, ð¹) denote the quotient set of this equivalence relation. Exercise 6.10 ([10, 124]). Prove that if ðž â ðœðð , and ð¹ is a set of non1 degenerate11 planes in Graff(ð â 1, ð), then |Î(ðž, ð¹)| ⥠2 ð, so long as 1
|ðž||ð¹| > ðð+1 . If ð = 2, then |Î(ðž, ð¹) ⥠2 ð when |ðž||ð¹| > ð8/3 .
11 Here,
non-degenerate means that the normal vector to the plane is nonisotropic.
10.1090/car/037/07
Chapter
7
Combinatorics in finite fields While I am primarily captivated by the study of the distance set problem and its variants in Euclidean and discrete settings, there are many other problems in geometric combinatorics that are worth exploring. We give only a brief and terse introduction to each topic. This list is not meant to be exhaustive, but these are the problems that I personally find most captivating in the area. The motivated reader should explore topics as deeply as possible, and we provide sufficient resources for one to do so.
7.1 Incidence theory Let ðœð denote the ð-dimensional vector space over some field ðœ. Given a set ð â ðœð of points and set Î of curves in ðœð , we define an incidence as a pair (ð, ðŸ) â ð à Πsuch that ð â ðŸ (see Figure 7.1). A major problem in geometric combinatorics is to determine the number of incidences between finite point-sets and finite sets of lines. As geometric as this problem feels, counting incidences turns out to have major combinatorial and number-theoretic implications. For example, we have already seen that bounding the number of incidences of points and lines in â2 can be used to give a nontrivial1 bound on the ErdÅs unit-distance problem (see (2.5)). Recall that the main driver of this result was an incidence result of Szemerédi and Trotter which we restate here for convenience. Theorem 7.1.1 (Szemerédi-Trotter, Version 1). Let ð â â2 be a finite point set, and suppose ð¿ is a finite set of lines in the plane â2 . Then the 1 In
fact, this method gives the current world record!
129
130
Chapter 7. Combinatorics in finite fields
Figure 7.1. A set of points 8 points and 6 lines determining 11 incidences.
number of incidences ðŒ(ð, ð¿) satisfies the bound ðŒ(ð, ð¿) ⪠|ð|2/3 |ð¿|2/3 + |ð| + |ð¿|. In particular if we consider a set of ð points ð and ð lines ð¿ in â2 , then we have the bound ðŒ(ð, ð¿) ⪠ð4/3 . The Szemerédi-Trotter bound can also be worded as follows: Theorem 7.1.2 (Szemerédi-Trotter, Version 2). Let ð¿ be a finite set of lines in â2 . Let ðð denote the set of points in â2 that lie in at least ð lines. Then |ðð | ⪠|ð¿|2 ðâ3 + |ð¿|ðâ1 . The Szemerédi-Trotter bound is sharp up to constants, at least when ð and ð¿ are comparable in cardinality. Example 7.1.3. Let ð = {(ð¥, ðŠ) â â€2 ⶠ1 †ð¥ †ð, 1 †ðŠ †2ð2 } and ð¿ = {(ð¡, ðð¡ + ð) ⶠð, ð â â€, 1 †ð †ð, 1 †ð †ð2 }. Then |ð| â |ð¿| â ð3 , and ðŒ(ð, ð¿) â ð4 . There are numerous incidence results for higher dimensions ([144], for example), though none have proven as useful (so far) as SzemerédiTrotter. My favorite proof of the Szemerédi-Trotter theorem is a clever
7.1. Incidence theory
131
graph-theoretic proof originally given by Székely ([146]), and we outline that proof now.
7.1.1 Proof of Szemerédi-Trotter (Theorem 7.1.1) We refer the reader unfamiliar with graph theory to Section 6.5 for a terse introduction to the necessary terminology. We assume all graphs are undirected, finite (meaning that the graph has a finite number of vertices), and simple. A cycle is a path that starts and ends on the same vertex, and which traverses no other vertex more than once. Recall that a graph is planar if its crossing number satisfies cr(ðº) = 0. That is, a graph is planar if and only if it can be drawn in the plane so that no edges intersect. For a planar drawing of a planar graph, there is a well-defined number of faces or regions bounded by edges, including the outer unbounded region (see Figure 7.2).
Figure 7.2. A planar graph with 5 vertices, 6 edges, and 3 faces (including the unbounded face).
Proposition 7.1.4. Let ðº = (ð, ðž) be planar, finite, simple, and connected. If ðº contains at least 3 vertices, then the number of faces |ð¹| satisfies 3|ð¹| †2|ðž|. Proof. If ðº has 3 vertices and 2 edges, then ðº has 1 face, in which case 3(1) †2(2). Otherwise, each face is determined by at least 3 edges, and each edge corresponds to at most 2 faces. The result follows.
132
Chapter 7. Combinatorics in finite fields
Proposition 7.1.5. Let ðº = (ð, ðž) be planar, finite, simple, and connected. If ðº contains at least 3 vertices, then |ðž| †3|ð|. Proof. This follows from the previous proposition together with the wellknown Euler formula on planar graphs: A planar graph with |ðž| edges, |ð| vertices, and |ð¹| faces satisfies the formula |ð¹| + |ð| â |ðž| = 2.
Proposition 7.1.6. Let ðº be a finite, simple, connected graph. Then cr(ðº) > |ðž| â 3|ð|. Proof. Suppose for a contradiction that cr(ðº) †|ðž| â 3|ð|. Then the planar subgraph ðº â² obtained by removing the cr(ðº) †|ðž| â 3|ð| overlapping edges would have more than |ðž| â (|ðž| â 3|ð|) = 3|ð| edges, contradicting Proposition 7.1.5. The key ingredient is the following result colloquially referred to as the Crossing Number Inequality. Theorem 7.1.7. Let ðº = (ð, ðž) be a finite simple graph, and suppose that |ðž| ⥠4|ð|. Then |ðž|3 cr(ðº) â« . |ð|2 We now prove Szemerédi-Trotter via the Crossing Number Inequality. Let ð be a finite point set in which we will make the vertices of our graph ðº. Now consider a finite set of lines ð¿. We say that two points ð¥, ðŠ â ð are consecutive on some line â â ð¿ if there is no third point ð§ on the line segment ð¥ðŠ. Finally, we connect two vertices ð¥, ðŠ â ð with an edge if and only if ð¥ and ðŠ are consecutive on some line. We have ðŒ(ð, ð¿) = |ðž| + |ð¿|, by noticing that any line containing ð points will determine ð â 1 edges to the graph, and then adding this over all lines. In particular, |ðž| = ðŒ(ð, ð¿) â |ð¿|. If |ðž| ⥠4|ð|, we apply the Crossing Number Inequality to see that (ðŒ(ð, ð¿) â |ð¿|)3 . cr(ðº) â« |ð|2
7.1. Incidence theory
133
On the other hand, notice that cr(ðº) †(|ð¿| ) â |ð¿|2 , since each crossing 2 comes from a pair of lines. It follows that (ðŒ(ð, ð¿) â |ð¿|)3 ⪠|ð¿|2 , |ð|2 and it follows that ðŒ(ð, ð¿) ⪠|ð¿|2/3 |ð|2/3 + |ð¿|. If |ðž| < 4|ð|, then ðŒ(ð, ð¿) â |ð¿| < 4|ð|, so that ðŒ(ð, ð¿) < 4|ð| + |ð¿|. This completes the proof of Theorem 7.1.1. Notice that the only fact about lines we needed in the proof is that two distinct lines can intersect in at most a finite number of points. Thus, the Szemerédi-Trotter Theorem can be extended to any set of curves which do not intersect themselves in more than a constant number of points.
7.1.2 Incidences over finite fields We are now ready to discuss incidences in finite fields. At this point we should be comfortable with points and lines in ðœðð , and incidences in ðœðð again refer to ordered pairs of points ð and lines â such that ð â â. We first note that nothing close to as strong as the Szemerédi-Trotter can hold in finite fields. Example 7.1.8. Let ð = ðœ2ð , and consider the set of all lines ð¿ in ðœ2ð . Note that |ð| = ð2 , while |ð¿| = ð2 + ð â ð2 = |ð|. Then ðŒ(ð, ð¿) â ð3 = (ð2 )3/2 , and thus the exponent 3/2 is best possible in finite fields. The part of the proof of Szemerédi-Trotter that breaks down in the finite field setting is the notion of âconsecutiveness,â which was the condition we used to determine the edge set in our graph. Thus an alternate approach is needed. Proposition 7.1.9. Suppose that we have a set of points ð â ðœ2ð and a set of lines ð¿ in ðœ2ð . Then the number of incidences ðŒ between the set of points ð and the set of lines ð¿ satisfies ðŒ †|ð| + |ð|1/2 |ð¿| and ðŒ †|ð¿| + |ð¿|1/2 |ð|. In particular, if we have a set of ð points and ð lines in ðœ2ð , then, ðŒ ⪠ð 3/2 .
134
Chapter 7. Combinatorics in finite fields
Proof. We prove only the first inequality. We will use Iverson-bracket notation: 1 ð¥ââ [ð¥ â â] = { 0 otherwise. Let ð(â) denote the number of points on the line â, so that ð(â) = â [ð¥ â â]. ð¥âð
Thus we have ðŒ = â ð(â) = â â [ð¥ â â]. ââð¿
ââð¿ ðâð
By Cauchy-Schwarz, we have 2 2
ðŒ = ( â ð(â)) †â 12 â ð(â)2 = |ð¿| â ð(â)2 . ââð¿
ââð¿
ââð¿
(7.1)
ââð¿
Now we see that â ð(â)2 = â â [ð¥ â â] â [ðŠ â â] ââð¿
ââð¿ ð¥âð
ðŠâð
= â â [ð¥ â â]2 + â â [ð¥ â â][ðŠ â â] ââð¿ ð¥âð
ââð¿ ð¥,ðŠâð ð¥â ðŠ
where the first sum handles the situation when ð¥ = ðŠ. Clearly, â â [ð¥ â â]2 = â â [ð¥ â â] = ðŒ. ââð¿ ð¥âð
ââð¿ ð¥âð
Also, since there is at most one line that passes through the distinct points ð¥ and ðŠ, it follows that â â [ð¥ â â][ðŠ â â] †|ð|2 . ââð¿ ð¥,ðŠâð ð¥â ðŠ
Combining these results with (7.1), we see that ðŒ 2 †|ð¿| â
(ðŒ + |ð|2 ). Completing the square in ðŒ yields 2
(ðŒ â
|ð¿|2 |ð¿| . ) †|ð¿||ð|2 + 2 4
7.1. Incidence theory
135
Taking square roots and applying the inequality âð¥ + ðŠ †âð¥ + âðŠ (which is trivially valid for all nonnegative ð¥ and ðŠ), we finally see that ðŒ †|ð¿|1/2 |ð| + |ð¿|. The second inequality follows similarly, and it uses only the fact that two distinct lines can have at most one common point. As before the proof can be modified slightly to accommodate other curves (such as circles, arcs, etc.) in the plane so long as the curves do not self-intersect too often. We have already seen numerous (often ingenious) applications of incidence theory used to solve many combinatorial problems. In the next section we will explore my favorite such application: a connection discovered by G. Elekes ([41]) between incidence theory and the so-called sum-product problem of ErdÅs and Szemerédi. Other applications include partial solutions to the unit-distance problem (see (2.5)) as we discussed above. We will spend the rest of the section detailing the most recent incidence bounds in finite fields and other fields. These bounds will be useful in turn in later sections. The first nontrivial incidence result for finite fields was given by Bourgain, Katz, and Tao ([14]). In what follows, ððœ3ð can be viewed as the plane ðœ2ð plus the line at infinity2 . Theorem 7.1.10. Let ð be prime, and consider a set of points ð and a set of lines ð¿ in ððœð3 . Let ðŒ â (0, 2), and set ð = ððŒ . If |ð|, |ð¿| †ð, then ðŒ(ð, ð¿) âªð ð 3/2âð for some ð > 0. The relation between ð and ðŒ was nontransparent from their proof. However an explicit estimate was first given Helfgott and Rudnev ([77]). Theorem 7.1.11. Let ð be prime, and consider a set of points ð and a set of lines ð¿ in ððœð3 . If |ð| = |ð¿| = ð, where ð < ð, then 3
1
ðŒ(ð, ð¿) ⪠ð 2 â 10678 . 2 Really,
details.
ððœ3ð is the set ðœ3ð ⧵ {(0, 0, 0)} quotiented out by dilations. See [14] for more
136
Chapter 7. Combinatorics in finite fields
This result was later improved by Jones ([91]). Theorem 7.1.12. Let ð be prime, and consider a set of points ð and a set of lines ð¿ in ððœð3 . If |ð| = |ð¿| = ð, where ð < ð, then 3
1
ðŒ(ð, ð¿) ⪠ð 2 â 806 +ð(1) . This result was further improved numerous times, with the best known results now belonging to Stevens-de Zeeuw ([145]). Theorem 7.1.13. Let ð = ðð be the power of a prime, and consider a set of points ð and a set of lines ð¿ in ððœ3ð . If |ð| = |ð¿| = ð, where ð < ð15/11 , then 3 1 ðŒ(ð, ð¿) ⪠ð 2 â 30 +ð(1) . Theorem 7.1.13 is only a corollary of the work of Stevens-de Zeeuw, and we encourage the reader to read their paper as it contains many important mathematical ideas, along with a terse recent history of incidence results over general (not necessarily finite) fields. Again many incidence results hold for sets of curves other than lines, so long as the curves do not self-intersect, and as long as we can bound the number of curves on which a single point may lie. Next we explore incidences in higher dimensions and with general curves. In what follows, a plane in ðœ3ð is a set of the form {ð¥â + ð¡ðŠ â + ð ð§ â ⶠð¡, ð â ðœð }, where ð¥,â ðŠ,â ð§ â â ðœ3ð , and where ðŠ â and ð§ â are nonzero. Theorem 7.1.14 ([130]). Let ð = ðð be a power of an odd prime, and consider a set ð points and a set ð planes in ðœ3ð where |ð| = ð and |ð| = ð. Assume that every line in ðœ3ð contains ð or fewer points of ð. Assume also that ð ⥠ð and ð ⪠ð2 . Then we have the bound ðŒ(ð, ð) ⪠ðâð + ðð. Theorem 7.1.14 served as the basis for the proof of Theorem 7.1.13 along with other important results (such as [131]). Theorem 7.1.15 (Cilleruelo, Iosevich, Lund, Roche-Newton, Rudnev, [20]). Let ð â ðœðð be a set of points, and ð® a set of spheres in ðœðð . Then |ð||ð®| | | |ðŒ(ð, ð®) â | < |ð||ð®|ðð . | ð | â
7.2. Sum-product phenomena
137
Theorem 7.1.15 was used to establish some pinned-distance results in ðœðð , and it was also used to establish an analogue of Beckâs Theorem for Circles in ðœ2ð . The Stephens-De Zeeuw bound (Theorem 7.1.13) and a related result ([131]) were used to prove the following variant (a so-called âsmall setâ variant) of the Finite Field Distance Problem. Theorem 7.1.16 ([86]). Let ðŽ â ðœð2 , where |ðŽ| †ð7/6 , and ð â¡ 3 (mod 4). Then, 1
149
|Î(ðŽ)| â« |ðŽ| 2 + 4214 . Finally, there has also been an incidence result over cyclic rings. Theorem 7.1.17 ([125]). Let â€ð be the ring of integers modulo ð, and let ðŸ(ð), ð(ð), and ð(ð) denote the smallest prime factor of ð, the Euler totient function, and the number of positive divisors of ð, respectively. Then for a collection of points ð and lines ð¿ in â€2ð , we have ðŒ(ð, ð¿) â€
|ð||ð¿| 2ð(ð)ð2 |ð||ð¿| . + ð ð(ð) â ðŸ(ð)
7.2 Sum-product phenomena Given a subset ðŽ of a ring (ð
, +, â
), we define the sum set, product set, difference set, and quotient set3 of ðŽ as ðŽ + ðŽ = {ð + ð ⶠð, ð â ðŽ}, ðŽ â
ðŽ = {ð â
ð ⶠð, ð â ðŽ}, ðŽ â ðŽ = {ð â ð ⶠð, ð â ðŽ}, ðŽ/ðŽ = {ððâ1 ⶠð, ð â ðŽ}, respectively. If ðŽ â â and |ðŽ| = ð, then we have the trivial bounds 2ð â 1 †|ðŽ + ðŽ|, |ðŽ â
ðŽ| †(
ð+1 ). 2
3 In order for the quotient set of ðŽ to be well defined, one needs to ensure that all elements of ðŽ are invertible.
138
Chapter 7. Combinatorics in finite fields
As an example consider the set ðŽ = {ð, ð + ð, . . . , ð + (ð â 1)ð}, an arithmetic progression of length ð. Then ðŽ + ðŽ = {2ð, 2ð + ð, . . . , 2(ð + (ð â 1)ð)}, so that |ðŽ + ðŽ| = 2ð â 1. On the other hand, it has been shown by Ford ([56]) that ð2 |ðŽ â
ðŽ| â , (log ð)ð¿ (log log ð)3/2 1+log log 2
= 0.086071 . . . .4 Thus while the sum set is where ð¿ = 1 â log 2 small, the product set is large. This result of Ford also settled the socalled Multiplication Table Problem. Given an ðÃð multiplication table, how many unique entries exist on that table? This is really asking for the exact size of the product set ðŽ â
ðŽ, where ðŽ = {1, . . . , ð}. On the other hand, if ðŽ = {ð, ðð, ðð2 , . . . ðððâ1 } is a geometric progression, then we have |ðŽ â
ðŽ| = 2ð â 1, while |ðŽ + ðŽ| = (ð+1 ). Thus in these 2 cases, if either the sum set or the product set is small, then the other is large. ErdÅs and Szemerédi5 conjectured the following: Conjecture 7.2.1 (Sum-Product Conjecture, [51]). Let ðŽ â â€, with |ðŽ| = ð. Then for all ð > 0, we have max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â«ð ð2âð . In the same paper ErdÅs and Szemerédi proved the following: Theorem 7.2.2. Let ðŽ â â€, with |ðŽ| = ð. Then there exists ð > 0 such that max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« ð1+ð . Notice that the bound max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« ð1+ð is exponentially better than the trivial bound max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« ð, but no attempt 4 Fordâs result postdates the ErdÅs-Szemerédi Problem, but ErdÅs already knew ([50]) that |ðŽ â
ðŽ| â ð2 /(log ð)ð¿+ð(1) where ð¿ = 0.086 . . . is the value appearing in Fordâs result. 5 Mel Nathanson once told me that this is really only ErdÅsâ conjecture (and not Szemerédiâs). However, due to ErdÅsâ prolific career, it is helpful to append someone elseâs name to the conjecture in order to help distinguish this conjecture of ErdÅs from his many others. For this reason and since historically this has been referred to as the ErdÅsSzemerédi Problem, we choose to retain both names.
7.2. Sum-product phenomena
139
at an explicit value of ð > 0 was made until Nathansonâs work ([115]), where he showed that every set ðŽ â †with |ðŽ| = ð must satisfy 1
max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« ð1+ 31 . Ford ([55]) refined Nathansonâs work to achieve the bound 1
max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« ð1+ 15 . Around the same time, Elekes ([41]) changed the way we think about the sum-product problem by connecting two seemingly disparate areas of mathematics in number theory and extremal geometry. Theorem 7.2.3. Let ðŽ â â with |ðŽ| = ð. Then |ðŽ + ðŽ||ðŽ â
ðŽ| â« ð5/2 . In particular, max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« ð5/4 . Proof. We may assume that 0 â ðŽ without affecting our asymptotic bounds. Consider the set of points ð = (ðŽ â
ðŽ) à (ðŽ + ðŽ) and the set of lines ð¿ = {ðð¥ + ð ⶠð â ðŽ,
1 â ðŽ} . ð
Then |ð| = |ðŽ + ðŽ||ðŽ â
ðŽ| while |ð¿| = ð2 . Note that each line is incident to ð points in ð. Hence, ðŒ(ð, ð¿) = ð3 . Thus, ð3 ⪠(|ðŽ â
ðŽ||ðŽ + ðŽ|)2/3 (ð2 )2/3 by the Szemerédi-Trotter inequality. The result follows. More than simply improving the Sum-Product results, Elekesâ short proof drastically changed our viewpoint of the sum-product problem, and efforts to improve on the sum-product conjecture ([101, 138, 141, 142]) have relied on incidence theory ever since. Note also that Elekesâ result works over any set of reals ðŽ â â, so that ðŽ need not contain only integers. Unfortunately, there is a natural boundary of |ðŽ + ðŽ||ðŽ â
ðŽ| â« ð3 , so that 3/2 is the optimal exponent we could achieve (for the sumproduct problem) with this method. It is known ([138]) that max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« |ðŽ|4/3+5/5277 for all ðŽ â â.
140
Chapter 7. Combinatorics in finite fields
Interestingly, there are numerous results that state when either the sum set or the product set is small, then the other must be large. Results of this type, specifically showing that the product set is large when the sum set is small, are sometimes called Freiman-type results, as the first such result was the work of Gregory Freiman ([57]). Theorem 7.2.4 (Freimanâs Theorem). Let ð denote a positive integer, and suppose ðŽ â â€, where |ðŽ| = ð. If ð ⥠3, and if |ðŽ+ðŽ| = 2ðâ1+ð †3ðâ4, then ðŽ is a subset of an arithmetic progression of length ð + ð. This result is highly combinatorial in nature, and improvements have long been sought by many mathematicians ([11, 133]). Nathanson and Tenenbaum ([116]) were the first to explicitly evaluate the size of the product set when the sum set is small. Theorem 7.2.5. Let ðŽ â â€, with |ðŽ| = ð. If |ðŽ + ðŽ| †3ð â 4, then |ðŽ â
ðŽ| ⥠ð2 /(log ð)2 if ð is sufficiently large. A quantitive version of the âsmall sum set, large product setâ phenomenon was also given by Elekes and Ruzsa ([42]). Theorem 7.2.6. Let ðŽ â â, where |ðŽ| = ð ⥠2. Then, |ðŽ + ðŽ|4 |ðŽ â
ðŽ| â«
ð6 . log ð
In particular, if |ðŽ + ðŽ| ⪠ð, then |ðŽ â
ðŽ| â«
ð2 . log ð
Note that Theorem 7.2.6 is sharp in the sense that ð6 cannot be replaced by any smaller power of ð. Nonetheless, their result only gives an inferior sum-product estimate of max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« ð6/5âð for any ð > 0. On the other hand, Elekesâ Theorem (Theorem 7.2.3) was not a sharp inequality, and yet it yielded a superior sum-product estimate. The âsmall product set implies large sum setâ results are more difficult. Nonetheless Chang ([16]) gave the following result: Theorem 7.2.7. Let ðŽ â â€, with |ðŽ| = ð. If |ðŽ â
ðŽ| < ðŒð, then |ðŽ + ðŽ| > 36âðŒ ð2 . Before we explore some questions related to the Sum-Product Problem, we first review some of the results achieved and variants considered
7.2. Sum-product phenomena
141
towards a better understanding of the sum-product phenomena. For example the following was achieved by Iosevich, Roche-Newton, and Rudnev ([88]). The following relied heavily on the Guth-Katz distance theorem (see (2.4)): Theorem 7.2.8. Let ðŽ â â with |ðŽ| = ð. Then, |ðŽ â
ðŽ + ðŽ â
ðŽ + ðŽ â
ðŽ| â« ð2 / log ð.
7.2.1 Sum-product phenomena in rings It is time to delve into other analogues of the problem. Indeed, the sum set ðŽ + ðŽ and product set ðŽ â
ðŽ are well defined when ðŽ â ð
, where ð
is any ring. Naturally, we will primarily consider the case ð
= ðœð , where ð = ðâ for an odd prime ð. There are numerous idiosyncrasies for the sum-product problem over finite fields. For example, if ðŽ â ðœð is a subring, then the sum set and product set satisfy ðŽ + ðŽ = ðŽ â
ðŽ = ðŽ. Thus we must not allow ðŽ to be âtoo closeâ to being a subring of ð
. For subsets of finite rings ðŽ â ð
, we have the trivial bounds |ðŽ| †max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) †min(|ð
|, (
|ðŽ| + 1 )) . 2
Fortunately, there are many methods available to us to ensure that ðŽ is not close to being a subring, including cardinality conditions (supposing ð1/2 ⪠|ðŽ| ⪠ð1âð for some ð > 0, for instance). The first pure sum-product result in finite fields was work of Bourgain, Katz, and Tao ([14]). Theorem 7.2.9. Let ðŽ â ðœð for a prime ð, where |ðŽ| ⪠ð1âð for some ð > 0. Then there exists ð¿ = ð¿(ð) > 0 such that max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« |ðŽ|1+ð¿ . Note 7.2.10. Theorem 7.2.9 does not establish an explicit relationship between ð and ð¿. Also, the condition |ðŽ| ⪠ð1âð was sufficient in this case as the prime field ðœð has only the trivial subrings: itself and {0}. The original statement of Theorem 7.2.9 also assumed that |ðŽ| â« ðð , though this condition was later shown ([13]) to be unnecessary.
142
Chapter 7. Combinatorics in finite fields
Hart, Iosevich, and Solymosi ([75]) established the first explicit sumproduct bound. Theorem 7.2.11. Let ðŽ â ðœð with ð1/2 ⪠|ðŽ| ⪠ð7/10 . Then, max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« ðâ1/4 |ðŽ|3/2 . In particular if |ðŽ| â ð7/10 , then max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« |ðŽ|8/7 . Additionally, if |ðŽ| â« ð3/4 and |ðŽ + ðŽ| ⪠|ðŽ|, then |ðŽ â
ðŽ| â« ð. The proof of Theorem 7.2.11 essentially amounts to using the following incidence bound for hyperbolas. Theorem 7.2.12 ([75]). Let ðž, ð¹ â ðœ2ð , and ð â ðœâð . Then |{(ð¥, ðŠ) â ðž à ð¹ ⶠ(ð¥1 â ðŠ1 )(ð¥2 â ðŠ2 ) = ð}| ⪠ðâ1 |ðž||ð¹| + ð1/2 â|ðž||ð¹|. Additional results exist for small subsets of finite fields. Theorem 7.2.13 (Garaev, [59]). Let ðŽ â ðœð , where ð is prime. Then for sets ðŽ with |ðŽ| < ð7/13 /(log ð)4/13 , max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« |ðŽ|15/14+ð(1) . Garaevâs inequality has been improved many times ([12, 95, 96]) with the current best result belonging to Roche-Newton, Rudnev, and Shkredov ([127]). Theorem 7.2.14. Let ðŽ â ðœð , where ð = ðâ for a prime ð. Then we have max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« |ðŽ|6/5 , so long as |ðŽ| < ð5/8 . A similar result holds if ðŽ + ðŽ is replaced by ðŽ â ðŽ. In arbitrary finite fields, the strongest known result is the following result by Jones and Roche-Newton ([92]). Theorem 7.2.15. Let ðŽ â ðœð , where ðŽ is such that |ðŽ â© ððº| †|ðº|1/2 for all subfields ðº â ðœð and for all ð â ðœð . Then max(|ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« |ðŽ|12/11âð(1) . Note 7.2.16. Again the condition that |ðŽâ©ððº| †|ðº|1/2 in Theorem 7.2.15 ensures that ðŽ is sufficiently far from being a subring.
7.2. Sum-product phenomena
143
Another such result concerns the set ðŽ(ðŽ + 1) = {ð(ð + 1) ⶠð, ð â ðŽ}. Theorem 7.2.17 ([92]). Let ðŽ â ðœð , where ð is prime. Then for sets ðŽ with |ðŽ| < ð1/2 , |ðŽ(ðŽ + 1)| â« |ðŽ|57/56âð(1) . This result was an improvement on [60], and both papers also included results for sets of real numbers ðŽ â â. Theorem 7.2.18 ([127]). Let ðŽ â ðœð with ð = ðâ for a prime ð. For sets ðŽ with |ðŽ| ⪠ð18/35 , max(|ðŽ + ðŽ + ðŽ|, |ðŽ â
ðŽ|) â« |ðŽ|16/13 . We mention one more analogue of the sum-product phenomena in finite fields. We define ðŽ â
ðŽ + ðŽ â
ðŽ = {ðð + ðð ⶠð, ð, ð, ð â ðŽ}. It is convenient to denote ðŽ + ðŽ = ðŽ2 , and ðŽ + ðŽ = 2ðŽ. Thus we would write ðŽ â
ðŽ + ðŽ â
ðŽ = 2ðŽ2 . Likewise we define ðâµ ððŽð = ðŽ +âµââµ â¯+ ðŽð , ââµ âµâµâ ðâtimes
where ðŽð = {ð1 . . . ðð ⶠð1 , . . . , ðð â ðŽ} is the ð-fold product set. In this context the sum-product problem can then be framed similarly to the ErdÅs-Falconer distance problem. Problem 7.2.19 (Dot-Product Problem). Let ðŽ â ðœð . Find the minimum exponent ðŒ â (0, 1) such that we are guaranteed |ððŽð | â« ð whenever |ðŽ| â« ððŒ . In general ðŒ > 1/2 as otherwise ðŽ could be a subring of ðœð . 1
1
Theorem 7.2.20. ([74]) Let ðŽ â ðœð with |ðŽ| > ð 2 + 2ð . Then ðœâð â ððŽ2 .
144
Chapter 7. Combinatorics in finite fields
Proof. For ðž â ðœðð , we define Î (ðž) = {ð¥ â
ðŠ ⶠð¥, ðŠ â ðž}, where ð¥ â
ðŠ = ð¥1 ðŠ1 + ⯠+ ð¥ð ðŠ ð denotes the standard inner product. Let ð(ð) = |{(ð¥, ðŠ) â ðž à ðž ⶠð¥ â
ðŠ = ð}|. We will show that ð â Î (ðž) for all ð â ðœâð so long as ðž is sufficiently large, which is more than what we need to show. We will make the connection between these results at the end of the proof. As usual, we use the exponential sums method:
ð(ð) = ðâ1 â â ð(ð â
(ð¥ â
ðŠ â ð)) ð âðœð ð¥,ðŠâðž
=ð
â1
â â ð(ð (ð¥ â
ðŠ â ð)) + ðâ1 â â ð(ð (ð¥ â
ðŠ â ð)) ð â 0 ð¥,ðŠâðž
ð =0 ð¥,ðŠâðž â1
2
= ð |ðž| + ð
ð . Note that by rearranging our sums and applying the triangle inequality, we have | | ð
ð = ðâ1 â â â ð(ð (ð¥ â
ðŠ â ð)) †ðâ1 â || â â ð(ð (ð¥ â
ðŠ â ð))|| . | ð¥âðž ðŠâðž ð â 0 ð¥âðž |ðŠâðž ð â 0 Squaring ð
ð and applying the Cauchy-Schwarz inequality yields 2
|ð
ð | †ð
â2
| | ( â 1 â
|| â â ð(ð (ð¥ â
ðŠ â ð))||) | |ðŠâðž ð â 0 ð¥âðž
â€ð
â2
| | ( â 1 ) ( â || â â ð(ð (ð¥ â
ðŠ â ð))|| ) | ð¥âðž ð¥âðž |ðŠâðž ð â 0
2
2
2
2
| | †ð |ðž| â || â â ð(ð (ð¥ â
ðŠ â ð))|| . |ðŠâðž ð â 0 | ð¥âðœð ð â2
7.2. Sum-product phenomena
145
Using only the fact that a complex number ð§ = ð¥ + ððŠ satisfies |ð§| = ð§ð§ = (ð¥ + ððŠ)(ð¥ â ððŠ), we see that 2
| | â || â â ð(ð (ð¥ â
ðŠ â ð))|| |ðŠâðž ð â 0 | ð¥âðœð ð = â
â
â ð(ð¥ â
(ð ðŠ â ð â² ðŠâ² ))ð(ð(ð â² â ð ))
ð ,ð â² â 0 ðŠ,ðŠâ² âðž ð¥âðœð ð
= â ð(ð(ð â² â ð )), ð ,ð â² â 0 ðŠ,ðŠâðž ð ðŠ=ð â² ðŠâ²
where we regrouped and then applied orthogonality. Note that ð , ð â² â ðœâð . Thus we substitute ð = ð /ð â² and ð = ð â² to see that |ð
ð |2 †ððâ2 |ðž| â ð(ðð(1 â ð)) â ðŒ + ðŒðŒ, ð,ðâðœâð ðŠ,ðŠâ² âðž ððŠ=ðŠâ²
where ðŒ = ððâ2 |ðž| â 1 = |ðž|2 ððâ2 (ð â 1) †|ðž|2 ððâ1 , ðâðœâð ðŠ,ðŠâ² âðž ðŠ=ðŠâ²
and where ðŒðŒ = ððâ2 |ðž|
ð(ðð(1 â ð)) â ððâ2 |ðž| â 1
â
ðâ 0,1 ðŠ,ðŠâ² âðž ððŠ=ðŠâ²
ðâðœð ;ðâ 0,1 ðŠ,ðŠâ² âðž ððŠ=ðŠâ²
= âððâ2 |ðž| â 1. ðâ 0,1 ðŠ,ðŠâ² âðž ððŠ=ðŠâ²
Hence ðŒðŒ < 0. This shows that |ð
ð |2 †ðŒ + ðŒðŒ †ðŒ †ððâ1 |ðž|2 , and hence ð(ð) = ðâ1 |ðž|2 + ð
ð , where |ð
ð | †ð
ðâ1 2
|ðž|.
146
Chapter 7. Combinatorics in finite fields ð+1
Therefore, ð(ð) > 0 (ensuring ð â ðð(ðž)) whenever |ðž| > ð 2 . In particular if ðž = ðŽ à ⯠à ðŽ is the ð-fold Cartesian product of a set ðŽ â ðœð , then Î (ðž) = ððŽ2 . It follows that if |ðŽ| > ð
ð+1 2ð
, then ðœâð â ððŽ2 .
In [26], Theorem 7.2.20 was extended to the case â€ð . Theorem 7.2.21. Let ð = ðâ for a prime ð, and suppose ðŽ â â€ð satisfies |ðŽ| > ð
(2ââ1)ð+1 2âð
. Then, â€Ãð â ððŽ2 .
As before, the proof of Theorem 7.2.21 mimicked the proof of Theorem 7.2.20 in spirit, though some of the technical difficulties were greater.
7.3 Kakeya conjecture Another problem in the âbig sets must be interestingâ category is the Kakeya Conjecture. A Kakeya-Besicovith set (which for convience we will call simply a Kakeya set) is a compact set ðž â â2 containing a line segment of length 1 in every direction. Of course the first such set that springs to mind is a circle with diameter 1. Is there another set with a smaller area? It turns out that an equilateral triangle with sidelength 1 is also a Kakeya set, and it has a smaller area. A deltoid of diameter 1 has area smaller still. How small can the area be of such a Kakeya set?
Theorem 7.3.1 (Besicovitch, [8]). There exist Kakeya sets with measure 06 . 6 See
Section 2.2 for what it means for a set to have measure 0.
7.3. Kakeya conjecture
147
This is far from the end of the story as there is much more to consider. What happens in higher dimensions? Do there exist Kakeya sets in âð of measure zero? What do they look like? The so-called Kakeya conjecture is simple to state, given that we remember the Hausdorff dimension: Conjecture 7.3.2. Let ðž â âð be a Kakeya set. Then dimð» (ðž) = ð. Again, one should not worry too much about the notion of Hausdorff dimension if this concept is foreign to you. Simply put, the Kakeya conjecture asserts that Kakeya sets must all be large in terms of their (Hausdorff) dimension. Roy Davies ([32]) showed that a Kakeya set ðž â â2 satisfies dimð» (ðž) = 2, but the conjecture is unsolved in dimensions ð ⥠3. The best known bounds for Kakeya sets in higher dimensions involve the Minkowski dimension, which is closely related to the Hausdorff dimension, but it is much easier to define. For a compact set ðž â âð , let ð ð¿ (ðž) denote the minimum number of ð¿-balls7 needed to cover ðž. Then the upper Minkowski dimension is log(ð ð¿ (ðž)) dimð (ðž) = lim sup , log(1/ð¿) + ð¿â0 while the lower Minkowski dimension is log(ð ð¿ (ðž)) , dimð (ðž) = lim inf + log(1/ð¿) ð¿â0 and thus the Minkowski dimension is dimð (ðž) = lim+ ð¿â0
log(ð ð¿ (ðž)) , log(1/ð¿)
provided the limit exists. Note 7.3.3. There are many sets for which the Hausdorff dimension and the Minkowski dimension are equal, but in general one has the inequality 0 †dimð» (ðž) †dimð (ðž) †dimð (ðž) †ð 7A
ð¿-ball is simply a ball with radius ð¿.
148
Chapter 7. Combinatorics in finite fields
for all sets ðž â âð . Thus the condition that dimð» ðž = ð is stronger than the condition dimð ðž = ð. The best known bounds for the Minkowski dimension of a Kakeya set in âð are as follows ([94, 97, 103]). Theorem 7.3.4. Let ðž â âð be a Kakeya set. Then, 5/2 + 10â10 ⧠⪠⪠3 + 10â10 ⪠dimð (ðž) â¥
⚠⪠(2 â â2)(ð â 4) + 3 ⪠⪠ð+ðâ1 â© ð
ð=3 ð=4 5 †ð †23 ð ⥠24.
Here, ð = 1.675 . . . is the root of ð¥3 = 4ð¥ â 2 with ð > 1. The following bounds for the Hausdorff dimension of Kakeya sets are also known ([97, 158]). Theorem 7.3.5. Let ðž â âð be a Kakeya set. Then dimð» (ðž) ⥠{
ð+2 2
(2 â â2)(ð â 4) + 3
ð = 3, 4 ð ⥠5.
The proofs of these results are very technical in nature (notice that we havenât even defined the Hausdorff dimension!), and so we will not give the proofs here. The interested reader should consult the original sources for such proofs. However, Tom Wolff ([157]) devised a finite field analogue of the Kakeya conjecture, and we will explore that analogue in detail below. Even in finite fields, a Kakeya set refers to a set containing a line in every direction. We first recall some basic facts about lines in finite fields, and we define directions in finite fields. Recall a line in ðœðð is a set of the form â = {ð + ð¡ð ⶠð¡ â ðœð } â ðœðð , where ð, ð â ðœðð and ð â 0.â Notice that every line contains precisely ð points. Now, we say two lines â1 and â2 in ðœðð are in the same direction
7.3. Kakeya conjecture
149
if there exists a vector ð¥ â ðœðð such that â1 + ð¥ = {ð + ð¥ + ð¡ð ⶠð¡ â ðœð } satisfies â1 + ð¥ = â2 . âBeing in the same directionâ turns out to be an equivalence relation on the set of all lines. Directions are the resulting set of equivalence classes. Moreover, the number of directions in ðœðð is given by ðð â 1 , ðâ1 since given any point, there are ðð â 1 other points in ðœðð through which we can draw a line, and there are exactly ð points on a line (which means there are ð â 1 other points on a line if we exclude our starting point). Thus, there are exactly ð + 1 directions in ðœ2ð and approximately ððâ1 directions in ðœðð . We are now ready to state the Finite Field Kakeya Conjecture. Recall that the notion of dimension in âð can be replaced by cardinality in ðœðð . Loosely speaking, if ðž â âð has some algebraic or geometric property whenever dimð» (ðž) > ðŒ, then one should expect8 the same for sets ðž â ðœðð with |ðž| â« ððŒ . Conjecture 7.3.6. Let ðž â ðœðð be a set containing a line in every direction. â there More precisely, suppose that ðž â ðœðð is such that for all ð â ðœðð ⧵ {0}, ð ð exists ð â ðœð such that ð + ð¡ð â ðž for all ð¡ â ðœð . Then |ðž| â«ð ð . This conjecture has been resolved in the affirmative ([40]), but before we discuss this result, it is instructive to consider Wolffâs results. Theorem 7.3.7 (Wolff, [158]). Let ðž â ðœðð be a Kakeya set. Then |ðž| â«ð ð In particular a Kakeya set ðž â
ðœ2ð
ð+2 2
.
satisfies |ðž| â« ð2 .
Proof. We only prove the result in the case that ð = 2, and in fact, we prove a strengthened version of the result: Lemma 7.3.8. Let ðž â ðœ2ð satisfy the condition that ðž contains at least ð/2 points on a line in a set of lines having ð different directions. Then |ðž| â« ðð. 8 This is a vast overgeneralization, but the number of instances of this duality transcends mere coincidence.
150
Chapter 7. Combinatorics in finite fields
We first prove the lemma. Let {â1 , . . . , âð } be the set of ð lines. Since |ðž â© â| ⥠ð/2 for all lines â, it follows that ð
1 ðð †â |ðž â© âð |. 2 ð=1 Squaring our previous inequality, we see that 2
ð 2
(ðð) ⪠( â |ðž â© âð |) ð=1 ð
ð
= â â |ðž â© âð ||ðž â© âð | ð=1 ð=1 ð
ð
= â â â ðâð (ð¥)ðâð (ð¥) ð¥âðž ð=1 ð=1 ð
ð
†|ðž| â â |âð â© âð |, ð=1 ð=1
where ðâ is the characteristic function of the line â so that ðâ (ð¥) = 1 if ð¥ â â and 0 otherwise. Using that two lines intersect in at most 1 point, we see that ð
(ðð)2 ⪠|ðž| ( â |âð | + â |âð â© âð |) ð=1
ðâ ð
†|ðž|(ðð + ð(ð â 1)) from which the Lemma follows, since there are ð = ð + 1 directions when ð = 2. Thus we have shown that |ðž| â« ð2 for Kakeya sets ðž â ðœ2ð , and this proves Theorem 7.3.7 when ð = 2. Wolffâs bounds were steadily improved upon by numerous authors ([14, 111, 128, 149]), whose proofs were quite difficult and typically involved some harmonic analysis, additive combinatorics, and even some algebraic geometry. It was widely believed that the finite field Kakeya problem was nearly as difficult as the Euclidean version until Zeev Dvir shocked the harmonic analysis world with a full (algebraic) proof of the
7.3. Kakeya conjecture
151
finite field Kakeya conjecture, whose difficulty even some high school students9 would understand. Theorem 7.3.9 (Dvir, [40]). Let ðž â ðœðð be a Kakeya set. Then |ðž| â«ð ðð . Proof. We need the following two Propositions10 . Proposition 7.3.10 (Paramater Counting Argument). Suppose that ðž â ðœðð . Then there exists a nonzero polynomial ð in ð variables such that ð(ð¥) = 0 for all ð¥ â ðž and such that deg(ð) †ð|ðž|1/ð . Proposition 7.3.11 (Vanishing Proposition). Let ð ð·,ð be the vector space of polynomials over ðœð in ð variables with degree at most ð·. If ð â ð ð·,ð vanishes on a set ðž where |ðž| ⥠ð· + 1, then ð â¡ 0 is the zero polynomial. Note 7.3.12. Analogues of these two preceding Propositions hold over any field, though we have only stated them for finite fields. Suppose that ðž â ðœðð is a Kakeya set and that |ðž| †(10ð)âð ðð . By Proposition 7.3.10 there must exist a nonzero polynomial ð in ð variables that vanishes on ðž and has degree ð· â deg(ð) †ð|ðž|1/ð < ð. We write ð = ðð· + ð
, where ðð· is nonzero and homogeneous of degree ð· and where deg(ð
) < ð·. Let ð â ðœðð ⧵ {(0, . . . , 0)}, and choose ð â ðœðð ⧵ {(0, . . . , 0)} so that {ð + ð¡ð ⶠð¡ â ðœð } â ðž. Then the polynomial ð(ð¡) = ð(ðð¡ + ð) = 0 for all ð¡ â ðœð . Note that deg(ð) †ð· < ð. By Proposition 7.3.11, ð must be the zero polynomial, so that every coefficient of ð must be zero. But the ð¡ð· coefficient in ð is nothing else but ðð· (ð), and hence ðð· (ð) = 0 for all ð â ðœðð ⧵ {(0, . . . , 0)}. Of course ðð· (0, . . . , 0) = 0, since ðð· is homogeneous. Thus, ðð· also vanishes at every point in ðœðð . Since ð· < ð, ðð· is the zero polynomial, which is a contradiction. Thus, we must have |ðž| ⥠(10ð)âð ðð for Kakeya sets ðž â ðœðð . 9 At a conference I once attended, Nets Katz was giving a detailed talk of his work on the âJointsâ problem (this was the joint work with Larry Guth that eventually lead to the solution of the ErdÅs distance problem), describing most of the steps of the proof as âhigh school mathematicsâ. Halfway through the lecture, another mathematician stood up and shouted, âWhere the hell did you go to high school?â 10 The following proof appears in Larry Guthâs excellent book Polynomial Methods in Combinatorics ([68]).
152
Chapter 7. Combinatorics in finite fields
And the story in finite fields is done! Despite being a short proof, the proof remains mysterious to me (and I hope to you too). Why did the proof work? Does any modification of the proof apply to the Euclidean Kakeya Conjecture? What is it about the Kakeya problem that lent itself to such an easy solution? What other combinatorial problems can be solved with the polynomial method? What happens in â€ðð ? Note that lines in â€ðð are defined similarly to the finite field case: for ð, ð â â€ðð (and ð â (0, . . . , 0)) we define â = {ð + ðð¡ ⶠð¡ â â€ð }. Note that lines need not have ð points anymore (such as when gcd(ð, ð) > 1), so this problem is even more interesting (and difficult)! There has been some progress on the analogue for the Kakeya problem for the modules over the ring â€ð , at least in the case ð = 2 when ð is the power of a prime. Theorem 7.3.13 (Dummit and Hablicsek, [39]). Let ð be a prime and suppose that ðž â â€ð2ð is a Kakeya set. Then |ðž| â¥
ð2ð . 2ð
There are various other results in the area (see [45], for example), but this area is ripe for further exploration.
7.4 Waringâs theorem The classical Waring problem, dating back to the late 18th century, asks one to find the minimum number ð so that for a fixed ð ⥠2, every nonnegative integer ð can be written as a sum of ð ðth powers. We use ð(ð) to denote the minimum such value of ð. Lagrange showed that ð(2) †4, which together with the trivial observation that 7 cannot be written as a sum of 3 squares, shows that ð(2) = 4. That ð(ð) exists for all ð ⥠2 was shown by Hilbert ([80]), and an (almost) exact evaluation of ð(ð) is known (see [153] for details): It is known that ð
ð(ð) = 2ð + â(3/2) â â 2 so long as ð
2ð {(3/2)ð } + â(3/2) â †2ð ,
7.4. Waringâs theorem
153
and ð
ð
ð(ð) = 2ð + â(3/2) â + â(4/3) â â ðŽ otherwise, where ðŽ is either 2 or 3 depending on whether or not ð satisfies a certain inequality. Here ââ
â is the usual floor function and {ð¥} = ð¥ â âð¥â is the fractional part of ð¥. The modern formulation of Waringâs problem is to fix an integer ð ⥠2 and determine the minimal integer ð such that every sufficiently large integer can be written as the sum of ð ðth powers. As is customary, we use ðº(ð) to denote the minimal such integer. Only two values of the function are known precisely: ðº(2) = 4 is a corollary of Lagrangeâs four squares theorem (and the trivial fact that, if ð â¡ 7 (mod 8), then ð is not the sum of three squares), and ðº(4) = 16 is work of Davenport ([31]). We know only that ðº(3) â {4, 5, 6, 7}, for example ([30]). Finding upper bounds for ðº(ð) for certain values of ð and asymptotic bounds are areas of active research. The best known general asymptotic bound for ðº(ð) is work of Wooley ([159]): ðº(ð) †ð(log ð + log log ð + 2 + ð(log log ð/ log ð)). Let ðº 1 (ð) be the smallest integer ð such that almost all11 integers ð are the sum of ð ðth powers. Even with the weakening of the definition the precise values of ðº 1 (ð) are only known for six values of ð. Of course Lagrangeâs four squares theorem and the observation that if ð¥ â¡ 7 (mod 8), then ð¥ is not the sum of three squares implies that ðº 1 (2) = 4. The other known values of the function ðº 1 are ðº 1 (3) = 4, ðº 1 (4) = 15, ðº 1 (8) = 32, ðº 1 (16) = 64, and ðº 1 (32) = 128 ([153]). If ð is a subset of the ring and ð is a positive integer, we let Î(ð, ð) denote the smallest integer ð such that for all ð â ð, there exists a solution {ð1 , . . . , ðð } to the equation ð ð = ð1ð + ⯠+ ðð .
11 More
precisely if ðž â â is the exceptional set, then lim
ðââ
|ðž â© {1, . . . , ð}| = 0. ð
(7.2)
154
Chapter 7. Combinatorics in finite fields
In the case that no such minimum exists, then we put Î(ð, ð) = +â. Thus, the classical version of Waringâs problem is determining ð(ð) = Î(ð, â€+ ). A natural case to consider beyond the above versions of Waringâs problem is when ð = â€ð , the ring of integers mod ð. For example we know ([139]) that Î(2, â€ð ) †2 if and only if ð has the property that, if a prime ð satisfies ð2 ⣠ð, then ð â¡ 1 (mod 4). Furthermore, we have Î(2, â€ð ) †3 if and only if 8 †ð, and of course Î(2, â€ð ) †4 for all ð by Lagrangeâs four squares theorem. We further set Î(ð) = max Î(ð, â€ð ). â€ð
Note that this quantity Î(ð) exists for each ð (Exercise 7.9). Of course we have Î(2) = 4, but cubes are more interesting as it turns out that the following holds: Proposition 7.4.1. Î(3, â€ð ) †2 if and only if 7 †ð and 9 †ð, and Î(3, â€ð ) †3 if and only if 9 †ð. Finally Î(3) = 4. The first to calculate Î(ð) for small values of ð were Hardy and Littlewood ([72]), who calculated Î(ð) for 2 †ð †100, with a followup and some corrections from Sekigawa-Koyama ([137]) who calculated Î(ð) for 2 †ð †200. For example it turns out that Î(4) = 15, Î(5) = 5, and Î(6) = 9. Much remains unknown about Î(ð) despite it being evaluated for small values of ð. See [46] for an evaluation of (a slight variation of) Î(ð) and some related problems. The main tools used in this context are the Chinese Remainder Theorem (Exercise 5.6) and an elementary version of Henselâs Lemma: Lemma 7.4.2 (Henselâs Lemma). Let ð(ð¥) be a polynomial with integer coefficients, let ð be prime, and let ð be a positive integer. If ð â †satisfies ð(ð) â¡ 0 (mod ðð ) and if ðâ² (ð) ⢠0 (mod ð), then there exists ð â †such that ð(ð) â¡ 0 (mod ðð+1 ) and ð â¡ ð (mod ð). More robust and technical (and more useful) versions of Henselâs Lemma exist (see [126], for example).
7.5. Rothâs theorem and the cap-set problem
155
7.4.1 Waringâs problem for finite fields The finite field version of the problem is much the same as that of the ring of integers modulo ð, except that Î(ð, ðœð ) need not always exist. For example, the cubes in ðœ4 are {0, 1}, so that every element outside of the subfield {0, 1} cannot be written as a sum of any number of cubes. Thus Î(3, ðœ4 ) = +â. We start by noting that Î(ð, ðœðð ) exists if and only if for all ð ⣠ð where ðð â1
ð â ð, then ðð â1 †ð ([9]), and we have already seen that Î(ð, ðœð ) exists for all primes ð. We summarize some of the basic known results in the following theorem. Theorem 7.4.3. Let ð = ðð , and assume that ð and ð are such that Î(ð, ðœð ) exists. Put ð = gcd(ð, ð â 1). Then: Î(ð, ðœð ) = Î(ð, ðœð )
(7.3)
Î(ð, ðœð ) †ð
(7.4)
Î(ð, ðœð ) = 1 for all ðœð ⺠ð = 1
(7.5)
Î(2, ðœð ) = 2
(7.6)
Î(ð, ðœð ) †ð Î(ð, ðœð ) ⪠ð
if ð > ð 1/2
2ð ð â1
.
(7.7) (7.8)
The result Î(2, ðœð ) = 2 follows from the pigeonhole principle. The bound Î(ð, ðœð ) ⪠ð1/2 is known as the second Heilbronn conjecture, and it was established by Cipra, Cochrane, and Pinner ([21]) in 2006. Analogous to what we did for the ring of integers mod ð, set Îâ (ð) = max Î(ð, ðœð ) where the maximum is taken over all prime powers ð such that Î(ð, ðœð ) exists. For example it is known ([33]) that Îâ (4) = 5, Îâ (5) = 5, and Îâ (6) = 6. What happens to Îâ (ð) as ð â â?
7.5 Rothâs theorem and the cap-set problem Often cited as one of Martin Gardnerâs main influences, Henry Dudeney was a mathematician and author who specialized in creating mathematical games. My favorite of his games is his aptly titled âno-three-in-lineâ
156
Chapter 7. Combinatorics in finite fields
game: Find the maximum number of points, ð(ð), on an ð Ã ð grid, so that no three points are collinear.
Figure 7.3. A set of 20 points on a 10 Ã 10 grid so that no line drawn contains any 3 points.
Note that Figure 7.3 shows that ð(10) ⥠20. The pigeonhole principle implies that ð(ð) †2ð in general (see Exercise 7.15). The bound ð(ð) = 2ð has been achieved for some small values of ð, but it has been ðð conjectured ([119]) that ð(ð) ⪠+ ð(ð) in general. This problem is â3
surprisingly difficult even for some small values of ð, and the reader is highly encouraged to find a solution for a 9 Ã 9 grid. One of the earliest and best-known problems in Ramsey theory is the following. Consider any 9 numbers, say {1, . . . , 9}. Now assign to each element in {1, . . . , 9} one of two colors, say red and blue. A natural question which arises is whether such a coloring must necessarily contain a 3-term arithmetic progression. For example: 1ðµ 2ð
3ð
4ðµ 5ðµ 6ð
7ð
8 ðµ 9ð
ⶠ1ðµ 2ð
3ð
4ðµ 5ðµ 6ð
7ð
8 ðµ 9ð
7.5. Rothâs theorem and the cap-set problem
157
and 1ðµ 2ðµ 3ð
4ð
5ðµ 6ðµ 7ð
8 ðµ 9ð
ⶠ1ðµ 2ðµ 3ð
4ð
5ðµ 6ðµ 7ð
8 ðµ 9ð
. Is there a way to color these 9 integers using only blue or red so that no three-term arithmetic progression has the same color? It turns out there is no such coloring, though in order to prove this, you must show that every coloring of {1, . . . , , 9} (of which there are 512 possible colorings) cannot be so colored. Indeed there is nothing special about having 9 integers, 2 colors, or even that the arithmetic progression has 3-terms. In 1927, Dutch mathematician Bartel Leendert van der Waerden ([152]) proved the following general result. Theorem 7.5.1 (van der Waerdenâs Theorem). Fix integers ð ⥠3 and ð ⥠2. Then there exists a constant ð(ð, ð) such that for ð ⥠ð(ð, ð) if the set {1, . . . , ð} is ð-colored in the sense that each element in {1, . . . , ð} corresponds to a color {ð¶1 , . . . , ð¶ð }, then there must exist at least one subset of {1, . . . , ð} containing an arithmetic progression of length ð, where all elements in the arithmetic progression have the same color. In our example above, we have ð(2, 3) = 9. The constant ð(ð, ð) is called van der Waerdenâs constant in his honor. Of course lower bounds on ð(ð, ð) can be exhibited by examples, but upper bounds are much more difficult as you have to show that any sufficiently large set of integers cannot be colored using fewer colors. For general ð and ð, Gowers ([63]) gave the terrifying-looking bound ð2
ð(ð, ð) †22
2ð+9
.
It is remarkable that any finite bound can be achieved for ð(ð, ð), though 101233
Gowersâ bound only shows that ð(2, 3) †1010 . Yikes! Ron Graham has conjectured ([67]) that we should have the more modest bound 2
ð(2, ð) †2ð . A problem closely related to van der Waerdenâs theorem is Rothâs theorem. Consider any subset of integers ðŽ â {1, 2, 3, . . . } with positive upper density in the sense that lim sup ðââ
|ðŽ â© [1, ð]| > 0. ð
158
Chapter 7. Combinatorics in finite fields
Does ðŽ necessarily contain 3-term arithmetic progressions? What about 4-term arithmetic progressions, or even ð-term arithmetic progressions? Roth showed this must be the case for ð = 3 ([129]), and Szemerédi gave an affirmative answer for every ð ⥠4 in his groundbreaking work [147]. For convenience we state these results together in the following theorem. Theorem 7.5.2 (Roth-Szemerédi). Let ðŽ â †have positive upper density. Then ðŽ necessarily contains a ð-term arithmetic progression for any ð ⥠3. Szemerédiâs Theorem is a very deep and difficult result. Famously, ErdÅs called Szemerédiâs proof a âmasterpiece of combinatorial reasoning.â There are many different proofs of Szemerédiâs Theorem (at least 10 different proofs)12 . Not only is Szemerédiâs Theorem important, but even some proofs of Szemerédiâs Theorem are important in their own right (in particular, Gowersâ proof ([63])). Interestingly, Szemerédiâs Theorem has been extended to the primes by Green and Tao. Theorem 7.5.3 (Green-Tao, [65]). Let ð« = {2, 3, 5, . . . } be the primes, and as usual let ð(ð) = |ð«â©{1, . . . , ð}| be the prime counting function. If ðŽ â ð« has positive upper density in the primes in the sense that lim sup ðââ
|ðŽ â© {1, . . . , ð}| >0 ð(ð)
then ðŽ contains some ð-term arithmetic progression for each ð ⥠3. Notice that the primes ð« have zero density in the positive integers, and yet the primes still contain arbitrarily long arithmetic progressions. Improved bounds for Rothâs Theorem have long been sought, with the best known bounds belonging to Thomas Bloom ([11]).
12 The shortest proof of Szemerédiâs Theorem is six pages and belongs to Terry Tao and Ben Green ([65]), though as Terry notes on his blog âTechnically, one might call this the shortest proof of Szemerédiâs theorem in the literature (and would be something like the sixteenth such genuinely distinct proof, by our count), but that would be cheating quite a bit, primarily due to the fact that it assumes the inverse conjecture for the Gowers norm, our current proof of which is checking in at about 100 pages. . . â ([150]).
7.5. Rothâs theorem and the cap-set problem
159
Theorem 7.5.4. Let ðŽ â {1, . . . , ð} be a set not containing any 3-term arithmetic progression. Then, |ðŽ| âª
ð (log log ð)4 . log ð
It has been said ([64]) that it is âby no means unreasonable to think that some of the methods introduced in [Bloomâs] paper could be useful in proving a bound |ðŽ| ⪠ð(log ð)â1âð which would be strong enough to conclude the first nontrivial case ð = 3 1 of ErdÅsâ celebrated conjecture that, if ðŽ â â is a set with âðâðŽ ð = â, then ðŽ contains infinitely many arithmetic progressions of length ðâ.13 We should note that the Green-Tao Theorem is the first theorem exhibiting a set of zero density (so that Szemérediâs Theorem does not apply) whose sum of reciprocals diverges and which contains arbitrarily long arithmetic progressions. We return to Dudeneyâs No-Three-In-Line Problem from the start of the section. How would the results change if instead of considering a 10 à 10 grid, we worked over â€210 ? This would mean identifying the left and right edges and identifying the top and bottom edges, so that we are now playing the game not on a plane, but on a torus. This version of Dudeneyâs problem has actually been well studied and is called the cap-set problem. Here a âcapâ refers to a 3-term arithmetic progression meaning a set of the forn {ð¥, ðŠ, ð¥ + ðŠ}. Problem 7.5.5 (Cap-Set Problem). What is the largest subset ðŽ â ðœð3 so that ðŽ does not contain any 3-term arithmetic progressions? Just like the ErdÅs-distance problem, it is too much to ask for a precise bound on such sets ðŽ for varying ð, but we would like to better understand the asymptotics. There is nothing special about ðœ3 , and indeed we can consider this question over any finite field ðœð or even any ring â€ð . 13 Excitingly, Thomas Bloom and Olof Sisack have recently submitted a paper claiming to give this bound, establishing this remarkable conjecture of ErdÅs for three-term arithmetic progressions!
160
Chapter 7. Combinatorics in finite fields
If you want to get an intuitive handle on such a problem, I suggest playing a card game called Set®. In this game the deck consists of 81 cards, each card having one of three colors, one of three shapes, one of three numbers, and one of three shadings. One must form a âsetâ which consists of three cards all with three of the four same identical properties, or each characteristic being different. Thus, â1 red filled diamond, 2 red filled diamonds, and 3 red filled diamondsâ form a set as they all share the same shape, color, and shading. Likewise, â1 red filled diamond, 2 green shaded circles, and 3 purple blank squigglesâ also form a set as each card has properties different from every other. This game is really just the cap-set problem for ðœ43 . Again a natural question arises. How many cards can you have such that you cannot form a set? Pelegrino ([120]) showed that it is possible to have 20 cards with no sets, but any collection of 21 cards must contain a set, thus settling the question. In general what is the largest subset of ðœðð containing no three-term arithmetic progressions? More precisely define ð3 (ð) to be the largest cardinality of a set ðŽ â â€ðð so that ðŽ does not contain any 3-term arithmetic progressions. When ð = 1, this is very similar to Rothâs Theorem discussed above (Theorem 7.5.2). For example, consider the cap-set problem in ðŽ â â€10 . Then ðŽ = {1, 2, 4, 9} does not contain any APs of length 3, which shows that ð3 (ðŽ) ⥠4. Again upper bounds are more difficult to achieve. Notice that the set ðŽ = {1, 2, 4, 8} may at first glance look like a cap set in â€10 , but it is not, as {4, 8, 2} is a 3-term AP in â€10 . Can you find any cap sets in â€10 consisting of 5 or more elements? The cap-set problem was recently solved by Ellenberg and Gijswijt ([44]) wherein they proved the following. Theorem 7.5.6. Suppose that ðŽ â â€ð3 is a set not containing any 3-term arithmetic progressions. Then |ðŽ| = ð((2.756)ð ). This Theorem has an interesting history (see [4], for example). In a series of papers, it was shown that ð3 (ð) = ð(3ð ), and then ð3 (ð) = ð(3ð /ð). In a much praised paper, Bateman and Katz ([4]) showed that
7.6. The spectral gap theorem
161
ð3 (ð) = ð(3ð /ð1+ð ) for some ð > 0. Thus, Ellenbergâs and Gijswijtâs result that ð3 (ð) = ð(2.756 . . . )ð was quite a breakthrough! We should note that Ellenberg first posted a solo-authored paper of this result, though Dion Gijswijt happened to prove the same result independently around the same time. Thus, their proofs, which both utilized the Polynomial Method along with similar ideas from a paper of Croot, Lev, and Pach ([23]) on cap-sets in â€ð4 , were combined and submitted as a joint paper. Truly, math is a social enterprise!
7.6 The spectral gap theorem Algebraic graph theory is a subject which aims to bring the methods of algebra (specifically group theory and linear algebra) to the study of graphs. The Spectral Gap Theorem is an oft underused and undervalued tool from algebraic graph theory. It has been used to study a wide variety of problems in combinatorics, including the ErdÅs-distance problem, Waringâs problem, and the sum-product problem. We explore some of these results below, but first we describe the Spectral Gap Theorem and the related notion of Cayley Graphs. The various terms that appear in the statement of the theorem will be explained in the course of giving its proof. Theorem 7.6.1 (Spectral Gap Theorem for Symmetric Graphs). Let ðº be a ð-regular simple graph with ð eigenvalues ð1 ⥠⯠⥠ðð . Let ð ðâ = ( max |ðð |) , ð 2â€ðâ€ð and let ð, ð â ð(ðº) be subsets of vertices. If â|ð||ð | > ðâ then there exists an ð â ð edge in ðº (that is, there exists an edge incident to a vertex in ð and a vertex in ð ). In particular if |ð| > ðâ , then there exist two distinct vertices ð¥, ðŠ â ð such that there is an edge from ð¥ to ðŠ. Proof. 14 Let ðž = (ððð ) be the 0 â 1 matrix, called the adjacency matrix, where ððð = 1 if there exists an edge from ð¥ð to ð¥ð , and ððð = 0 otherwise. Since ðº is undirected, the matrix ðž is symmetric, and hence ðž has real eigenvalues. Let ð£ 1 , . . . , ð£ ð be orthonormal eigenvectors of ðž 14 This
proof was first shown to me by Jonathan Pakianathan.
162
Chapter 7. Combinatorics in finite fields
with eigenvalues ð1 ⥠ð2 ⥠⯠⥠ðð . It turns out (see Exercise 7.16) that the degree ð = ð1 is the largest eigenvalue and corresponds to the eigenvector (1, . . . , 1). Any real ð-dimensional vector ð£ can be written ð ð£ = âð=1 âšð£, ð£ð â©ð£ð where âšâ
, â
â© is the dot product. Dotting this equation ð
with itself yields the Plancherel identity âšð£, ð£â© = âð=1 âšð£, ð£ð â©2 . Let 1ð be the 0 â 1 vector whose ðth entry is 1 when ð£ ð â ð and zero otherwise. Define 1ð similarly. The number of directed ð â ð edges is given by ð ð 1ð ðž1ð = âš1ð , ðž1ð â© = â ðð âš1ð , ð£ð â©âš1ð , ð£ð â©. ð=1
Note that âš1ð , ð£ 1 â© = |ð|/âð. Hence, peeling off the ð = 1 term, we see that: ð |ð||ð |ð ð 1ð ðž1ð = + â ðð âš1ð , ð£ð â©âš1ð , ð£ð â©. ð ð=2 If we let ð
ð
= â ðð âš1ð , ð£ð â©âš1ð , ð£ð â© ð=2
then by Cauchy-Schwarz we have 1/2
| | |ð
| †max |ðð | ||ââš1ð , ð£ð â©2 || 2â€ðâ€ð |ð |
1/2
| | |ââš1 , ð£ â©2 | ð ð | | |ð |
.
Thus by Plancherel, we see that |ð
| †max |ðð |â|ð||ð |. 2â€ðâ€ð
ð|ð||ð |
Since the number of ð âð edges is ð +ð
, then there will be an ð âð edge so long as ð|ð||ð | > max |ðð |â|ð||ð |. ð 2â€ðâ€ð The theorem follows immediately. The Spectral Gap Theorem has an analogue when ðº is directed, though in general the eigenvalues of ðº are complex. We refer the interested reader to [2], for example.
7.6. The spectral gap theorem
163
We next introduce the notion of Cayley graphs and explore a Spectral Gap Theorem for Cayley digraphs (directed graphs). Let ðº denote an additive abelian group, and suppose that ð â ðº is a subset. We then consider the Cayley graph ð¢(ð, ðž) = ð¶ððŠ(ðº, ð) where ð = ðº and where two vertices ð¥, ðŠ â ðº are connected if and only if ðŠ â ð¥ â ð. Note that if ð is a symmetric set (that is, if ð¥ â ð, then âð¥ â ð), the Cayley graph is undirected (possibly with loops if we allow 0 â ð). In general the Cayley graph is a digraph. First we describe the Spectral Gap Theorem in this context. Theorem 7.6.2. Let ð¶ððŠ(ðº, ð) be the Cayley digraph of the abelian group ðº over the set ð â ðº. Let {ð1 , . . . , ðð } denote the set of additive characters on ðº, where ð1 â¡ 1 denotes the trivial character. Define ðâ =
| | ð max || â ðð (ð )|| . |ð| 2â€ðâ€ð |ð âð |
Let ð and ð be sets of vertices from the Cayley graph. Then ð¶ððŠ(ðº, ð) has an ð â ð edge if the geometric mean â|ð||ð | > ðâ ; if |ð| > ðâ , then there exist two distinct vertices ð¥, ðŠ â ð such that there is an edge from ð¥ to ðŠ. We note that ð¶ððŠ(ðº, ð) is connected if and only if ð is a generating set (i.e., a subset whose elements and inverses span the group ðº). Cayley digraphs have a beautiful correspondence between the characters of the underling group ðº and the eigenvalues of the adjacency matrix of ð¶ððŠ(ðº, ð). Proposition 7.6.3. Let ð¶ððŠ(ðº, ð) be the Cayley graph of the abelian group ðº with respect to the set ð. Let Î be the eigenvalues of the adjacency matrix of ð¶ððŠ(ðº, ð), and let ð denote the characters of ðº. Then each eigenvalue ð â Î corresponds to a sum of the form ð = â ð(ð ), ð âð
for some ð â ð. A priori, it is not clear that the number of eigenvalues of ð¶ððŠ(ðº, ð) and the number of characters on ðº should be equal in number. However this does occur since we have |ðº|-many eigenvectors, and they are
164
Chapter 7. Combinatorics in finite fields
all distinct. In particular, the adjacency matrix of ð¶ððŠ(ðº, ð) is always diagonalizable. See [34] for a nice survey of these and similar results.
7.6.1 Applications of the spectral gap theorem Finally we describe some problems15 which employ the malleable Spectral Gap Theorem in various settings. When studying Waringâs problem for a fixed integer ð ⥠2 and for a fixed ring ð
, we define ðº(ð, ðž) to be the graph with vertex set ð = ð
, where two vertices ð¥ and ðŠ are connected by an edge if and only if ð¥ â ðŠ is a ðth power in ð
. Note that this graph need not be symmetric. In particular its symmetry will depend on whether or not â1 is a ðth power in ð
. This same choice of graph also gives a result analogous to the Furstenberg-SarkÅzy Theorem on squares ([58, 135]). We first recall the classical theorem. Theorem 7.6.4. Let ðž â â€+ have positive natural density in the sense that |ðž â© {1, . . . ð}| lim > 0. ð ðââ Then ðž contains two distinct elements ð¥, ðŠ â ðž such that ð¥ â ðŠ is a square. A finite field analogue was obtained by DemiroÄlu ([33]): Theorem 7.6.5. Let ðž â ðœð satisfy |ðž| >
ðð . âðâ1
Then ðž contains two
distinct elements ð¥ and ðŠ such that ð¥ â ðŠ is a ðth power in ðœð . Other similar results include the following. Theorem 7.6.6. Let ðž â ð2 (ðœð ) be such that |ðž| â« ð5/2 . Then for all ð¡ â ðœâð , there exist matrices ðŽ and ðµ in ðž such that det(ðŽ â ðµ) = ð¡. The graph used here is the natural one: Fix ð¡ â ðœâð . The vertex set is ð2 (ðœð ), and two matrices (i.e., vertices) ðŽ and ðµ are connected if and only if ðµ â ðŽ has determinant ð¡. Note that we must have |ðž| â« ð2 as the set ð ð ð¹ = {[ ] ⶠð, ð â ðœð } 0 0 15 Many of these results are work of my former graduate student YeÅim DemiroÄlu Karabulat.
7.6. The spectral gap theorem
165
has cardinality ð2 , and yet the difference of any two matrices from ð¹ has determinant zero. An immediate corollary is the following: 4 Corollary 7.6.7. Let ðŽ, ðµ, ð¶, ð· â ðœð be such that â|ðŽ||ðµ||ð¶||ð·| â« ð3/4 . Then (ðŽ â ðµ)(ð¶ â ð·) = ðœâð .
In particular if |ðŽ| â« ð3/4 , then (ðŽ â ðŽ)(ðŽ â ðŽ) = ðœð . This theorem on sums and products is just shy of the following remarkable result of Murphy and Petridis ([113]). Just like the finite field distance problem, you are able to get away with less if you only want a positive proportion of ðœð rather than the entire field. Theorem 7.6.8. Let ðŽ â ðœð with |ðŽ| > ð2/3âð¿ for any ð¿ < 1/13, 542. ð Then |(ðŽ â ðŽ)(ðŽ â ðŽ)| > 2 . Another result achieved using the Spectral Gap Theorem is as follows. Theorem 7.6.9. Every matrix ðŽ â ðð (ðœð ) is a sum of 2 invertible matrices except in the case that ðð (ðœð ) = ðœ2 (that is if ð = 1 and ð = 2). Moreover if ð = 2, then every matrix in ð2 (ðœð ) is the sum of two matrices of determinant 1. Theorem 7.6.9 along with The Artin-Weddenburn Theorem can be used to reconstruct the following classic result of Henriksen ([78]): Theorem 7.6.10. Let ð
be a finite ring (with identity) of odd order. Then every element of ð
is the sum of two units of ð
. More results using the Spectral Gap Theorem appear in the literature, but hopefully this gives an idea of exactly how flexible and hence how useful the Spectral Gap Theorem can be. There is so much more we could explore in combinatorics in finite fields and beyond, but I strongly recommend you start looking up some of the references in this book and dig into some of these problems, especially the unsolved problems (including the finite field distance problem)! There is much work left to do. As Richard Feynman ([54]) once said, âSome new ideas are here needed!â
166
Chapter 7. Combinatorics in finite fields
7.7 Exercises: Chapter 7 Exercise 7.1. Prove that we have the trivial incidence bound ðŒ(ð, ð¿) †|ð¿||ð| for any set of points and lines over any field ðœ. Exercise 7.2. Show that taking ð = ðœ2ð and ð¿ the set of all possible lines contradicts the Szemerédi-Trotter Theorem thereby showing that Theorem 7.1.1 does not hold in finite fields. Exercise 7.3. Let ðŽ â â€, and notice that since addition is commutative and since subtraction is not commutative one would expect to have |ðŽ â ðŽ| ⥠|ðŽ + ðŽ| for most sets ðŽ. Nonetheless find a set ðŽ â †such that |ðŽ â ðŽ| < |ðŽ + ðŽ|. Such sets are called MSTD sets (More Sums Than Differences). Exercise 7.4. Prove the following analogue of the Kakeya conjecture for circles: If ðž â ðœðð contains a sphere of radius ð¡ for all ð¡ â ðœâð , then |ðž| â«ð ðð . Exercise 7.5. ([41]) Let ðž + (ðŽ) denote the additive energy of ðŽ â â: ðž + (ðŽ) = {(ð, ð, ð, ð) â ðŽ à ðŽ à ðŽ à ðŽ ⶠð + ð = ð + ð}. Use Cauchy-Schwarz to prove |ðŽ|4 †ðž + (ðŽ)|ðŽ + ðŽ|. (Compare to Lemma 6.2.2.) Exercise 7.6. ([3]) Prove that |ðŽ/ðŽ + ðŽ| â« |ðŽ|3/2 for all sets ðŽ â â ⧵ {0}. Exercise 7.7. Recall that the Minkowski dimension of a set ðž â âð is given by log ðð , dimð (ðž) = lim+ ðâ0 log(1/ð) where ðð is the number of ð-balls needed to cover the set ðž. Show that the 1 set ðž = {1/ð}ðââ€+ has Minkowski dimension dimð (ðž) = 2 . Exercise 7.8. Adapt the previous example to find a set ðž â â with Minkowski dimension dimð (ðž) = ðŒ for any ðŒ â [0, 1]. Exercise 7.9. Prove that Î(ð) exists for all ð ⥠2.
7.7. Exercises: Chapter 7
167
Exercise 7.10. Prove that Î(ð, â€ð ) = 1 if and only if gcd(ð, ð â 1) = 1. Conclude that (7.5) holds: Î(ð, ðœð ) = 1 for all finite fields ðœð if and only if ð = 1. Exercise 7.11. Prove that (7.6) holds using the pigeonhole principle. ð
ð
Exercise 7.12. Let ð = ð11 . . . ðââ be the prime factorization of ð. Then, Î(ð, â€ð ) †ð if and only if Î(ð, â€ððð ) †ð for ð = 1, . . . , â. ð
Exercise 7.13. Prove that if ð and ðâ² are positive integers such that ð ⣠ðâ² , and if Î(ð, ð) > ð, then Î(ð, ðâ² ) > ð. Exercise 7.14. Let ð
âð = {ð¥ð ⶠð¥ â â€ðâ } ⧵ {0}. Then, ð
âð â â€ðÃâ if ð ⥠â. That is, if ð ⥠â then all the nonzero ðth power residues in â€ðâ are units. Exercise 7.15. Let ð(ð) be as in the âNo-three-in-lineâ game (see Section 7.5). Prove that ð(ð) †2ð for all ð. Exercise 7.16. Let ðº be a regular (undirected) graph of degree ð (meaning every vertex has degree ð). Prove that if the spectrum of the adjacency matrix of ðº is {ð1 , . . . , ðð }, then we have max |ðð | = ð.
1â€ðâ€ð
Bibliography [1] Amol Aggarwal, On unit distances in a convex polygon, Discrete Math. 338 (2015), no. 3, 88â92, DOI 10.1016/j.disc.2014.10.009. MR3291870 [2] Noga Alon and Joel H. Spencer, The probabilistic method, 4th ed., Wiley Series in Discrete Mathematics and Optimization, John Wiley & Sons, Inc., Hoboken, NJ, 2016. MR3524748 [3] Antal Balog, A note on sum-product estimates, Publ. Math. Debrecen 79 (2011), no. 34, 283â289, DOI 10.5486/PMD.2011.5104. MR2907965 [4] Michael Bateman and Nets Hawk Katz, New bounds on cap sets, J. Amer. Math. Soc. 25 (2012), no. 2, 585â613, DOI 10.1090/S0894-0347-2011-00725-X. MR2869028 [5] Mike Bennett, Alex Iosevich, and Jonathan Pakianathan, Three-point configurations determined by subsets of ðœ2ð via the Elekes-Sharir paradigm, Combinatorica 34 (2014), no. 6, 689â706, DOI 10.1007/s00493-014-2978-6. MR3296181 [6] Michael Bennett, Jeremy Chapman, David Covert, Derrick Hart, Alex Iosevich, and Jonathan Pakianathan, Long paths in the distance graph over large subsets of vector spaces over finite fields, J. Korean Math. Soc. 53 (2016), no. 1, 115â126, DOI 10.4134/JKMS.2016.53.1.115. MR3450941 [7] Michael Bennett, Derrick Hart, Alex Iosevich, Jonathan Pakianathan, and Misha Rudnev, Group actions and geometric combinatorics in ðœðð , Forum Math. 29 (2017), no. 1, 91â110, DOI 10.1515/forum-2015-0251. MR3592595 [8] A. S. Besicovitch, The Kakeya problem, Amer. Math. Monthly 70 (1963), 697â706, DOI 10.2307/2312249. MR157266 [9] M. Bhaskaran, Sums of ðth powers in algebraic and Abelian number fields, Arch. Math. (Basel) 17 (1966), 497â504, DOI 10.1007/BF01899421. MR204400 [10] P. Birklbauer, A. Iosevich, and T. Pham, Distances from points to planes, Acta Arith. 186 (2018), no. 3, 219â224, DOI 10.4064/aa171110-23-8. MR3879390 [11] T. F. Bloom, A quantitative improvement for Rothâs theorem on arithmetic progressions, J. Lond. Math. Soc. (2) 93 (2016), no. 3, 643â663, DOI 10.1112/jlms/jdw010. MR3509957 [12] J. Bourgain and A. Glibichuk, Exponential sum estimates over a subgroup in an arbitrary finite field, J. Anal. Math. 115 (2011), 51â70, DOI 10.1007/s11854-011-0023-x. MR2855033 [13] J. Bourgain, A. A. Glibichuk, and S. V. Konyagin, Estimates for the number of sums and products and for exponential sums in fields of prime order, J. London Math. Soc. (2) 73 (2006), no. 2, 380â398, DOI 10.1112/S0024610706022721. MR2225493
169
170
Bibliography
[14] J. Bourgain, N. Katz, and T. Tao, A sum-product estimate in finite fields, and applications, Geom. Funct. Anal. 14 (2004), no. 1, 27â57, DOI 10.1007/s00039-004-0451-1. MR2053599 [15] Peter Brass, William Moser, and János Pach, Research problems in discrete geometry, Springer, New York, 2005. MR2163782 [16] Mei-Chu Chang, The ErdÅs-Szemerédi problem on sum set and product set, Ann. of Math. (2) 157 (2003), no. 3, 939â957, DOI 10.4007/annals.2003.157.939. MR1983786 [17] Jeremy Chapman, M. Burak ErdoÄan, Derrick Hart, Alex Iosevich, and Doowon Koh, Pinned distance sets, ð-simplices, Wolffâs exponent in finite fields and sumproduct estimates, Math. Z. 271 (2012), no. 1-2, 63â93, DOI 10.1007/s00209-0110852-4. MR2917133 [18] F. R. K. Chung, The number of different distances determined by ð points in the plane, J. Combin. Theory Ser. A 36 (1984), no. 3, 342â354, DOI 10.1016/00973165(84)90041-4. MR744082 [19] Fan R. K. Chung, E. Szemerédi, and W. T. Trotter, The number of different distances determined by a set of points in the Euclidean plane, Discrete Comput. Geom. 7 (1992), no. 1, 1â11, DOI 10.1007/BF02187820. MR1134448 [20] Javier Cilleruelo, Alex Iosevich, Ben Lund, Oliver Roche-Newton, and Misha Rudnev, Elementary methods for incidence problems in finite fields, Acta Arith. 177 (2017), no. 2, 133â142, DOI 10.4064/aa8225-10-2016. MR3600755 [21] James Arthur Cipra, Todd Cochrane, and Christopher Pinner, Heilbronnâs conjecture on Waringâs number (mod ð), J. Number Theory 125 (2007), no. 2, 289â297, DOI 10.1016/j.jnt.2006.12.001. MR2332590 [22] H. S. M. Coxeter, Projective geometry, Springer-Verlag, New York, 1994. Revised reprint of the second (1974) edition. MR1335229 [23] Ernie Croot, Vsevolod F. Lev, and Péter Pál Pach, Progression-free sets in â€ð4 are exponentially small, Ann. of Math. (2) 185 (2017), no. 1, 331â337, DOI 10.4007/annals.2017.185.1.7. MR3583357 [24] David J. Covert, Results on the ErdÅs-Falconer distance problem in â€ðð for odd ð, J. Math. Anal. Appl. 426 (2015), no. 2, 727â733, DOI 10.1016/j.jmaa.2015.01.063. MR3314855 [25] David Covert, Derrick Hart, Alex Iosevich, Steven Senger, and Ignacio UriarteTuero, A Furstenberg-Katznelson-Weiss type theorem on (ð + 1)-point configurations in sets of positive density in finite field geometries, Discrete Math. 311 (2011), no. 6, 423â430, DOI 10.1016/j.disc.2010.10.009. MR2799893 [26] David Covert, Alex Iosevich, and Jonathan Pakianathan, Geometric configurations in the ring of integers modulo ðâ , Indiana Univ. Math. J. 61 (2012), no. 5, 1949â1969, DOI 10.1512/iumj.2012.61.4751. MR3119606 [27] David Covert, Doowon Koh, and Youngjin Pi, On the sums of any ð points in finite fields, SIAM J. Discrete Math. 30 (2016), no. 1, 367â382, DOI 10.1137/140982441. MR3466193 [28] David Covert, Doowon Koh, and Youngjin Pi, The ð-resultant modulus set problem on algebraic varieties over finite fields, Finite Fields Appl. 48 (2017), 68â86, DOI 10.1016/j.ffa.2017.07.012. MR3705735 [29] David Covert, Doowon Koh, and Youngjin Pi, The generalized ð-resultant modulus set problem in finite fields, J. Fourier Anal. Appl. 25 (2019), no. 3, 1026â1052, DOI 10.1007/s00041-018-9619-1. MR3953496
Bibliography
171
[30] H. Davenport, On Waringâs problem for cubes, Acta Math. 71 (1939), 123â143, DOI 10.1007/BF02547752. MR026 [31] H. Davenport, On Waringâs problem for fourth powers, Ann. of Math. (2) 40 (1939), 731â747, DOI 10.2307/1968889. MR253 [32] Roy O. Davies, Some remarks on the Kakeya problem, Proc. Cambridge Philos. Soc. 69 (1971), 417â421, DOI 10.1017/s0305004100046867. MR272988 [33] YeÅim DemiroÄlu Karabulut, Waringâs problem in finite rings, J. Pure Appl. Algebra 223 (2019), no. 8, 3318â3329, DOI 10.1016/j.jpaa.2018.11.003. MR3926213 [34] YeÅim DemiroÄlu Karabulut, Unit-graphs and special unit-digraphs on matrix rings, Forum Math. 30 (2018), no. 6, 1397â1412, DOI 10.1515/forum-2017-0262. MR3871453 [35] Reinhard Diestel, Graph theory, 5th ed., Graduate Texts in Mathematics, vol. 173, Springer, Berlin, 2017, DOI 10.1007/978-3-662-53622-3. MR3644391 [36] Rainer Dietmann, On the ErdÅs-Falconer distance problem for two sets of different size in vector spaces over finite fields, Monatsh. Math. 170 (2013), no. 3-4, 343â359, DOI 10.1007/s00605-012-0469-7. MR3055791 [37] X. Du, L. Guth, Y. Ou, H. Wang, B. Wilson, and R. Zhang, Weighted restriction estimates and application to Falconer distance set problem, Preprint arXiv:1802.10186. [38] Xiumin Du and Ruixiang Zhang, Sharp ð¿2 estimates of the Schrödinger maximal function in higher dimensions, Ann. of Math. (2) 189 (2019), no. 3, 837â861, DOI 10.4007/annals.2019.189.3.4. MR3961084 [39] Evan P. Dummit and Márton Hablicsek, Kakeya sets over non-Archimedean local rings, Mathematika 59 (2013), no. 2, 257â266, DOI 10.1112/S002557931300003X. MR3081770 [40] Zeev Dvir, On the size of Kakeya sets in finite fields, J. Amer. Math. Soc. 22 (2009), no. 4, 1093â1097, DOI 10.1090/S0894-0347-08-00607-3. MR2525780 [41] György Elekes, On the number of sums and products, Acta Arith. 81 (1997), no. 4, 365â367, DOI 10.4064/aa-81-4-365-367. MR1472816 [42] Gy. Elekes and I. Z. Ruzsa, Few sums, many products, Studia Sci. Math. Hungar. 40 (2003), no. 3, 301â308, DOI 10.1556/SScMath.40.2003.3.4. MR2036961 [43] György Elekes and Micha Sharir, Incidences in three dimensions and distinct distances in the plane, Combin. Probab. Comput. 20 (2011), no. 4, 571â608, DOI 10.1017/S0963548311000137. MR2805398 [44] Jordan S. Ellenberg and Dion Gijswijt, On large subsets of ðœðð with no three-term arithmetic progression, Ann. of Math. (2) 185 (2017), no. 1, 339â343, DOI 10.4007/annals.2017.185.1.8. MR3583358 [45] Jordan S. Ellenberg, Richard Oberlin, and Terence Tao, The Kakeya set and maximal conjectures for algebraic varieties over finite fields, Mathematika 56 (2010), no. 1, 1â 25, DOI 10.1112/S0025579309000400. MR2604979 [46] Christian Elsholtz, The number Î(ð) in Waringâs problem, Acta Arith. 131 (2008), no. 1, 43â49, DOI 10.4064/aa131-1-3. MR2373515 [47] M. Burak ErdogÌan, A bilinear Fourier extension theorem and applications to the distance set problem, Int. Math. Res. Not. 23 (2005), 1411â1425, DOI 10.1155/IMRN.2005.1411. MR2152236 [48] Paul Erdös, On the Sum and Difference of Squares of Primes, J. London Math. Soc. 12 (1937), no. 2, 133â136, DOI 10.1112/jlms/s1-12.1.133. MR1574651
172
Bibliography
[49] P. Erdös, On sets of distances of ð points, Amer. Math. Monthly 53 (1946), 248â250, DOI 10.2307/2305092. MR15796 [50] P. ÃrdeÅ¡, An asymptotic inequality in the theory of numbers (Russian, with English summary), Vestnik Leningrad. Univ. 15 (1960), no. 13, 41â49. MR0126424 [51] P. ErdÅs and E. Szemerédi, On sums and products of integers, Studies in pure mathematics, BirkhÀuser, Basel, 1983, pp. 213â218. MR820223 [52] K. J. Falconer, On the Hausdorff dimensions of distance sets, Mathematika 32 (1985), no. 2, 206â212 (1986), DOI 10.1112/S0025579300010998. MR834490 [53] K. J. Falconer, The geometry of fractal sets, Cambridge Tracts in Mathematics, vol. 85, Cambridge University Press, Cambridge, 1986. MR867284 [54] R. Feynman, Interview with Yorkshire Television program, âTake the World from Another Point of View,â (1972). [55] Kevin Ford, Sums and products from a finite set of real numbers, Ramanujan J. 2 (1998), no. 1-2, 59â66, DOI 10.1023/A:1009709908223. Paul ErdÅs (1913â1996). MR1642873 [56] Kevin Ford, The distribution of integers with a divisor in a given interval, Ann. of Math. (2) 168 (2008), no. 2, 367â433, DOI 10.4007/annals.2008.168.367. MR2434882 [57] G. A. FreıÌman, On the addition of finite sets (Russian), Dokl. Akad. Nauk SSSR 158 (1964), 1038â1041. MR0168529 [58] Harry Furstenberg, Ergodic behavior of diagonal measures and a theorem of Szemerédi on arithmetic progressions, J. Analyse Math. 31 (1977), 204â256, DOI 10.1007/BF02813304. MR498471 [59] M. Z. Garaev, An explicit sum-product estimate in ðœð , Int. Math. Res. Not. IMRN 11 (2007), Art. ID rnm035, 11. MR2344270 [60] Moubariz Z. Garaev and Chun-Yen Shen, On the size of the set ðŽ(ðŽ + 1), Math. Z. 265 (2010), no. 1, 125â132, DOI 10.1007/s00209-009-0504-0. MR2606952 [61] Julia Garibaldi, Alex Iosevich, and Steven Senger, The ErdÅs distance problem, Student Mathematical Library, vol. 56, American Mathematical Society, Providence, RI, 2011, DOI 10.1090/stml/056. MR2721878 [62] Chris Godsil and Gordon Royle, Algebraic graph theory, Graduate Texts in Mathematics, vol. 207, Springer-Verlag, New York, 2001, DOI 10.1007/978-1-4613-0163-9. MR1829620 [63] W. T. Gowers, A new proof of Szemerédiâs theorem, Geom. Funct. Anal. 11 (2001), no. 3, 465â588, DOI 10.1007/s00039-001-0332-9. MR1844079 [64] B. Green, Review of A quantitative improvement for Rothâs theorem on arithmetic progressions, MathSciNet review. MR3509957. [65] Ben Green and Terence Tao, The primes contain arbitrarily long arithmetic progressions, Ann. of Math. (2) 167 (2008), no. 2, 481â547, DOI 10.4007/annals.2008.167.481. MR2415379 [66] Allan Greenleaf and Alex Iosevich, On triangles determined by subsets of the Euclidean plane, the associated bilinear operators and applications to discrete geometry, Anal. PDE 5 (2012), no. 2, 397â409, DOI 10.2140/apde.2012.5.397. MR2970712 [67] Ron Graham, On the growth of a van der Waerden-like function, Integers 6 (2006), A29, 5. MR2264844 [68] Larry Guth, Polynomial methods in combinatorics, University Lecture Series, vol. 64, American Mathematical Society, Providence, RI, 2016, DOI 10.1090/ulect/064. MR3495952
Bibliography
173
[69] Larry Guth, Alex Iosevich, Yumeng Ou, and Hong Wang, On Falconerâs distance set problem in the plane, Invent. Math. 219 (2020), no. 3, 779â830, DOI 10.1007/s00222019-00917-x. MR4055179 [70] Larry Guth and Nets Hawk Katz, On the ErdÅs distinct distances problem in the plane, Ann. of Math. (2) 181 (2015), no. 1, 155â190, DOI 10.4007/annals.2015.181.1.2. MR3272924 [71] Brandon Hanson, Ben Lund, and Oliver Roche-Newton, On distinct perpendicular bisectors and pinned distances in finite fields, Finite Fields Appl. 37 (2016), 240â264, DOI 10.1016/j.ffa.2015.10.002. MR3426588 [72] G. H. Hardy and J. E. Littlewood, Some Problems of âPartitio Numerorumâ (VIII): The Number Gamma(k) in Waringâs Problem, Proc. London Math. Soc. (2) 28 (1928), no. 7, 518â542, DOI 10.1112/plms/s2-28.1.518. MR1575871 [73] Derrick Hart and Alex Iosevich, Ubiquity of simplices in subsets of vector spaces over finite fields (English, with English and Russian summaries), Anal. Math. 34 (2008), no. 1, 29â38, DOI 10.1007/s10476-008-0103-z. MR2379694 [74] Derrick Hart and Alex Iosevich, Sums and products in finite fields: an integral geometric viewpoint, Radon transforms, geometry, and wavelets, Contemp. Math., vol. 464, Amer. Math. Soc., Providence, RI, 2008, pp. 129â135, DOI 10.1090/conm/464/09080. MR2440133 [75] Derrick Hart, Alex Iosevich, and Jozsef Solymosi, Sum-product estimates in finite fields via Kloosterman sums, Int. Math. Res. Not. IMRN 5 (2007), Art. ID rnm007, 14, DOI 10.1093/imrn/rnm007. MR2341599 [76] Derrick Hart, Alex Iosevich, Doowon Koh, and Misha Rudnev, Averages over hyperplanes, sum-product theory in vector spaces over finite fields and the ErdÅs-Falconer distance conjecture, Trans. Amer. Math. Soc. 363 (2011), no. 6, 3255â3275, DOI 10.1090/S0002-9947-2010-05232-8. MR2775806 [77] Harald Andrés Helfgott and Misha Rudnev, An explicit incidence theorem in ðœð , Mathematika 57 (2011), no. 1, 135â145, DOI 10.1112/S0025579310001208. MR2764161 [78] Melvin Henriksen, Two classes of rings generated by their units, J. Algebra 31 (1974), 182â193, DOI 10.1016/0021-8693(74)90013-1. MR349745 [79] Do Duy Hieu and Van Thang Pham, Distinct distances on regular varieties over finite fields, J. Number Theory 173 (2017), 602â613, DOI 10.1016/j.jnt.2016.10.003. MR3581934 [80] David Hilbert, Beweis fÃŒr die Darstellbarkeit der ganzen Zahlen durch eine feste Anzahl ðð¡ðð Potenzen (Waringsches Problem) (German), Math. Ann. 67 (1909), no. 3, 281â300, DOI 10.1007/BF01450405. MR1511530 [81] Alex Iosevich, A view from the top, Student Mathematical Library, vol. 39, American Mathematical Society, Providence, RI, 2007. Analysis, combinatorics and number theory, DOI 10.1090/stml/039. MR2355095 [82] Alex Iosevich and Doowon Koh, The ErdÅs-Falconer distance problem, exponential sums, and Fourier analytic approach to incidence theorems in vector spaces over finite fields, SIAM J. Discrete Math. 23 (2008/09), no. 1, 123â135, DOI 10.1137/060669875. MR2452874 [83] Alex Iosevich and Doowon Koh, Extension theorems for the Fourier transform associated with nondegenerate quadratic surfaces in vector spaces over finite fields, Illinois J. Math. 52 (2008), no. 2, 611â628. MR2524655
174
Bibliography
[84] Alex Iosevich and Doowon Koh, Extension theorems for paraboloids in the finite field setting, Math. Z. 266 (2010), no. 2, 471â487, DOI 10.1007/s00209-009-0580-1. MR2678639 [85] Alex Iosevich and Doowon Koh, Extension theorems for spheres in the finite field setting, Forum Math. 22 (2010), no. 3, 457â483, DOI 10.1515/FORUM.2010.025. MR2652707 [86] Alex Iosevich, Doowon Koh, Thang Pham, Chun-Yen Shen, and Le Anh Vinh, A new bound for the ErdÅs distinct distances problem in the plane over prime fields, Acta Arith. 193 (2020), no. 2, 165â174, DOI 10.4064/aa190214-10-4. MR4066548 [87] A. Iosevich and M. Rudnev, ErdÅs distance problem in vector spaces over finite fields, Trans. Amer. Math. Soc. 359 (2007), no. 12, 6127â6142, DOI 10.1090/S0002-9947-0704265-1. MR2336319 [88] Alex Iosevich, Oliver Roche-Newton, and Misha Rudnev, On an application of Guth-Katz theorem, Math. Res. Lett. 18 (2011), no. 4, 691â697, DOI 10.4310/MRL.2011.v18.n4.a8. MR2831835 ð+1 [89] Alex Iosevich and Steven Senger, Sharpness of Falconerâs 2 estimate, Ann. Acad. Sci. Fenn. Math. 41 (2016), no. 2, 713â720, DOI 10.5186/aasfm.2016.4145. MR3525395 [90] Henryk Iwaniec and Emmanuel Kowalski, Analytic number theory, American Mathematical Society Colloquium Publications, vol. 53, American Mathematical Society, Providence, RI, 2004, DOI 10.1090/coll/053. MR2061214 [91] Timothy G. F. Jones, An improved incidence bound for fields of prime order. part A, European J. Combin. 52 (2016), no. part A, 136â145, DOI 10.1016/j.ejc.2015.09.004. MR3425971 [92] Timothy G. F. Jones and Oliver Roche-Newton, Improved bounds on the set ðŽ(ðŽ + 1), J. Combin. Theory Ser. A 120 (2013), no. 3, 515â526, DOI 10.1016/j.jcta.2012.11.001. MR3007133 [93] Haim Kaplan, JiÅà MatouÅ¡ek, Zuzana Safernová, and Micha Sharir, Unit distances in three dimensions, Combin. Probab. Comput. 21 (2012), no. 4, 597â610, DOI 10.1017/S0963548312000144. MR2942731 [94] Nets Hawk Katz, Izabella Åaba, and Terence Tao, An improved bound on the Minkowski dimension of Besicovitch sets in ð3 , Ann. of Math. (2) 152 (2000), no. 2, 383â446, DOI 10.2307/2661389. MR1804528 [95] Nets Hawk Katz and Chun-Yen Shen, A slight improvement to Garaevâs sum product estimate, Proc. Amer. Math. Soc. 136 (2008), no. 7, 2499â2504, DOI 10.1090/S00029939-08-09385-4. MR2390519 [96] Nets Hawk Katz and Chun-Yen Shen, Garaevâs inequality in finite fields not of prime order, Online J. Anal. Comb. 3 (2008), Art. 3, 6. MR2375606 [97] Nets Hawk Katz and Terence Tao, New bounds for Kakeya problems, J. Anal. Math. 87 (2002), 231â263, DOI 10.1007/BF02868476. Dedicated to the memory of Thomas H. Wolff. MR1945284 [98] Nets Hawk Katz and Gábor Tardos, A new entropy inequality for the ErdÅs distance problem, Towards a theory of geometric graphs, Contemp. Math., vol. 342, Amer. Math. Soc., Providence, RI, 2004, pp. 119â126, DOI 10.1090/conm/342/06136. MR2065258
Bibliography
175
[99] Doowon Koh and Chun-Yen Shen, Sharp extension theorems and Falconer distance problems for algebraic curves in two dimensional vector spaces over finite fields, Rev. Mat. Iberoam. 28 (2012), no. 1, 157â178, DOI 10.4171/RMI/672. MR2904136 [100] Doowon Koh and Hae-Sang Sun, Distance sets of two subsets of vector spaces over finite fields, Proc. Amer. Math. Soc. 143 (2015), no. 4, 1679â1692, DOI 10.1090/S0002-99392014-12386-0. MR3314081 [101] S. V. Konyagin and I. D. Shkredov, New results on sums and products in â (Russian, with Russian summary), Tr. Mat. Inst. Steklova 294 (2016), no. Sovremennye Problemy Matematiki, Mekhaniki i MatematicheskoÄ±Ì Fiziki. II, 87â98, DOI 10.1134/S0371968516030055. English version published in Proc. Steklov Inst. Math. 294 (2016), no. 1, 78â88. MR3628494 [102] Izabella Åaba, From harmonic analysis to arithmetic combinatorics, Bull. Amer. Math. Soc. (N.S.) 45 (2008), no. 1, 77â115, DOI 10.1090/S0273-0979-07-01189-5. MR2358378 [103] I. Åaba and T. Tao, An improved bound for the Minkowski dimension of Besicovitch sets in medium dimension, Geom. Funct. Anal. 11 (2001), no. 4, 773â806, DOI 10.1007/PL00001685. MR1866801 [104] Rudolf Lidl and Harald Niederreiter, Finite fields, 2nd ed., Encyclopedia of Mathematics and its Applications, vol. 20, Cambridge University Press, Cambridge, 1997. With a foreword by P. M. Cohn. MR1429394 [105] Mark Lewko, New restriction estimates for the 3-d paraboloid over finite fields, Adv. Math. 270 (2015), 457â479, DOI 10.1016/j.aim.2014.11.008. MR3286541 [106] Mark Lewko, Finite field restriction estimates based on Kakeya maximal operator estimates, J. Eur. Math. Soc. (JEMS) 21 (2019), no. 12, 3649â3707, DOI 10.4171/JEMS/910. MR4022712 [107] Allison Lewko and Mark Lewko, Endpoint restriction estimates for the paraboloid over finite fields, Proc. Amer. Math. Soc. 140 (2012), no. 6, 2013â2028, DOI 10.1090/S0002-9939-2011-11444-8. MR2888189 [108] JiÅà MatouÅ¡ek, The number of unit distances is almost linear for most norms, Adv. Math. 226 (2011), no. 3, 2618â2628, DOI 10.1016/j.aim.2010.09.009. MR2739786 [109] Pertti Mattila, Spherical averages of Fourier transforms of measures with finite energy; dimension of intersections and distance sets, Mathematika 34 (1987), no. 2, 207â228, DOI 10.1112/S0025579300013462. MR933500 [110] Pertti Mattila, Geometry of sets and measures in Euclidean spaces, Cambridge Studies in Advanced Mathematics, vol. 44, Cambridge University Press, Cambridge, 1995. Fractals and rectifiability, DOI 10.1017/CBO9780511623813. MR1333890 [111] Gerd Mockenhaupt and Terence Tao, Restriction and Kakeya phenomena for finite fields, Duke Math. J. 121 (2004), no. 1, 35â74, DOI 10.1215/S0012-7094-04-12112-8. MR2031165 [112] Leo Moser, On the different distances determined by ð points, Amer. Math. Monthly 59 (1952), 85â91, DOI 10.2307/2307105. MR46663 [113] Brendan Murphy and Giorgis Petridis, A second wave of expanders in finite fields, Combinatorial and additive number theory. II, Springer Proc. Math. Stat., vol. 220, Springer, Cham, 2017, pp. 215â238, DOI 10.1007/978-3-319-68032-3_15. MR3754906
176
Bibliography
[114] Brendan Murphy and Giorgis Petridis, An example related to the ErdÅs-Falconer question over arbitrary finite fields, Bull. Hellenic Math. Soc. 63 (2019), 38â39. MR3955393 [115] Melvyn B. Nathanson, On sums and products of integers, Proc. Amer. Math. Soc. 125 (1997), no. 1, 9â16, DOI 10.1090/S0002-9939-97-03510-7. MR1343715 [116] Melvyn B. Nathanson and Gérald Tenenbaum, Inverse theorems and the number of sums and products (English, with English and French summaries), Astérisque 258 (1999), xiii, 195â204. Structure theory of set addition. MR1701198 [117] Daniel Oberlin and Richard Oberlin, Unit distance problems, Amer. J. Math. 137 (2015), no. 1, 251â270, DOI 10.1353/ajm.2015.0004. MR3318091 [118] Tuomas Orponen, On the distance sets of Ahlfors-David regular sets, Adv. Math. 307 (2017), 1029â1045, DOI 10.1016/j.aim.2016.11.035. MR3590535 [119] E. Pegg Jr., Math Games: Chessboard Tasks, Apr. 11, 2005. http://www.maa.org/ editorial/mathgames/mathgames_04_11_05.html. [120] Giuseppe Pellegrino, Sul massimo ordine delle calotte in ð 4,3 (Italian), Matematiche (Catania) 25 (1970), 149â157 (1971). MR363952 [121] Yuval Peres and Wilhelm Schlag, Smoothness of projections, Bernoulli convolutions, and the dimension of exceptions, Duke Math. J. 102 (2000), no. 2, 193â251, DOI 10.1215/S0012-7094-00-10222-0. MR1749437 [122] Giorgis Petridis, Pinned algebraic distances determined by Cartesian products in ðœð2 , Proc. Amer. Math. Soc. 145 (2017), no. 11, 4639â4645, DOI 10.1090/proc/13649. MR3691983 [123] Duc Hiep Pham, Thang Pham, and Le Anh Vinh, An improvement on the number of simplices in ðœðð , Discrete Appl. Math. 221 (2017), 95â105, DOI 10.1016/j.dam.2016.12.023. MR3612592 [124] Thang Pham, Nguyen Duy Phuong, Nguyen Minh Sang, Claudiu Valculescu, and Le Anh Vinh, Distinct distances between points and lines in ðœ2ð , Forum Math. 30 (2018), no. 4, 799â808, DOI 10.1515/forum-2016-0248. MR3824792 [125] Pham Van Thang and Le Anh Vinh, ErdÅs-Rényi graph, Szemerédi-Trotter type theorem, and sum-product estimates over finite rings, Forum Math. 27 (2015), no. 1, 331â 342, DOI 10.1515/forum-2011-0161. MR3334065 [126] Alain M. Robert, A course in ð-adic analysis, Graduate Texts in Mathematics, vol. 198, Springer-Verlag, New York, 2000, DOI 10.1007/978-1-4757-3254-2. MR1760253 [127] Oliver Roche-Newton, Misha Rudnev, and Ilya D. Shkredov, New sumproduct type estimates over finite fields, Adv. Math. 293 (2016), 589â605, DOI 10.1016/j.aim.2016.02.019. MR3474329 [128] Keith McKenzie Rogers, The finite field Kakeya problem, Amer. Math. Monthly 108 (2001), no. 8, 756â759, DOI 10.2307/2695620. MR1865664 [129] K. F. Roth, On certain sets of integers, J. London Math. Soc. 28 (1953), 104â109, DOI 10.1112/jlms/s1-28.1.104. MR51853 [130] Misha Rudnev, On the number of incidences between points and planes in three dimensions, Combinatorica 38 (2018), no. 1, 219â254, DOI 10.1007/s00493-016-33296. MR3776354 [131] Misha Rudnev and Ilya D. Shkredov, On the restriction problem for discrete paraboloid in lower dimension, Adv. Math. 339 (2018), 657â671, DOI 10.1016/j.aim.2018.10.002. MR3866909
Bibliography
177
[132] Hans Salié, Ãber die Kloostermanschen Summen ð(ð¢, ð£; ð) (German), Math. Z. 34 (1932), no. 1, 91â109, DOI 10.1007/BF01180579. MR1545243 [133] Tom Sanders, On Rothâs theorem on progressions, Ann. of Math. (2) 174 (2011), no. 1, 619â636, DOI 10.4007/annals.2011.174.1.20. MR2811612 [134] Peter Sarnak, Some applications of modular forms, Cambridge Tracts in Mathematics, vol. 99, Cambridge University Press, Cambridge, 1990, DOI 10.1017/CBO9780511895593. MR1102679 [135] A. SárkÅzy, On difference sets of sequences of integers. I, Acta Math. Acad. Sci. Hungar. 31 (1978), no. 1-2, 125â149, DOI 10.1007/BF01896079. MR466059 [136] Wolfgang M. Schmidt, Equations over finite fields. An elementary approach, Lecture Notes in Mathematics, Vol. 536, Springer-Verlag, Berlin-New York, 1976. MR0429733 [137] Hiroshi Sekigawa and Kenji Koyama, Nonexistence conditions of a solution for the congruence ð¥1ð + ⯠+ ð¥ð ð â¡ ð (mod ðð ), Math. Comp. 68 (1999), no. 227, 1283â 1297, DOI 10.1090/S0025-5718-99-01067-4. MR1627821 [138] George Shakan, On higher energy decompositions and the sum-product phenomenon, Math. Proc. Cambridge Philos. Soc. 167 (2019), no. 3, 599â617, DOI 10.1017/s0305004118000506. MR4015652 [139] Charles Small, Waringâs problem mod ð, Amer. Math. Monthly 84 (1977), no. 1, 12â 25, DOI 10.2307/2318299. MR424734 [140] J. Spencer, E. Szemerédi, and W. Trotter Jr., Unit distances in the Euclidean plane, Graph theory and combinatorics (Cambridge, 1983), Academic Press, London, 1984, pp. 293â303. MR777185 [141] József Solymosi, On the number of sums and products, Bull. London Math. Soc. 37 (2005), no. 4, 491â494, DOI 10.1112/S0024609305004261. MR2143727 [142] József Solymosi, Bounding multiplicative energy by the sumset, Adv. Math. 222 (2009), no. 2, 402â408, DOI 10.1016/j.aim.2009.04.006. MR2538014 [143] J. Solymosi and Cs. D. Tóth, Distinct distances in the plane, Discrete Comput. Geom. 25 (2001), no. 4, 629â634, DOI 10.1007/s00454-001-0009-z. The Micha Sharir birthday issue. MR1838423 [144] József Solymosi and Van Vu, Distinct distances in high dimensional homogeneous sets, Towards a theory of geometric graphs, Contemp. Math., vol. 342, Amer. Math. Soc., Providence, RI, 2004, pp. 259â268, DOI 10.1090/conm/342/06146. MR2065269 [145] Sophie Stevens and Frank de Zeeuw, An improved point-line incidence bound over arbitrary fields, Bull. Lond. Math. Soc. 49 (2017), no. 5, 842â858, DOI 10.1112/blms.12077. MR3742451 [146] László A. Székely, Crossing numbers and hard ErdÅs problems in discrete geometry, Combin. Probab. Comput. 6 (1997), no. 3, 353â358, DOI 10.1017/S0963548397002976. MR1464571 [147] E. Szemerédi, On sets of integers containing no ð elements in arithmetic progression, Acta Arith. 27 (1975), 199â245, DOI 10.4064/aa-27-1-199-245. MR369312 [148] Endre Szemerédi and William T. Trotter Jr., Extremal problems in discrete geometry, Combinatorica 3 (1983), no. 3-4, 381â392, DOI 10.1007/BF02579194. MR729791 [149] Terence Tao, Some recent progress on the restriction conjecture, Fourier analysis and convexity, Appl. Numer. Harmon. Anal., BirkhÀuser Boston, Boston, MA, 2004, pp. 217â243, DOI 10.1198/106186003321335099. MR2087245
178
Bibliography
[150] Ben Green and Terence Tao, Yet another proof of Szemerédiâs theorem, An irregular mind, Bolyai Soc. Math. Stud., vol. 21, János Bolyai Math. Soc., Budapest, 2010, pp. 335â342, DOI 10.1007/978-3-642-14444-8_8. MR2815622 [151] Gábor Tardos, On distinct sums and distinct distances, Adv. Math. 180 (2003), no. 1, 275â289, DOI 10.1016/S0001-8708(03)00004-5. MR2019225 [152] B. L. van der Waerden, Beweis einer Baudetschen Vermutung, Nieuw. Arch. Wisk. (in German) 15 (1927), 212â216. [153] R. C. Vaughan and T. D. Wooley, Waringâs problem: a survey, Number theory for the millennium, III (Urbana, IL, 2000), A K Peters, Natick, MA, 2002, pp. 301â340, DOI 10.1198/004017002320256684. MR1956283 [154] André Weil, On some exponential sums, Proc. Nat. Acad. Sci. U.S.A. 34 (1948), 204â 207, DOI 10.1073/pnas.34.5.204. MR27006 [155] Wikimedia, Image taken from https://commons.wikimedia.org/w/index. php?title=File:Fano_plane.svg&oldid=357042840 [156] Wikimedia, Image taken from https://commons.wikimedia.org/wiki/File: Cantor_set_in_seven_iterations.svg [157] Thomas Wolff, Decay of circular means of Fourier transforms of measures, Internat. Math. Res. Notices 10 (1999), 547â567, DOI 10.1155/S1073792899000288. MR1692851 [158] Thomas Wolff, An improved bound for Kakeya type maximal functions, Rev. Mat. Iberoamericana 11 (1995), no. 3, 651â674, DOI 10.4171/RMI/188. MR1363209 [159] Trevor D. Wooley, Large improvements in Waringâs problem, Ann. of Math. (2) 135 (1992), no. 1, 131â164, DOI 10.2307/2946566. MR1147960 [160] Joshua Zahl, Breaking the 3/2 barrier for unit distances in three dimensions, Int. Math. Res. Not. IMRN 20 (2019), 6235â6284, DOI 10.1093/imrn/rnx336. MR4031237
Index
Additive character, 35 Additive energy, 166 Algebraic method, 31 Arithmetic progression, 138 Bézoutâs little theorem, 19 Binary operation, 8 Cap-set problem, 159 Cauchy-Schwarz inequality, 16 Cayley graph, 163 Character, 35, 81 Characteristic (field), 14 Characteristic function, 37 Chinese remainder theorem, 86, 103, 154 Completing the square, 44, 103 Convolution, 72 Direction (of line), 4 Distance graph, 121 Distance set Euclidean, 22 finite field, 28 integers mod ð, 76 Dot-product problem, 143 Dvirâs theorem, 151
Elekes-Sharir framework, 23, 111 Energy set, 112 Equivalence class, 2 Equivalence relation, 2 ErdÅs distance problem, 22 ErdÅs unit-distance problem, 23 ErdÅs-Falconer distance problem finite field, 31 integer mod ð, 76 Euclidâs algorithm, 19 Extension field, 14 Falconer distance problem, 25 Fano plane, 8 Fermatâs two-squares theorem, 33 Field, 13 Field trace, 36 Finite configuration, 105 Finite field distance problem, 29 Finite field restriction problem, 72 Fourier transform 179
180 finite field, 37 integer mod ð, 77 Freimanâs theorem, 140 Gauss sum, 44, 82, 83 General linear group, 13, 127 Graph theory, 120 adjacent, 121 chromatic number, 121, 126 connected, 121 crossing number, 121 degree, 121 directed graph, 121 finite graph, 121 incident, 121 loop, 121 multigraph, 121 path, 121 regular graph, 121 simple graph, 121 undirected graph, 121 Group, 12 Hölderâs inequality, 16 Hausdorff dimension, 25, 147 Henselâs lemma, 154 Incidence, 24 Incidence theory, 129 Integers mod ð, 2 Interpolation, 103 Inversion theorem finite field, 38 integers mod ð, 77 ð-resultant set, 94 ð-simplex, 106
Index ð-star, 125 Kakeya conjecture, 31, 146 Kloosterman sum, 57, 88 Kummer sum, 94 Lagrangeâs four-square theorem, 23 Lebesgue measure, 25 Legendre symbol, 53 Line, 4, 5 Minkowski dimension, 147 MSTD set, 166 Multiplication table problem, 138 multiplicative character, 36 Multiplicative subgroup of the integers mod ð, 12 Nondegenerate, 107 Null vector, 58 Order (of a Finite Field), 14 Orthogonal group, 13, 127 Orthogonality finite field, 37 integers mod ð, 77 Parameter counting argument, 151 Pigeonhole principle, 1 Pinned distance set, 98 Plancherelâs theorem finite field, 38 integers mod ð, 77 Polynomial ring, 11 Product set, 137 Projective plane, 4
Index Quotient set, 2, 137 Ramsey theory, 156 Regular variety, 97 Restriction theory, 70 Ring, 10 Rothâs theorem, 157 Salem set, 51 Salié sum, 88 Semerédiâs theorem, 158 Special Euclidean group, 112 Special orthogonal group, 13 Spectral gap theorem, 161 Speherical average, 42 Sphere of radius ð¡ finite field, 6, 37
181 integer mod ð, 77 Splitting field, 14 Steinhaus theorem, 27 Subfield, 14 Sum set, 137 Sum-product problem, 24, 135, 138 Szemerédi-Trotter theorem, 24, 129, 130 Triangle, 106 Triangle inequality, 16 van der Waerdenâs theorem, 157 Vanishing proposition, 151 Waringâs theorem, 152
AMS / MAA THE CARUS MATHEMATICAL MONOGRAPHS
ErdoËs asked how many distinct distances must there be in a set of n points in the plane. Falconer asked a continuous analogue, essentially asking what is the minimal Hausdorff dimension required of a compact set in order to guarantee that the set of distinct distances has positive Lebesgue measure in R . The ï¬nite ï¬eld distance problem poses the analogous question in a vector space over a ï¬nite ï¬eld. The problem is relatively new but remains tantalizingly out of reach. This book provides an accessible, exciting summary of known results. The tools used range over combinatorics, number theory, analysis, and algebra. The intended audience is graduate students and advanced undergraduates interested in investigating the unknown dimensions of the problem. Results available until now only in the research literature are clearly explained and beautifully motivated. A concluding chapter opens up connections to related topics in combinatorics and number theory: incidence theory, sum-product phenomena, Waringâs problem, and the Kakeya conjecture.
For additional information and updates on this book, visit www.ams.org/bookpages/car-37
CAR/37