Next Article in Journal
Espresso Crema Analysis with f-AnoGAN
Previous Article in Journal
An Optimized Method for Solving the Green Permutation Flow Shop Scheduling Problem Using a Combination of Deep Reinforcement Learning and Improved Genetic Algorithm
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Homothetic Covering of Crosspolytopes

1
Department of Applied Mathematics, Harbin University of Science and Technology, Harbin 150080, China
2
School of Mathematics, North University of China, Taiyuan 030051, China
*
Author to whom correspondence should be addressed.
These authors contributed equally to this work.
Mathematics 2025, 13(4), 546; https://doi.org/10.3390/math13040546
Submission received: 21 January 2025 / Revised: 1 February 2025 / Accepted: 3 February 2025 / Published: 7 February 2025
(This article belongs to the Section B: Geometry and Topology)

Abstract

:
The exact value of Γ m ( K ) , which is the least positive number γ such that a convex body K can be covered by m translates of γ K , is usually difficult to obtain. We present exact values of Γ 14 ( B 1 3 ) , Γ 11 ( B 1 4 ) , Γ 2 n ( B 1 n ) , Γ 2 n + 1 ( B 1 n ) , and Γ 2 n + 2 ( B 1 n ) , where B 1 n is the unit ball of R n endowed with the taxicab norm.

1. Introduction

A compact convex set K whose interior int K is not empty is called a convex body. Denote by ext K the set of extreme points of K, cl K the closure of K, and bd K the boundary of K. Let K n be the set of all convex bodies in R n . We denote by c ( K ) the smallest number of translates of int K needed to cover K. Concerning the least upper bound of c ( K ) for each K K n , there is a long-standing conjecture:
Conjecture 1 (Hadwiger’s covering conjecture).
For each K K n , we have
c ( K ) 2 n ;
the equality holds if and only if K is a parallelotope.
This conjecture has been studied by many authors, and it is completely solved only in the two-dimensional case (cf. [1]). M. Lassak [2] proved that c ( K ) 8 holds for each centrally symmetric convex body in K 3 . A. Prymak [3] showed that c ( K ) 14 holds for each K K 3 . More details and references about this conjecture can be found in [4,5,6]. For each K K n and each m Z + , set
Γ m ( K ) : = inf γ 0 | c i | i [ m ] R n s . t . K i [ m ] ( c i + γ K ) ,
where [ m ] = i Z | 1 i m . By standard compactness arguments, one can show that “inf” in the definition of Γ m ( K ) can be replaced with “min”. The map Γ m ( · ) : K n [ 0 , 1 ] is called the m-covering functional. A set C of m points satisfying K Γ m ( K ) K + C is called an m-optimal configuration of K.
Estimating Γ p ( K ) for special convex bodies K K n plays an important role in Chuanming Zong’s quantitative program to attack Conjecture 1; see [7]. However, it is not easy to obtain m-optimal configurations even for convex bodies like the Euclidean unit disk and simplices (cf. Figure 6 in [8,9]). Several algorithms have been introduced to find near-optimal configurations for convex bodies; see [10,11,12]. In most cases, it is more difficult to show the optimality of a nice configuration. We will find several optimal configurations for crosspolytopes and prove the optimality theoretically. Since SageMath provides facilities to handle convex polytopes, many details in our proofs can be checked by computer programs.
A convex body K K n is said to be an n-dimensional crosspolytope if there exist n linearly independent vectors v 1 , , v n such that K = conv { ± v 1 , , ± v n } . Each n-dimensional crosspolytope is affinely equivalent to
B 1 n = x = ( x 1 , x 2 , x n ) | x 1 = i [ n ] | x i | 1 ,
which is the closed unit ball of 1 n . In [13], Y. Lian showed that
Γ 10 ( B 1 3 ) = Γ 11 ( B 1 3 ) = Γ 12 ( B 1 3 ) = Γ 13 ( B 1 3 ) = 3 5 ,
and claimed that
Γ 14 ( B 1 3 ) = Γ 15 ( B 1 3 ) = Γ 16 ( B 1 3 ) = Γ 17 ( B 1 3 ) = 4 7 .
We shall show that the correct value of Γ 14 ( B 1 3 ) is 7 / 13 . For each m [ 2 n , 2 n + 2 ] , an m-optimal configuration for B 1 n is presented. The exact value of Γ 11 ( B 1 4 ) is also determined. Our methods are different from the ones used in [14].
For each x R n , denote by p i ( x ) the i-th coordinate of x. Denote by # A the cardinality of a finite set A. Let e i | i [ n ] be the canonical basis of R n . A nonempty subset A of 1 n is said to be ε-separated for some positive ε if the distance (with respect to the taxicab norm) between each pair of points in A is at least ε .
Our main results are the following:
Theorem 1.
Γ 14 ( B 1 3 ) = 7 / 13 .
Theorem 2.
Γ 2 n ( B 1 n ) = Γ 2 n + 1 ( B 1 n ) = Γ 2 n + 2 ( B 1 n ) = ( n 1 ) / n .
It follows that Γ 11 ( B 1 4 ) Γ 10 ( B 1 4 ) = 3 / 4 . In fact, the corresponding equality holds.
Theorem 3.
Γ 11 ( B 1 4 ) = 3 / 4 .

2. Homothetic Covering of B 1 3

2.1. Notations

We collect in this subsection several notations and basic results that will be used. Set
O 1 = x | p i ( x ) 0 , i [ 3 ] , O 2 = x | p 1 ( x ) 0 , p 2 ( x ) 0 , p 3 ( x ) 0 , O 3 = x | p 1 ( x ) 0 , p 2 ( x ) 0 , p 3 ( x ) 0 , O 4 = x | p 1 ( x ) 0 , p 2 ( x ) 0 , p 3 ( x ) 0 ,
O 5 = O 4 , O 6 = O 3 , O 7 = O 2 , and O 8 = O 1 . Two orthants O and O of R 3 are said to be adjacent (denoted by O O ) if the intersection of them is two-dimensional. The adjacency between orthants is collected in Table 1.
For each j [ 8 ] , set f j = ( sgn p 1 ( x ) , sgn p 2 ( x ) , sgn p 3 ( x ) ) , where x int O j . For example, we have f 1 = ( 1 , 1 , 1 ) , f 2 = ( 1 , 1 , 1 ) , f 3 = ( 1 , 1 , 1 ) , and f 4 = ( 1 , 1 , 1 ) . Let
F = ± f i | i [ 4 ]   and   F i = f F | e i | f = 1 , i [ 3 ] .
Clearly,
B 1 3 = x R 3 | x | f 1 , f F
and
F = F j F j , j [ 3 ] .
For each α R and each g R 3 { o } , set
H g , α = x | x | g = α , H g , α + = x | x | g α , and H g , α = x | x | g α .
Let G be the group generated by permutations of the standard basis vectors and reflections with respect to coordinate hyperplanes, and Q 1 , Q 2 , Q 3 , and Q 4 be the linear transforms in G with matrices
0 0 1 0 1 0 1 0 0 , 1 0 0 0 0 1 0 1 0 , 1 0 0 0 1 0 0 0 1 , and 0 1 0 1 0 0 0 0 1 ,
respectively. It can be verified that, for each γ > 0 and each m R 3 , we have
γ B 1 3 + m = x + m | x + m | f m | f + γ , f F .
Thus, for each y γ B 1 3 + m and each f F , we have
m | f + γ 2 γ + y | f .
Let P 0 = { ± 6 e 1 , ± 6 e 2 , ± 6 e 3 } , P 1 = { ( 2 , 2 , 2 ) , ( 5 , 5 , 4 ) , ( 4 , 5 , 5 ) , ( 5 , 4 , 5 ) } , and P = P 0 P 1 ( P 1 ) . We will show that P is a 14-optimal configuration of 13 B 1 3 . Set
K i = 6 e i + 7 B 1 3 , i [ 3 ] .
By (1), we have
K i = 6 e i + x R 3 | x | f 7 , f F = x + 6 e i | x R 3 and x | f 7 , f F = x R 3 | x | f 13 and x | f 1 , f F i .
Similarly,
K i = x | x | f 1 and x | f 13 , f F i .
Let k [ 8 ] and k 1 , k 2 , k 3 [ 8 ] be integers such that, for each i [ 3 ] , O k O k i and f k i differs from f k only in its i-th coordinate. Put
G k = { f k 1 , f k 2 , f k 3 } .
The following lemma can also be verified by a computer program based on SageMath; see [15].
Lemma 1.
For each k [ 8 ] , the closure of
U k : = ( 13 B 1 3 ( 7 B 1 3 + P 0 ) ) O k = ( 13 B 1 3 ( ( i [ 3 ] K i ) ( i [ 3 ] ( K i ) ) ) ) O k
is the simplex S k : = x | x | f 1 , f G k ; x | f k 13 contained in int O k .
Proof. 
It suffices to show that U k = Δ : = x | x | f > 1 , f G k ; x | f k 13 .
Let x U k . We have x | f 13 , f F . For each i [ 3 ] , there exist g k , i , h k , i F i such that x | g k , i > 1 and x | h k , i > 1 . Clearly,
x | f k i x | g k , i > 1 , i [ 3 ] .
Thus, x Δ . Therefore, U k Δ .
Clearly, Δ 13 B 1 3 int O k . Note that, for each i [ 3 ] , we have G k F i and G k F i . By (6) and (7), we have
Δ ( ( i [ 3 ] K i ) ( i [ 3 ] ( K i ) ) ) = .
Thus, Δ U k . This completes the proof. □
For each k [ 8 ] , set
T k = S k bd ( 13 B 1 3 )   and   T : = T k | k [ 8 ] .
There exists Q G such that S k = Q ( S 1 ) and T k = Q ( T 1 ) .
It can be verified that (see also [15])
E : = k [ 8 ] ext T k = { ( ± 1 , ± 6 , ± 6 ) , ( ± 6 , ± 1 , ± 6 ) , ( ± 6 , ± 6 , ± 1 ) } .
Set V : = v k | k [ 8 ] and C : = c k | k [ 8 ] , where, for each k [ 8 ] , v k is the point in
ext S k ext T k = { ( ± 1 , ± 1 , ± 1 ) } O k ,
and c k is the point in { ( ± 4 , ± 4 , ± 5 ) } T k . See Table 2 for more details.
Lemma 2.
Suppose that k [ 8 ] , m R 3 , and x ext T k ( m + γ B 1 3 ) . If f x F is given by
p i ( f x ) = sgn p i ( x ) , | p i ( x ) |   = 6 , sgn p i ( x ) , | p i ( x ) |   = 1 ,
then
m | f x γ 11 .
If γ B 1 3 + m contains another point y ext T k , then
1 γ m | f x γ 11 .
Proof. 
By (4), we have x | f x = 11 m | f x + γ , which implies (10).
Assume that | p i 0 ( x ) |   = 1 . For each y ext T k { x } , we have | p i 0 ( y ) |   = 6 and
y | f x = sgn p i 0 ( x ) p i 0 ( y ) + i [ 3 ] { i 0 } ( sgn p i ( x ) ) p i ( y ) = 1 .
By (4), we have y | f x = 1 m | f x + γ , from which (11) follows. □
Take y = ( 6 , 1 , 6 ) ext T 2 ( γ B 1 3 + m ) for example. We have f y = ( 1 , 1 , 1 ) = f 4 . From (10), it follows that m | f 4 γ 11 . By (4), we have
x | f 4 m | f 4 + γ 2 γ 11 , x γ B 1 3 + m .
If y = ( 6 , 1 , 6 ) ext T 1 ( γ B 1 3 + m ) , then f y = ( 1 , 1 , 1 ) = f 3 . By (10), we have
x | f 3 m | f 3 + γ 2 γ 11 , x γ B 1 3 + m .
We end this subsection with the following result.
Lemma 3.
Γ 14 ( B 1 3 ) 7 / 13 .
Proof. 
We only need to prove that 13 B 1 3 7 B 1 3 + P . By Lemma 1 and
k { 1 , 2 , 5 , 6 } S k = k { 3 , 4 , 7 , 8 } S k ,
it suffices to show that
k { 1 , 2 , 5 , 6 } S k 7 B 1 3 + P 1 .
Clearly,
S 1 = ( H f 1 , 7 + S 1 ) ( H f 1 , 7 S 1 ) .
It can be verified that
ext ( S 1 H f 1 , 7 + ) 7 B 1 3 + ( 5 , 5 , 4 )   and   ext ( S 1 H f 1 , 7 ) 7 B 1 3 + ( 2 , 2 , 2 ) .
Hence,
S 1 ( 7 B 1 3 + ( 2 , 2 , 2 ) ) ( 7 B 1 3 + ( 5 , 5 , 4 ) ) .
Note that
Q 1 ( S 1 ) = S 6 , Q 2 ( S 1 ) = S 5 , Q 1 ( 2 , 2 , 2 ) = Q 2 ( 2 , 2 , 2 ) = ( 2 , 2 , 2 ) , Q 1 ( 5 , 5 , 4 ) = ( 4 , 5 , 5 ) , Q 2 ( 5 , 5 , 4 ) = ( 5 , 4 , 5 ) .
These equalities and (15) show that
S 5 S 6 7 B 1 3 + { ( 4 , 5 , 5 ) , ( 5 , 4 , 5 ) , ( 2 , 2 , 2 ) } .
Clearly, Q 3 ( S 1 ) = S 2 . By Lemma 1,
S 2 = H f 2 , 13 H f 1 , 1 + H f 3 , 1 + H f 4 , 1 + .
Straightforward calculations show that (see Table 3 and [15])
ext ( S 2 H f 1 , 9 + ) 7 B 1 3 + ( 5 , 5 , 4 ) , ext ( S 2 H f 3 , 9 + ) 7 B 1 3 + ( 4 , 5 , 5 ) , ext ( S 2 H f 4 , 9 + ) 7 B 1 3 + ( 5 , 4 , 5 ) , ext ( S 2 H f 1 , 9 H f 3 , 9 H f 4 , 9 ) 7 B 1 3 + ( 2 , 2 , 2 ) .
Hence,
S 2 7 B 1 3 + { ( 2 , 2 , 2 ) , ( 5 , 5 , 4 ) , ( 4 , 5 , 5 ) , ( 5 , 4 , 5 ) } .

2.2. Auxiliary Lemmas

In the rest of Section 2, γ will always be a number in ( 0 , 7 ) . Let k , j [ 8 ] . S k ( T k , resp.) and S j ( T j , resp.) are said to be adjacent (denoted by S k S j or T k T j ) if they are contained in adjacent orthants. The following lemma is of fundamental importance for this section.
Lemma 4.
Suppose that k , j [ 8 ] are distinct, m R 3 , and S j S k . If T k ( γ B 1 3 + m ) ; then, S j ( γ B 1 3 + m ) = . In particular, a translate of γ B 1 3 intersects at most two members of T , which are adjacent.
Proof. 
Fix x T k ( γ B 1 3 + m ) . Clearly, x | f k = 13 .
Let y 13 B 1 3 O j ( γ B 1 3 + m ) . By (4) and (5), we have
y | f k m | f k + γ 2 γ + x | f k = 2 γ 13 < 1 .
Since S j S k , there are two possible cases.
Case 1. sgn p i ( y ) = sgn p i ( x ) , i [ 3 ] . Then, for each g F , we have
y | g i [ 3 ] | p i ( y ) | = y | f k < 1 .
Case 2. There exists a unique i 0 [ 3 ] such that sgn p i 0 ( x ) = sgn p i 0 ( y ) . Then, for each g sgn p i 0 ( x ) F i 0 , we have
y | g i [ 3 ] { i 0 } | p i ( y ) | sgn p i 0 ( x ) p i 0 ( y ) = y | f k < 1 .
In both cases, by Lemma 1, we have y S j .
To complete the proof, it is sufficient to note that among any three distinct orthants, there is a pair that are not adjacent. □
Lemma 5.
Let m R 3 .
(a) 
If there exists a k [ 8 ] such that T k γ B 1 3 + m , then m | f k + γ < 5 .
(b) 
If ( γ B 1 3 + m ) V , then, for each k [ 8 ] , γ B 1 3 + m contains at most two points in ext T k .
(c) 
For each k [ 8 ] , S k cannot be covered by a single translate of γ B 1 3 .
Proof. 
We only need to show (a) and (b).
(a). Take x ext T k . By Lemma 2, we have
8 < 1 γ m | f x < 4 .
Since a point is in ext T k if and only if its coordinates is a permutation of the coordinates of x, we have
sgn p 1 ( x ) p 1 ( m ) sgn p 2 ( x ) p 2 ( m ) sgn p 3 ( x ) p 3 ( m ) ( 8 , 4 ) , sgn p 1 ( x ) p 1 ( m ) sgn p 2 ( x ) p 2 ( m ) + sgn p 3 ( x ) p 3 ( m ) ( 8 , 4 ) , sgn p 1 ( x ) p 1 ( m ) + sgn p 2 ( x ) p 2 ( m ) sgn p 3 ( x ) p 3 ( m ) ( 8 , 4 ) .
It follows that
m | f k = i [ 3 ] sgn p i ( x ) p i ( m ) < 12 .
Thus,
m | f k + γ < γ 12 < 5 .
(b). Assume the contrary that there exists k [ 8 ] such that ext T k γ B 1 3 + m . Let v V ( γ B 1 3 + m ) . By (4), we have 3 v | f k m | f k + γ . By (16), m | f k + γ < 5 , which is a contradiction. □
Let T k and T j be adjacent, with the signs of their i 0 -th coordinates being opposite. Set
D k j = x ext T k | | p i 0 ( x ) | = 6 and N k j = x ext T k | | p i 0 ( x ) | = 1 .
Clearly, # D k j = # D j k = 2 and # N k j = # N j k = 1 . Let x D k j and y D j k . We write x y if | p i ( x ) | = | p i ( y ) | , i [ 3 ] . Otherwise, we write x y and we have x y 1 = ( 1 , 6 , 6 ) ( 6 , 1 , 6 ) 1 = 22 . Thus, if x y , then they cannot be covered by a single translate of γ B 1 3 .
Clearly, we have
Lemma 6.
Suppose that T k T j and B is a translate of γ B 1 3 .
(a) 
# ( B ( D k j D j k ) ) 2 and # ( B ( ext T k ext T j ) ) 4 .
(b) 
If # ( B ext T k ) = # ( B ext T j ) = 2 , then there exist x D k j B and y D j k B such that x y , and x N k j B and y N j k B .
Lemma 7.
Let m R 3 . Suppose that T k T j . If c j γ B 1 3 + m , then
(a) 
( γ B 1 3 + m ) D k j = ,
(b) 
# ( ( γ B 1 3 + m ) k [ 8 ] { j } ext T k ) 1 = # N k j .
Proof. 
(a). By applying a transformation in G if necessary, we may assume that k { 2 , 3 , 4 } and j = 1 . Suppose that the i 0 -th coordinates of T k and T 1 are with the opposite sign. Let x D k 1 ( γ B 1 3 + m ) . Then, one of the coordinates of x is 1 and p i 0 ( x ) = 6 . By Lemma 2, one coordinate of f x (see (9) for the definition) is 1 and the other two are 1. It follows that c 1 | f x 3 > 2 γ 11 . If c 1 γ B 1 3 + m , then, by (4), we have 2 γ 11 < c 1 | f x m | f x + γ , which contradicts (10). Hence, ( γ B 1 3 + m ) D k 1 = .
(b). It follows from Lemma 4 and Lemma 7a. □
Suppose that T k T j . Lemma 7 shows that a translate of γ B 1 3 containing c j cannot contain any point in D k j . Conversely, a translate of γ B 1 3 containing a point in D k j cannot contain c j . Particularly, in the case of Lemma 6b, ( γ B 1 3 + m ) C = .
Lemma 8.
Let B be a translate of γ B 1 3 .
(a) 
If { c 1 , ( 1 , 6 , 6 ) , ( 5 , 6 , 2 ) } B , then B C = { c 1 } .
(b) 
If { c 2 , ( 1 , 6 , 6 ) , ( 5 , 6 , 2 ) } B , then B C = { c 2 } .
Proof. 
By symmetry, we only need to prove (a). By Lemma 4, B C { c 1 , c 2 , c 3 , c 4 } . Lemma 7 and the fact ( 1 , 6 , 6 ) D 12 D 13 show that c 2 , c 3 B . The distance from c 4 = ( 4 , 4 , 5 ) to ( 5 , 6 , 2 ) is 14. Thus, c 4 B . □
Lemma 9.
Let m R 3 .
(a) 
If { ( 1 , 6 , 6 ) , ( 6 , 1 , 6 ) , ( 6 , 6 , 1 ) } γ B 1 3 + m , then v 1 γ B 1 3 + m .
(b) 
If { ( 1 , 6 , 6 ) , ( 6 , 1 , 6 ) , v 1 } γ B 1 3 + m , then { ( 5 , 6 , 2 ) , ( 6 , 5 , 2 ) } γ B 1 3 + m .
(c) 
If { ( 6 , 1 , 6 ) , ( 6 , 6 , 1 ) , v 1 } γ B 1 3 + m , then c 2 γ B 1 3 + m .
Proof. 
(a). By the hypothesis and (4), we have
1 · ( 6 p 1 ( m ) ) + ( 1 ) · ( 1 p 2 ( m ) ) + ( 1 ) · ( 6 p 3 ( m ) ) γ , ( 1 ) · ( 1 p 1 ( m ) ) + 1 · ( 6 p 2 ( m ) ) + ( 1 ) · ( 6 p 3 ( m ) ) γ , 1 · ( 6 p 1 ( m ) + 1 · ( 6 p 2 ( m ) ) + 1 · ( 1 p 3 ( m ) ) γ .
Adding these inequalities, we have
p 1 ( m ) p 2 ( m ) + p 3 ( m ) + 35 3 γ   and   m | f 2 + γ 4 γ 35 < 7 .
By v 1 | f 2 = 1 > m | f 2 + γ and (4), v 1 γ B 1 3 + m .
(b). By the hypothesis and (4), we have
1 · ( 6 p 1 ( m ) ) + ( 1 ) · ( 1 p 2 ( m ) ) + 1 · ( 6 p 3 ( m ) ) γ , ( 1 ) · ( 1 p 1 ( m ) ) + 1 · ( 6 p 2 ( m ) ) + 1 · ( 6 p 3 ( m ) ) γ , ( 1 ) · ( 1 p 1 ( m ) ) + ( 1 ) · ( 1 p 2 ( m ) ) + ( 1 ) · ( 1 p 3 ( m ) ) γ .
It follows that
p 1 ( m ) + p 2 ( m ) p 3 ( m ) + 19 3 γ   and   m | f 2 + γ 4 γ 19 < 7 .
By ( 5 , 6 , 2 ) | f 2 = ( 6 , 5 , 2 ) | f 2 = 9 > m | f 2 + γ and (4), we have ( 5 , 6 , 2 ) , ( 6 , 5 , 2 ) γ B 1 3 + m .
(c). By the hypothesis and (4), we have
1 · ( 6 p 1 ( m ) ) + ( 1 ) · ( 1 p 2 ( m ) ) + ( 1 ) · ( 6 p 3 ( m ) ) γ , 1 · ( 6 p 1 ( m ) ) + 1 · ( 6 p 2 ( m ) ) + 1 · ( 1 p 3 ( m ) ) γ , ( 1 ) · ( 1 p 1 ( m ) ) + ( 1 ) · ( 1 p 2 ( m ) ) + 1 · ( 1 p 3 ( m ) ) γ .
Therefore,
p 1 ( m ) + p 2 ( m ) p 3 ( m ) + 23 3 γ   and   m | f 3 + γ 4 γ 23 < 5 .
By c 2 | f 3 = 5 > m | f 3 + γ and (4), we have c 2 γ B 1 3 + m . □
Lemma 10.
Let m R 3 . If ( γ B 1 3 + m ) ext ( 13 B 1 3 ) , then
( γ B 1 3 + m ) ( k [ 8 ] S k ) = .
Proof. 
By symmetry, we only need to consider the case when 13 e 1 γ B 1 3 + m . By (5), we have
m | f + γ 2 γ + 13 e 1 | f = 2 γ 13 < 1 , f F 1 .
Let x γ B 1 3 + m . By (4), we have
x | f m | f + γ < 1 , f F 1 .
By (6), (7), and Lemma 1, we have x k [ 8 ] S k . □
In the rest of this section, M R 3 is a set satisfying 13 B 1 3 γ B 1 3 + M ,
M 0 = m M | ( γ B 1 3 + m ) ext ( 13 B 1 3 ) .
Let x , y ext ( 13 B 1 3 ) be distinct. Then, x y 1 = 26 , which implies that # M 0 6 . Set M 1 = M M 0 . By Lemma 10,
k = 1 8 S k γ B 1 3 + M 1 .
Set M 1 = γ B 1 3 + m | m M 1 .
Recall that A 2-regular graph is a graph where each vertex is connected to exactly two other vertices.
Lemma 11.
Let T T and M 1 M 1 be two sets satisfying # T = # M 1 = k [ 8 ] , G be the bipartite graph with bipartitions T and M 1 , where T T and B M 1 are connected by an edge if and only if T B . If
(a) 
T M 1 ,
(b) 
no member of T can be covered by a single translate of M 1 ,
(c) 
and G is 2-regular,
then, for each pair of distinct members T i and T j of T , no member of M 1 can contain two points in ext T i and two points in ext T j simultaneously.
Proof. 
By the hypothesis, each member of T is covered by exactly two members of M 1 , and each member of M 1 intersects exactly two members of T .
Assume the contrary that there exist B 1 M 1 and distinct T i , T j T such that B 1 contains two points in ext T i and two points in ext T j . By Lemma 4, T i T j . Without loss of generality, we may assume that i = 1 , j = 2 . By Lemma 6, we may also assume (exchange the first two coordinates if necessary) that
{ ( 6 , 6 , 1 ) , ( 6 , 1 , 6 ) , ( 6 , 1 , 6 ) , ( 6 , 6 , 1 ) } B 1 .
Let B 2 and B 3 be members of M 1 such that T 1 B 1 B 2 and T 2 B 1 B 3 . See Figure 1. By Lemma 7,
C B 1 =   and   B 2 B 3 .
Note that { ( 6 , 1 , 6 ) , ( 6 , 1 , 6 ) } B 1 . By (10), we have B 1 H f 4 , 3 H f 3 , 3 .
Then, T 1 H f 4 , 3 + B 2 and T 2 H f 3 , 3 + B 3 . It can be verified that (see also [15])
ext ( T 1 H f 4 , 3 + ) = { ( 5 , 6 , 2 ) , ( 5 , 2 , 6 ) , ( 1 , 6 , 6 ) } B 2 , ext ( T 2 H f 3 , 3 + ) = { ( 5 , 6 , 2 ) , ( 5 , 2 , 6 ) , ( 1 , 6 , 6 ) } B 3 .
From Lemma 8, it follows that
C B 2 = { c 1 }   and   C B 3 = { c 2 } .
By Lemma 7, B 1 B 2 B 3 contains at most eight points of T T ext T . Clearly, k > 3 . Since ( 3 k 8 ) / ( k 3 ) > 3 , there exists B 4 M 1 that contains two points in ext T p and two points in ext T q , where p , q { 1 , 2 } . Let B 5 and B 6 be two members of M 1 such that T p B 4 B 5 and T q B 4 B 6 . As above, we have B 5 B 6 . By (18), we have
B 5 , B 6 { B 1 , B 2 , B 3 , B 4 }   and   k 6 .
By Lemma 7 again, B 5 B 6 can cover at most two points in
( T T ext T ) ( B 1 B 2 B 3 B 4 ) .
Therefore, at least 3 k ( 8 + 6 + 2 ) = 3 k 16 points in T T ext T remain uncovered, while the number of the remaining translates of γ B 1 3 is k 6 . Thus, k > 6 .
If k = 7 , then there exists a member of M 1 containing five points in E, which contradicts Lemma 6. If k = 8 , then the remaining two members B 7 and B 8 of M 1 contain eight points of { T T } ext T . By Lemma 7, B 7 B 8 contains no point in C, which is a contradiction. □
Lemma 12.
If # M 1 = 8 , then there exist m 0 M 1 and k 0 [ 8 ] such that T k 0 γ B 1 3 + m 0 .
Proof. 
Let G be the bipartite graph defined as in Lemma 11 with bipartitions T and M 1 . By Lemma 4, the total number of edges is at most 16.
Assume the contrary that, for each m M 1 and k [ 8 ] , T k ( γ B 1 3 + m ) . Then, for each k [ 8 ] , the degree of T k in G is at least two. Thus, G is 2-regular.
Let
N = max # ( ( γ B 1 3 + m ) ( i [ 8 ] ext T k ) ) | m M 1 .
Clearly, N 24 / 8 = 3 . By Lemmas 4 and 6, N 4 . By Lemmas 6 and 11, we only need to consider the case when N = 3 . For each m M 1 , we have
( E ( γ B 1 3 + m ) ) ( γ B 1 3 + M 1 { m } ) = .
Then, for each m M 1 , there exists a unique { k , j } [ 8 ] such that γ B 1 3 + m contains one point in ext T k and two points in ext T j , where T j T k .
Let B 1 and B 2 be two members of M 1 such that T 1 B 1 B 2 . Without loss of generality, we may assume that
# ( B 2 ext T 1 ) = 1   and   # ( B 2 ext T 2 ) = 2 .
We claim that v 1 B 1 B 2 . Otherwise, there exists B in M 1 such that v 1 S 1 B . By Lemma 4, B ( k { 5 , 6 , 7 , 8 } T k ) = . From (19), it follows that # ( B k { 2 , 3 , 4 } ext T k ) = 3 , which contradicts Lemma 4 since T 2 , T 3 , and T 4 are pairwise nonadjacent.
There are two cases.
Case a. ( 6 , 6 , 1 ) B 2 . Then, D 21 B 2 . It follows that B 2 T 1 = N 12 = { ( 6 , 6 , 1 ) } . By Lemma 9a, v 1 B 2 . Thus,
{ ( 1 , 6 , 6 ) , ( 6 , 1 , 6 ) , v 1 } B 1 .
By Lemma 9b, ( 5 , 6 , 2 ) B 1 . Since | | ( 5 , 6 , 2 ) ( 6 , 1 , 6 ) | | 1 = 14 , ( 5 , 6 , 2 ) B 2 . Clearly, ( 5 , 6 , 2 ) = ( 1 / 5 ) ( 1 , 6 , 6 ) + ( 4 / 5 ) ( 6 , 6 , 1 ) T 1 , which is a contradiction.
Case b. ( 6 , 6 , 1 ) B 2 . Without loss of generality, we may assume (exchange the first two coordinates if necessary) that ( 6 , 1 , 6 ) B 2 . By Lemma 6, ( 1 , 6 , 6 ) B 2 . Then, either ( 6 , 1 , 6 ) or ( 6 , 6 , 1 ) is contained in B 2 .
If { ( 6 , 1 , 6 ) , ( 6 , 6 , 1 ) , ( 6 , 1 , 6 ) } B 2 , then, by ( 6 , 1 , 6 ) B 2 and Lemma 7, c 2 B 2 . By Lemma 7b and (19), there exists a member of M 1 containing c 2 that contains at most two points in E, which is a contradiction.
Now, assume that { ( 6 , 1 , 6 ) , ( 6 , 6 , 1 ) , ( 6 , 6 , 1 ) } B 2 . We claim that v 1 B 2 . Otherwise, (20) holds, which yields a contradiction as in Case a. From Lemma 9c, it follows that c 2 B 2 , which contradicts Lemma 7b and (19). □
Corollary 1.
If # M 1 = 8 , then there exist m M 1 and distinct k 0 , j 0 [ 8 ] such that
T k 0 γ B 1 3 + m   and   T j 0 ( γ B 1 3 + m ) .
Proof. 
Set
I = k [ 8 ] | m M 1 s . t . T k γ B 1 3 + m , J = m M 1 | k [ 8 ] s . t . T k γ B 1 3 + m .
Let
T 1 = T k | k I , T 2 = T k | k [ 8 ] I , M 1 1 = γ B 1 3 + m | m J ,   and   M 1 2 = γ B 1 3 + m | m M 1 J .
Assume the contrary that, for each m M 1 and each k [ 8 ] satisfying T k γ B 1 3 + m , we have
( i [ 8 ] { k } T i ) ( γ B 1 3 + m ) = .
Thus, # I # J .
Considering the bipartite graph G defined as in Lemma 11 with bipartitions T and M 1 . Each member of M 1 1 has degree 1 in G and, by Lemma 4, each member of M 1 2 has degree at most 2. Thus, the total number L of edges of G satisfies
L 2 ( 8 # J ) + # J = 16 # J .
Each member of T 1 has degree at least 1 and each member of T 2 has degree at least 2. Therefore, we have L 16 # I . It follows that L = 16 # I = 16 # J , and the subgraph G of G with bipartitions T 2 and M 1 2 is 2-regular.
By Lemma 5, we have [ 8 ] I . By Lemma 12, there exist k I and m J such that T k γ B 1 3 + m . From Lemma 5b, it follows that v k γ B 1 3 + m . Then, there exists an m M 1 such that v k γ B 1 3 + m . By Lemma 4, γ B 1 3 + m can only intersect with the three members of T that are adjacent to T k . Note that these three members of T are pairwise nonadjacent. From Lemma 5b and Lemma 4, it follows that γ B 1 3 + m contains at most two points in E ext T k and m M 1 J .
Clearly, ( # E 3 # I 2 ) / ( 8 # I 1 ) > 3 . By the pigeon hole principle and Lemma 6, there exist m 1 M 1 J and k 1 , k 2 [ 8 ] I such that γ B 1 3 + m 1 contains two points in ext T k 1 and two points in ext T k 2 , which contradicts Lemma 11. □

2.3. The Proof of Theorem 1

Suppose the contrary that there exists a set M R 3 with # M = 14 such that 13 B 1 3 γ B 1 3 + M . By (17), we have k = 1 8 S k γ B 1 3 + M 1 , where # M 1 8 .
By Corollary 1, without loss of generality, we may assume that there exists m 1 M 1 such that
T 1 γ B 1 3 + m 1   and   T 2 ( γ B 1 3 + m 1 ) .
Lemma 13.
conv { ( 3 , 1 , 3 ) , ( 1 , 3 , 3 ) , ( 3 , 3 , 1 ) , v 1 } S 1 ( γ B 1 3 + m 1 ) .
Proof. 
Take x T 2 ( γ B 1 3 + m 1 ) , then m 1 | f 2 + γ 13 = x | f 2 , which is equivalent to
p 1 ( m 1 ) p 2 ( m 1 ) + p 3 ( m 1 ) γ 13 .
By T 1 γ B 1 3 + m 1 , we have
( 6 p 1 ( m 1 ) ) ( 1 p 2 ( m 1 ) ) + ( 6 p 3 ( m 1 ) ) γ , ( 1 p 1 ( m 1 ) ) + ( 6 p 2 ( m 1 ) ) + ( 6 p 3 ( m 1 ) ) γ .
It follows that m 1 | f 1 + 22 3 γ 13 . Hence, m 1 | f 1 + γ < 7 .
By (4), we have
S 1 ( γ B 1 3 + m 1 ) S 1 x | x | f 1 < 7 .
It can be verified that (see also [15])
conv { ( 3 , 1 , 3 ) , ( 1 , 3 , 3 ) , ( 3 , 3 , 1 ) , v 1 } S 1 H f 1 , 7 S 1 ( γ B 1 3 + m 1 ) .
Set
a 2 = ( 6 , 5 , 2 ) , a 3 = ( 6 , 1 , 6 ) , a 4 = ( 1 , 6 , 6 ) , a 5 = ( 4 , 3 , 6 ) , a 6 = ( 3 , 4 , 6 ) , a 7 = ( 2 , 6 , 5 ) , a 8 = ( 6 , 4 , 3 ) ,
and A = a k | k [ 8 ] { 1 } . It can be verified that a k T k , k [ 8 ] { 1 } , and A is 14-separated. For each k [ 8 ] { 1 } , there exists m k M 1 such that a k γ B 1 3 + m k . Clearly, if k , j [ 8 ] { 1 } are distinct, then m k and m j are distinct. By | | a 2 ( 1 , 6 , 6 ) | | 1 = 14 , we have a 2 γ B 1 3 + m 1 . By Lemma 4, m 1 m k , k [ 8 ] { 1 } . Hence, # M 1 = 8 and M 1 = m k | k [ 8 ] . Set
B k = γ B 1 3 + m k , k [ 8 ] .
Put
A = Q 4 ( A ) = { a 3 , a 4 , a 5 , a 6 } { a 7 , a 8 , a 2 } ,
where a 7 = Q 4 ( a 7 ) = ( 6 , 2 , 5 ) T 7 , a 8 = Q 4 ( a 8 ) = ( 4 , 6 , 3 ) T 8 , and a 2 = Q 4 ( a 2 ) = ( 5 , 6 , 2 ) T 2 . Since Q 4 is a linear isometry on 1 3 , A is also 14-separated.
Clearly, if i , k [ 8 ] are distinct and T k B i , then by Lemma 4, T i T k .
Since A is 14-separated, a 2 B k , k [ 8 ] { 1 , 2 } . From | | a 2 ( 6 , 1 , 6 ) | | 1 = 14 , it follows that a 2 B 1 . Thus,
a 2 B 2 ( k [ 8 ] { 2 } B k ) .
On the one hand, neither T 1 nor T 2 is adjacent to any of T 7 and T 8 (see Table 1); thus, a 7 , a 8 B 1 B 2 . On the other hand, since A is 14-separated, a 7 , a 8 B k , k { 3 , 4 , 5 , 6 } . Thus, a 7 , a 8 B 7 B 8 . Hence, either
{ a 7 , a 7 } B 7   and   { a 8 , a 8 } B 8 ,
or
{ a 8 , a 7 } B 7   and   { a 8 , a 7 } B 8 .
By Lemma 4 and Table 1, T 2 B 1 B 2 B 5 B 6 . Since ( 1 , 6 , 6 ) , ( 6 , 1 , 6 ) B 1 and
| | ( 5 , 2 , 6 ) ( 1 , 6 , 6 ) | | = | | ( 6 , 2 , 5 ) ( 1 , 6 , 6 ) | | = 20 , | | ( 2 , 5 , 6 ) ( 6 , 1 , 6 ) | | = | | ( 2 , 6 , 5 ) ( 6 , 1 , 6 ) | | = 20 ,
we have
{ ( 5 , 2 , 6 ) , ( 6 , 2 , 5 ) , ( 2 , 5 , 6 ) , ( 2 , 6 , 5 ) } ( B 2 B 5 B 6 ) T 2 .
Lemma 14.
( 5 , 2 , 6 ) , ( 6 , 2 , 5 ) B 6 .
Proof. 
By Lemma 4 and Table 1, we obtain all possible locations of points listed in Table 4.
Suppose the contrary that one of the two points ( 5 , 2 , 6 ) and ( 6 , 2 , 5 ) is in B 6 . By
| | ( 5 , 2 , 6 ) p 1 | | 1 = | | ( 6 , 2 , 5 ) p 1 | | 1 = 16 ,
p 1 B 6 . By | | a 4 p 1 | | 1 = 16 and | | a 2 p 1 | | 1 = 14 , p 1 B 2 B 4 . Thus, p 1 B 8 T 6 . Since B 8 intersects both T 8 and T 6 , a 7 B 7 and a 8 B 8 .
Note that p 2 T 8 . By | | p 1 p 2 | | 1 = | | a 7 p 2 | | 1 = 14 , p 2 B 7 B 8 . Since B 6 intersects both T 2 and T 6 , p 2 B 6 . Thus, p 2 B 5 T 8 .
By Table 4 and | | p 2 p 3 | | 1 = | | a 8 p 3 | | 1 = | | a 3 p 3 | | 1 = 14 , we have p 3 B 2 .
By | | ( 5 , 2 , 6 ) p 4 | | 1 = | | ( 6 , 2 , 5 ) p 4 | | 1 = 18 , p 3 B 2 , | | p 3 p 4 | | 1 = 22 , and | | a 8 p 4 | | 1 = 14 , we have p 4 B 4 .
Clearly, p 5 O 1 . By Lemma 13, p 5 B 1 . Moreover, p 4 B 4 T 6 , p 3 B 2 T 5 , and neither O 6 nor O 5 is adjacent to O 1 . By Lemma 4, p 5 B 2 B 4 . Thus, p 5 B 3 .
Since p 6 T 3 , B 1 T 1 , and B 1 T 2 , we have p 6 B 1 . By | | p 5 p 6 | | 1 = 14 and | | a 7 p 6 | | 1 = 18 , p 6 B 3 B 7 . By p 2 B 5 T 8 , B 5 intersects both T 5 and T 8 , we have p 6 B 5 , a contradiction. □
Corollary 2.
( 2 , 5 , 6 ) , ( 2 , 6 , 5 ) B 5 .
Proof. 
Since 13 B 1 3 γ B 1 3 + M , we have 13 B 1 3 γ B 1 3 + Q 4 ( M ) . From T 2 = Q 4 ( T 2 ) and T 1 = Q 4 ( T 1 ) , it follows that
T 1 γ B 1 3 + Q 4 ( m 1 ) and ( γ B 1 3 + Q 4 ( m 1 ) ) T 2 .
Thus, Q 4 ( M ) is a configuration satisfying (22), where m 1 is replaced with Q 4 ( m 1 ) . Let m k be the point in Q 4 ( M ) satisfying a k γ B 1 3 + m k . Then, since
a 6 = Q 4 ( a 5 ) γ B 1 3 + Q 4 ( m 5 ) ,
we have m 6 = Q 4 ( m 5 ) . Applying Lemma 14 on Q 4 ( M ) , we have ( 5 , 2 , 6 ) , ( 6 , 2 , 5 ) γ B 1 3 + Q 4 ( m 5 ) . It follows that ( 2 , 5 , 6 ) , ( 2 , 6 , 5 ) B 5 . □
By | | a 8 p 3 | | 1 = | | a 3 p 3 | | 1 = 14 and Table 4, we have p 3 B 3 B 8 .
Lemma 15.
( 4 , 3 , 6 ) B 2 .
Proof. 
By Table 1, we only need to show that ( 4 , 3 , 6 ) B 1 B 5 B 6 .
By ( 6 , 1 , 6 ) T 1 and | | ( 4 , 3 , 6 ) ( 6 , 1 , 6 ) | | 1 = 16 , we have ( 4 , 3 , 6 ) B 1 .
Next, we claim that ( 4 , 3 , 6 ) B 5 . Otherwise, since | | ( 4 , 3 , 6 ) p 3 | | 1 = 14 , we would have p 3 B 2 . By | | ( 2 , 5 , 6 ) p 3 | | 1 = | | ( 2 , 6 , 5 ) p 3 | | 1 = 18 , (25), and Corollary 2, we have ( 2 , 5 , 6 ) , ( 2 , 6 , 5 ) B 6 .
By Table 1, ( 6 , 1 , 6 ) is contained in B 2 B 4 B 6 B 8 . By
| | ( 6 , 1 , 6 ) a 4 | | 1 = 22 , | | ( 6 , 1 , 6 ) a 2 | | 1 = 20 , and | | ( 6 , 1 , 6 ) ( 2 , 6 , 5 ) | | 1 = 14 ,
we have ( 6 , 1 , 6 ) B 8 . Hence, B 8 intersects both T 6 and T 8 . By Lemma 4, B 8 T 7 = , and therefore a 7 B 7 . Since | | a 7 ( 1 , 6 , 6 ) | | 1 = 20 , we have ( 1 , 6 , 6 ) B 7 . In the current situation, B 5 intersects both T 5 and T 2 , and B 6 intersects both T 6 and T 2 . By Lemma 4 and Table 1, ( 1 , 6 , 6 ) B 8 T 8 .
By Table 1, | | ( 6 , 6 , 1 ) ( 1 , 6 , 6 ) | | 1 = 22 , and
| | ( 6 , 6 , 1 ) a 2 | | 1 = | | ( 6 , 6 , 1 ) ( 2 , 5 , 6 ) | | 1 = 14 ,
we have ( 6 , 6 , 1 ) B 4 T 6 .
By Table 1 and Lemma 13, p 5 B k S 1 for some k { 2 , 3 , 4 } . Since B 4 intersects T 6 , B 2 intersects T 5 , O 5 O 1 , and O 6 O 1 , we have, by Lemma 4, p 5 B 3 ( S 1 T 1 ) .
Clearly, p 6 T 3 . Since B 1 intersects both T 1 and T 2 , p 6 B 1 . By
| | p 5 p 6 | | 1 = 14   and   | | a 7 p 6 | | 1 = 18 ,
p 6 B 5 T 3 . Hence, B 5 intersects T 2 , T 3 , and T 5 , which contradicts Lemma 4.
In the rest of this section, we show that ( 4 , 3 , 6 ) B 6 . Otherwise, by
| | p 1 ( 4 , 3 , 6 ) | | 1 = | | a 2 p 1 | | 1 = 14 ,
and | | a 4 p 1 | | 1 = 16 , we have p 1 B 8 . Since B 8 intersects both T 6 and T 8 , we have a 7 B 7 . By
| | p 2 p 1 | | 1 = 14   and   | | a 7 p 2 | | 1 = 14 ,
we have p 2 B 7 B 8 . By ( 4 , 3 , 6 ) B 6 T 2 and Lemma 4, p 2 B 6 . Thus, p 2 B 5 T 8 . By Table 4 and
| | p 2 p 3 | | 1 = | | a 3 p 3 | | 1 = | | a 8 p 3 | | 1 = 14 ,
we have p 3 B 2 T 5 . Since B 2 intersects both T 2 and T 5 , ( 5 , 6 , 2 ) B 2 . By Table 4,
| | ( 4 , 3 , 6 ) p 4 | | 1 = 16 ,   and   | | a 8 p 4 | | 1 = 14 ,
we have p 4 B 4 T 6 . Since B 4 intersects T 6 and O 6 O 1 , and B 2 intersects T 5 and O 5 O 1 , p 5 B 2 B 4 . By Lemma 13, p 5 B 1 . Thus, p 5 B 3 ( S 1 T 1 ) .
Clearly, p 6 T 3 . Since B 1 intersects both T 1 and T 2 , ( 6 , 5 , 2 ) B 1 . By
| | p 5 p 6 | | 1 = 14 , | | a 7 p 6 | | 1 = 18 ,
and Lemma 4, we have p 6 B 5 T 3 , contradicting the fact that B 5 intersects both T 5 and T 8 . □
Lemma 16.
If { ( 5 , 2 , 6 ) , ( 6 , 2 , 5 ) } B 5 , then { ( 5 , 2 , 6 ) , ( 6 , 2 , 5 ) } B 2 .
If { ( 2 , 5 , 6 ) , ( 2 , 6 , 5 ) } B 6 , then { ( 2 , 5 , 6 ) , ( 2 , 6 , 5 ) } B 2 .
Proof. 
We prove the first statement and the other one can be obtained by arguments as in the proof of Corollary 2.
By Lemma 14, either ( 5 , 2 , 6 ) B 2 or ( 6 , 2 , 5 ) B 2 .
Case a  ( 5 , 2 , 6 ) B 2 . By (4), for each f F 3 , we have
m 2 | f + γ ( 6 , 2 , 5 ) | f ( 5 , 2 , 6 ) | f ( 6 , 2 , 5 ) | f = 1 , 0 , 1 | f 0 .
By (4) and a 2 = ( 6 , 5 , 2 ) B 2 , for each f F 3 , we have
m 2 | f + γ ( 6 , 2 , 5 ) | f ( 6 , 5 , 2 ) | f ( 6 , 2 , 5 ) | f = ( 0 , 3 , 3 ) | f 0 .
By (4) again, we have { ( 5 , 2 , 6 ) , ( 6 , 2 , 5 ) } B 2 .
Case b  ( 6 , 2 , 5 ) B 2 . By (4), we have
m 2 | f + γ ( 5 , 2 , 6 ) | f ( 6 , 2 , 5 ) | f ( 5 , 2 , 6 ) | f = ( 1 , 0 , 1 ) | f 0 , f F 1 .
From ( 4 , 3 , 6 ) B 2 and (4), it follows that
m 2 | f + γ ( 5 , 2 , 6 ) | f ( 4 , 3 , 6 ) | f ( 5 , 2 , 6 ) | f = 1 , 1 , 0 | f 0 , f F 1 .
By (4), we have { ( 5 , 2 , 6 ) , ( 6 , 2 , 5 ) } B 2 . □
If { ( 5 , 2 , 6 ) , ( 6 , 2 , 5 ) } B 5 T 2 , then by
| | ( 5 , 2 , 6 ) ( 6 , 6 , 1 ) | | = 14   and   | | ( 6 , 2 , 5 ) ( 1 , 6 , 6 ) | | = 14 ,
we have ( 6 , 6 , 1 ) , ( 1 , 6 , 6 ) B 5 . Thus,
{ ( 6 , 6 , 1 ) , ( 1 , 6 , 6 ) } ( B 2 B 3 B 8 ) T 5 .
If { ( 2 , 5 , 6 ) , ( 2 , 6 , 5 ) } B 6 T 2 , then one can verify that
{ ( 6 , 6 , 1 ) , ( 6 , 1 , 6 ) } ( B 2 B 4 B 8 ) T 6 .
Lemma 17.
We have
{ ( 2 , 5 , 6 ) , ( 2 , 6 , 5 ) , ( 5 , 2 , 6 ) , ( 6 , 2 , 5 ) } B 2 .
Proof. 
We show that { ( 5 , 2 , 6 ) , ( 6 , 2 , 5 ) } B 2 . The inclusion { ( 2 , 5 , 6 ) , ( 2 , 6 , 5 ) } B 2 can be proved by similar arguments as in the proof of Corollary 2.
By Lemma 16, we only need to show that { ( 5 , 2 , 6 ) , ( 6 , 2 , 5 ) } B 5 . By Lemmas 4 and 13, ( 3 , 1 , 3 ) B 2 B 3 B 4 . Thus, it suffices to show that, if { ( 5 , 2 , 6 ) , ( 6 , 2 , 5 ) } is a subset of B 5 , then ( 3 , 1 , 3 ) B 2 B 3 B 4 .
First, we show that ( 3 , 1 , 3 ) B 2 . Otherwise, by
| | ( 2 , 5 , 6 ) ( 3 , 1 , 3 ) | | 1 = | | ( 2 , 6 , 5 ) ( 3 , 1 , 3 ) | | 1 = 14 ,
we have ( 2 , 5 , 6 ) , ( 2 , 6 , 5 ) B 2 T 2 . By Corollary 2, we have { ( 2 , 5 , 6 ) , ( 2 , 6 , 5 ) } B 6 . By Lemma 4 and the facts ( 3 , 1 , 3 ) B 2 S 1 , O 5 O 1 , and O 6 O 1 , we have
B 2 T 5 = B 2 T 6 = ,
which, together with | | a 4 ( 6 , 1 , 6 ) | | 1 = 22 and (27), shows that ( 6 , 1 , 6 ) B 8 T 6 .
Similarly, by | | ( 6 , 2 , 5 ) ( 1 , 6 , 6 ) | | 1 = 14 , | | a 3 ( 1 , 6 , 6 ) | | 1 = 22 , and (26), we have ( 1 , 6 , 6 ) B 8 T 5 . It follows that B 8 intersects both T 5 and T 6 , which are not adjacent, which is a contradiction.
Next, we show that ( 3 , 1 , 3 ) B 3 . Suppose that this is not true. From O 5 O 1 and Lemma 4, it follows that T 5 B 3 = . Thus, ( 6 , 6 , 1 ) B 3 . By
| | ( 6 , 6 , 1 ) a 2 | | 1 = 14 ,
we have ( 6 , 6 , 1 ) B 2 . By (26), we have ( 6 , 6 , 1 ) B 8 , which is impossible since | | ( 6 , 6 , 1 ) a 8 | | 1 = 16 .
In the rest of this section, we show that ( 3 , 1 , 3 ) B 4 . Otherwise, by
| | ( 3 , 1 , 3 ) ( 6 , 5 , 2 ) | | 1 = | | ( 3 , 1 , 3 ) ( 5 , 6 , 2 ) | | 1 = 14 ,
we have ( 6 , 5 , 2 ) , ( 5 , 6 , 2 ) B 4 . By Table 1, ( 5 , 6 , 2 ) B 1 B 6 B 7 . Since B 1 intersects both T 1 and T 2 and | | a 7 ( 5 , 6 , 2 ) | | 1 = 18 , we have ( 5 , 6 , 2 ) B 6 T 4 . By
| | a 4 ( 6 , 1 , 6 ) | | 1 = 22 , | | a 2 ( 6 , 1 , 6 ) | | 1 = 20 , and | | ( 5 , 6 , 2 ) ( 6 , 1 , 6 ) | | 1 = 14 ,
we have ( 6 , 1 , 6 ) B 8 T 6 . By (26),
| | a 2 ( 1 , 6 , 6 ) | | 1 = 20 , and | | a 3 ( 1 , 6 , 6 ) | | 1 = 22 ,
we have ( 1 , 6 , 6 ) B 8 T 5 . Then, B 8 intersects both T 5 and T 6 , which are not adjacent, which is a contradiction. □
Corollary 3.
We have
{ ( 3 , 1 , 3 ) , p 5 } B 3 B 4 .
Proof. 
By Table 1 and Lemma 13, these two points are in B 2 B 3 B 4 . By Lemma 17 and | | ( 5 , 2 , 6 ) p 5 | | 1 = | | ( 2 , 5 , 6 ) ( 3 , 1 , 3 ) | | 1 = 14 , we have p 5 , ( 3 , 1 , 3 ) B 2 . □
Lemma 18.
We have ( 3 , 3 , 1 ) B 2 .
Proof. 
By Table 1 and Lemma 13, ( 3 , 3 , 1 ) B 2 B 3 B 4 . We show that ( 3 , 3 , 1 ) B 3 , and ( 3 , 3 , 1 ) B 4 can be proved by similar arguments as in the proof of Corollary 2.
Otherwise, ( 3 , 3 , 1 ) B 3 S 1 . By Lemma 4, B 3 may only intersect T 1 and T 3 . Since
| | ( 2 , 6 , 5 ) ( 3 , 3 , 1 ) | | 1 = | | ( 2 , 5 , 6 ) ( 3 , 3 , 1 ) | | 1 = 14 ,
we have ( 2 , 6 , 5 ) , ( 2 , 5 , 6 ) B 3 . Since B 1 intersects both T 1 and T 2 , and
| | ( 2 , 6 , 5 ) a 5 | | 1 = | | ( 2 , 5 , 6 ) a 5 | | 1 = 16 ,
we have ( 2 , 5 , 6 ) , ( 2 , 6 , 5 ) B 7 T 3 . Thus, B 7 intersects both T 3 and T 7 . By Lemma 4, { a 8 , a 8 } B 8 .
Since B 3 intersects S 1 and T 3 , ( 6 , 6 , 1 ) , ( 6 , 1 , 6 ) B 3 . By
| | a 4 ( 6 , 6 , 1 ) | | 1 = 22 , | | ( 2 , 5 , 6 ) ( 6 , 6 , 1 ) | | 1 = 14 , | | a 8 ( 6 , 1 , 6 ) | | 1 = 16 ,   and   | | ( 2 , 6 , 5 ) ( 6 , 1 , 6 ) | | 1 = 14 ,
we have ( 6 , 6 , 1 ) B 8 T 7 and ( 6 , 1 , 6 ) B 4 T 7 .
Since O 7 O 1 , we have B 4 S 1 = . Thus, p 5 B 4 . By Lemma 13, p 5 B 1 . By (28) and | | p 5 ( 5 , 2 , 6 ) | | 1 = 14 , we have p 5 B 3 S 1 . By
| | p 5 p 6 | | 1 = 14 , | | ( 1 , 6 , 6 ) p 6 | | 1 = 20 , and | | p 6 a 7 | | 1 = 18 ,
we have p 6 B 5 T 3 .
Since B 8 intersects both T 8 and T 7 , ( 1 , 6 , 6 ) B 8 . By | | ( 3 , 3 , 1 ) ( 1 , 6 , 6 ) | | 1 = 18 and | | p 6 ( 1 , 6 , 6 ) | | 1 = 14 , we have ( 1 , 6 , 6 ) B 2 T 5 . By (28) and
| | ( 1 , 6 , 6 ) ( 6 , 2 , 5 ) | | 1 = 14 ,
this is impossible. □
Lemma 19.
If ( 6 , 1 , 6 ) , ( 1 , 6 , 6 ) B 2 , then v 2 B 2 .
Proof. 
From (4) and a 2 , a 2 B 2 , it follows that
| 6 p 1 ( m 2 ) |   +   | 1 p 2 ( m 2 ) |   +   | 6 p 3 ( m 2 ) |   γ ,
| 1 p 1 ( m 2 ) |   +   | 6 p 2 ( m 2 ) |   +   | 6 p 3 ( m 2 ) |   γ ,
| 6 p 1 ( m 2 ) |   +   | 5 p 2 ( m 2 ) |   +   | 2 p 3 ( m 2 ) |   γ ,
| 5 p 1 ( m 2 ) |   +   | 6 p 2 ( m 2 ) |   +   | 2 p 3 ( m 2 ) |   γ .
Taking the sum of (30) and (32), the sum of (31) and (33), and the sum of (30) and (31), and applying the triangle inequality, we obtain
| 6 p 1 ( m 2 ) |   γ 4 , | 6 p 2 ( m 2 ) |   γ 4 , and | 6 p 3 ( m 2 ) |   γ 5 .
Hence,
m 2 | f 2 + γ γ 10 + γ 10 + γ 11 + γ < 3 = ( 1 , 1 , 1 ) | f 2 .
By (4), we have v 2 B 2 . □
Lemma 20.
Either ( 3 , 1 , 3 ) B 3 or p 5 B 4 .
Proof. 
Assume the contrary that ( 3 , 1 , 3 ) B 3 S 1 and p 5 B 4 S 1 . By (4), we have
5 = ( 3 , 1 , 3 ) | f 3 m 3 | f 3 + γ .
By (16), T 3 B 3 . Clearly, B 1 T 3 = . Thus, by Table 1, we have
T 3 B 3 B 5 B 7 .
Similarly, we have
T 4 B 4 B 6 B 7 .
By Lemma 18, B 2 can only intersect T 1 and T 2 . Moreover, B 3 can only intersect T 1 and T 3 , and B 4 can only intersect T 1 and T 4 . It follows that
T 5 B 5 B 8 , T 6 B 6 B 8 , and T 7 B 7 B 8 .
By
| | a 4 ( 2 , 2 , 1 ) | | 1 = | | a 3 ( 2 , 2 , 1 ) | | 1 = 14 ,
the midpoint ( 2 , 2 , 1 ) of ( 3 , 3 , 1 ) and ( 1 , 1 , 1 ) is in B 7 B 8 . If ( 2 , 2 , 1 ) B 7 , then, by (4), we have
5 = ( 2 , 2 , 1 ) | f 7 m 7 | f 7 + γ .
By (16), we have T 7 B 7 . By (36), T 7 B 7 B 8 . In this situation, by Lemma 4, B 8 cannot intersect T 5 and T 6 . Thus, by (36),
T 5 B 5   and   T 6 B 6 .
If ( 2 , 2 , 1 ) B 8 then, by Lemma 4, B 8 cannot intersect T 5 and T 6 . By (36) again, we have (37). It follows that ( 6 , 6 , 1 ) B 6 T 6 .
By | | ( 6 , 1 , 6 ) ( 1 , 6 , 6 ) | | 1 = 22 and | | ( 6 , 1 , 6 ) ( 6 , 6 , 1 ) | | 1 = 22 , ( 6 , 1 , 6 ) B 1 B 6 . If ( 6 , 1 , 6 ) B 5 T 2 , then, since O 2 O 3 , B 5 does not intersect T 3 . By (34), B 7 T 3 . Since O 3 O 4 , by (35), B 6 T 4 . Since O 2 , O 3 , and O 4 are not adjacent to O 8 , by Lemma 4, S 8 B 8 , which contradicts Lemma 5b. Thus, ( 6 , 1 , 6 ) B 2 T 2 . Similarly, we have ( 1 , 6 , 6 ) B 2 T 2 .
By Lemma 19, v 2 B 1 B 5 B 6 . By T 1 B 1 , (37), and Lemma 5b, this is impossible. □
Clearly, by (29) and Lemma 20, ( 3 , 1 , 3 ) B 4 or p 5 B 3 .
Lemma 21.
If ( 3 , 1 , 3 ) B 4 , then ( 1 , 6 , 6 ) B 8 . If p 5 B 3 , then ( 6 , 1 , 6 ) B 8 .
Proof. 
Suppose that ( 3 , 1 , 3 ) B 4 . By B 1 T 2 ,
| | ( 3 , 1 , 3 ) ( 6 , 5 , 2 ) | | 1 = 14 , | | a 7 ( 6 , 5 , 2 ) | | 1 = 18 ,
and Lemma 4, ( 6 , 5 , 2 ) B 6 T 4 .
By | | ( 6 , 5 , 2 ) ( 1 , 6 , 6 ) | | 1 = 14 , ( 3 , 1 , 3 ) B 4 S 1 , and ( 3 , 3 , 1 ) B 2 S 1 , we have ( 1 , 6 , 6 ) B 8 T 6 .
Now, suppose that p 5 B 3 . As in the proof of Corollary 2, we have ( 3 , 1 , 3 ) Q 4 ( B 3 ) . By the first case, we have ( 1 , 6 , 6 ) ( γ B 1 3 + Q 4 ( m 8 ) ) T 6 , where m 8 is defined as in Corollary 2. Then, either ( 1 , 6 , 6 ) Q 4 ( B 7 ) T 6 or ( 1 , 6 , 6 ) Q 4 ( B 8 ) T 6 . If ( 1 , 6 , 6 ) Q 4 ( B 7 ) T 6 , then B 7 T 5 and m 8 B 7 , which contradicts Lemma 4. Hence, ( 6 , 1 , 6 ) B 8 T 5 . □
If ( 3 , 1 , 3 ) B 4 , then ( 1 , 6 , 6 ) B 8 T 6 . Hence, by Lemma 4, (23) holds. However, this is in contradiction to | | ( 1 , 6 , 6 ) a 8 | | 1 = 18 . If p 5 B 3 , then ( 6 , 1 , 6 ) B 8 T 5 , which is in contradiction to | | ( 6 , 1 , 6 ) a 8 | | 1 = 18 . This completes the proof of Theorem 1.

3. A ( 2 n + 2 ) -Optimal Configuration of B 1 n

In this section, we always assume n 3 . Let
A n = { ( ± 1 n , ± 1 n , , ± 1 n ) } B 1 n .
For each x , y A n and k [ n ] { 0 } , set
I ( x , y ) = # i [ n ] | p i ( x ) = p i ( y )   and   D x k = z A n | I ( x , z ) = k .
If γ < ( n 1 ) / n and c R n , then, for each pair x , y of points in ( γ B 1 n + c ) A n , we have
| | x y | | 1 2 γ < 2 ( n 1 ) n ,
which implies that I ( x , y ) 2 .
Let S be a maximal subset of A n such that I ( p , q ) 2 , p , q S . Then, for each p S , we have D p 0 = { p } , S A n ( D p 1 D p 0 ) and
S p S ( A n ( D p 1 D p 0 ) ) = A n p S ( D p 1 D p 0 ) A n ( S ) .
Therefore, we obtain that # S 2 n 1 .
Lemma 22.
If γ < ( n 1 ) / n and c R n , then # ( ( γ B 1 n + c ) A n ) < 2 n 1 .
Proof. 
Set S = ( γ B 1 n + c ) A n . Then, # S 2 n 1 . Assume the contrary that # S = # ( S ) = 2 n 1 . Then, by (38), p S D p 1 S . Clearly,
D p k = x A n | I ( p , x ) = k = D p n k = x A n | I ( p , x ) = n k .
It follows that
D p n 1 S , p S .
Let k [ n 1 ] and p S . Take x D p n k 1 . Without loss of generality, we may assume that
p i ( x ) = p i ( p ) i [ n k 1 ] .
Let u A n be the point satisfying
p n ( u ) = p n ( x ) = p n ( p )   and   p i ( x ) = p i ( u ) , i [ n 1 ] .
Then, u D p n k and x D u n 1 . Hence,
D p n k 1 u D p n k D u n 1 , p S , k [ n 1 ] .
For each k [ n ] , we show by induction that
D p n k S , p S .
The case when k = 1 follows from (39). Suppose that (41) holds for all k [ m ] , where m [ n 1 ] . For each p S , by (39) and (40), we have
D p n ( m + 1 ) u D p n m D u n 1 S .
Hence, (41) holds for each k [ n ] . It follows that D p 0 S , p S , which is impossible. □
Lemma 23.
For p , q A n , D p 1 = D q 1 if and only if p = q .
Proof. 
We only need to show that D p 1 D q 1 , when p q . If p = q , then, since n 3 , D p 1 D q 1 . Otherwise, there exists k [ n 1 ] such that p D q k . Without loss of generality, we may assume that p i ( p ) = p i ( q ) i [ k ] . Suppose that u A n is given by
p i ( u ) = p i ( p ) , i = 1 , p i ( p ) , i 1 .
Then, u D p 1 and, since p 1 ( u ) = p 1 ( q ) and p n ( u ) = p n ( q ) , u D q 1 . □
Lemma 24.
If c R 4 and γ < 3 / 4 , then # ( ( γ B 1 4 + c ) A 4 ) 5 .
Proof. 
Let S = ( γ B 1 4 + c ) A 4 . Then, I ( p , q ) 2 , p , q S . Assume the contrary that # S 6 . By (38), we have
# S # A 4 # ( ( p S D p 1 ) ( S ) ) # S .
Then, # ( ( p S D p 1 ) ( S ) ) 4 . Clearly, # D x 1 = 4 , x A 4 . By Lemma 23, we have # ( p S D p 1 ) 5 . Hence, p S D p 1 ( S ) . Thus, there exist p 0 , u 0 S such that u 0 D p 0 1 = D p 0 3 . Moreover, since S D p 0 2 D p 0 3 { p 0 } , # S 6 , and # D p 0 3 4 , there exists a point v 0 D p 0 2 S .
Let u D p 0 3 S and v D p 0 2 S . There exists a unique integer i ( u ) [ 4 ] such that p i ( u ) ( u ) = p i ( u ) ( p 0 ) . If p i ( u ) ( u ) = p i ( u ) ( v ) , then p i ( u ) ( v ) = p i ( u ) ( p 0 ) . Thus, there is exactly one integer j [ 4 ] { i ( u ) } such that p j ( v ) = p j ( p 0 ) = p j ( u ) . Then, for each i [ 4 ] { i ( u ) , j } , we have p i ( v ) = p i ( p 0 ) = p i ( u ) . Thus, v D u 1 , which is impossible. Therefore, p i ( u ) ( u ) = p i ( u ) ( v ) . In particular, we have
p i ( u ) ( v 0 ) = p i ( u ) ( u ) = p i ( u ) ( p 0 ) , u D p 0 3 S , p i ( u 0 ) ( v ) = p i ( u 0 ) ( u 0 ) = p i ( u 0 ) ( p 0 ) , v D p 0 2 S .
Since v 0 D p 0 2 S , # ( D p 0 3 S ) 2 . By u 0 D p 0 3 S , we have # ( D p 0 2 S ) 3 .
From # S 6 , it follows that # ( D p 0 3 S ) = 2 and # ( D p 0 2 S ) = 3 . Suppose that D p 0 3 S = { u 0 , u 1 } . For each v D p 0 2 S , we have
p i ( u 0 ) ( v ) = p i ( u 0 ) ( u 0 ) = p i ( u 0 ) ( p 0 )   and   p i ( u 1 ) ( v ) = p i ( u 1 ) ( u 1 ) = p i ( u 1 ) ( p 0 ) .
Thus, # ( D p 0 2 S ) = 1 , which is a contradiction. □
Proof of Theorem 2. 
Let
C = { ± ( 1 n , 0 , , 0 ) , ± ( 0 , 1 n , , 0 ) , ± ( 0 , 0 , , 1 n ) } .
First, we show that B 1 n ( ( n 1 ) / n ) B 1 n + C . It is clear that o ( ( n 1 ) / n ) B 1 n + c , c C . Thus, it is suffices to show that bd B 1 n ( ( n 1 ) / n ) B 1 n + C . For each x bd B 1 n , let j be an integer in [ n ] such that | p j ( x ) | = max i [ n ] | p i ( x ) | . Take c C such that p j ( c ) = 1 / n sgn ( p j ( x ) ) . Since x bd B 1 n , we have | p j ( x ) | 1 / n . Then,
| | x c | | 1 = i [ n ] { j } | p i ( x ) p i ( c ) |   +   | p j ( x ) p j ( c ) | = i [ n ] | p i ( x ) | 1 n = n 1 n .
Hence,
Γ 2 n + 2 ( B 1 n ) Γ 2 n + 1 ( B 1 n ) Γ 2 n ( B 1 n ) n 1 n .
Next, we show that Γ 2 n + 2 ( B 1 n ) ( n 1 ) / n . Otherwise, there exists γ ( 0 , ( n 1 ) / n ) and M R n with # M = 2 n + 2 such that B 1 n γ B 1 n + M . For each x A n and each point y ext B 1 n , we have x y 1 2 ( n 1 ) / n . Since ext B 1 n is 2-separated and # ( ext B 1 n ) = 2 n , A n is contained in at most two translates of γ B 1 n . Hence, there exists a c M such that # ( ( γ B 1 n + c ) A n ) 2 n 1 . By Lemma 22, this is a contradiction. □
Proof of Theorem 3. 
Since Γ 11 ( B 1 4 ) Γ 8 ( B 1 4 ) = 3 / 4 , we only need to show that Γ 11 ( B 1 4 ) 3 / 4 . Otherwise, there exists γ ( 0 , 3 / 4 ) and c k | k [ 11 ] R 4 such that B 1 4 γ B 1 4 + C . For each vertex v of B 1 4 and each point a A 4 , we have v a 1 3 / 2 > 2 γ . Then, we may assume, with loss of generality, that A 4 γ B 1 4 + c k | k [ 3 ] . By Lemma 24, # ( ( γ B 1 4 + { c k } ) A 4 ) 5 , k [ 3 ] . Since # A 4 = 16 , this is impossible. □

Author Contributions

Conceptualization, Y.L. and S.W.; methodology, Y.L. and F.C.; coding, Y.L.; validation, Y.L., F.C. and S.W.; writing—original draft preparation, Y.L.; writing—review and editing, S.W.; funding acquisition, S.W. All authors have read and agreed to the published version of the manuscript.

Funding

The authors are supported by the National Natural Science Foundation of China (grant numbers 12071444 and 12401125), and the Fundamental Research Program of Shanxi Province (grant numbers 20210302124657, 202103021224291, and 202303021221116).

Data Availability Statement

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Levi, F.W. Überdeckung eines Eibereiches durch Parallelverschiebung seines offenen Kerns. Arch. Math. 1955, 6, 369–370. [Google Scholar] [CrossRef]
  2. Lassak, M. Solution of Hadwiger’s covering problem for centrally symmetric convex bodies in E3. J. Lond. Math. Soc. 1984, 30, 501–511. [Google Scholar] [CrossRef]
  3. Prymak, A.; Shepelska, V. On the Hadwiger covering problem in low dimensions. J. Geom. 2020, 111, 42. [Google Scholar] [CrossRef]
  4. Boltyanski, V.; Martini, H.; Soltan, P.S. Excursions into Combinatorial Geometry, Universitext; Springer: Berlin/Heidelberg, Germany, 1997. [Google Scholar]
  5. Brass, P.; Moser, W.; Pach, J. Research Problems in Discrete Geometry; Springer: New York, NY, USA, 2005. [Google Scholar]
  6. Martini, H.; Soltan, V. Combinatorial problems on the illumination of convex bodies. Aequationes Math. 1999, 57, 121–152. [Google Scholar] [CrossRef]
  7. Zong, C. A quantitative program for Hadwiger’s covering conjecture. Sci. China Math. 2010, 53, 2551–2560. [Google Scholar] [CrossRef]
  8. Bezdek, K.; Khan, M.A. The geometry of homothetic covering and illumination. In Discrete Geometry and Symmetry; Springer: Cham, Switzerland, 2018; Volume 234. [Google Scholar]
  9. Yu, M.; Gao, S.; He, C.; Wu, S. Estimations of covering functionals of simplices. Math. Inequalities Appl. 2023, 26, 793–809. [Google Scholar] [CrossRef]
  10. Han, X.; Wu, S.; Zhang, L. An algorithm based on compute unified device architecture for estimating covering functionals of convex bodies. Axioms 2024, 13, 132. [Google Scholar] [CrossRef]
  11. He, C.; Lv, Y.; Martini, H.; Wu, S. A branch-and-bound approach for estimating covering functionals of convex bodies. J. Optim. Theory Appl. 2023, 196, 1036–1055. [Google Scholar] [CrossRef]
  12. Yu, M.; Lv, Y.; Zhao, Y.; He, C.; Wu, S. Estimations of covering functionals of convex bodies based on relaxation algorithm. Mathematics 2023, 11, 2000. [Google Scholar] [CrossRef]
  13. Lian, Y.; Zhang, Y. Covering the crosspolytope with its smaller homothetic copies. arXiv 2021, arXiv:2103.10004. [Google Scholar]
  14. Joós, A. Covering the Crosspolytope with Crosspolytopes. arXiv 2023, arXiv:2305.00569. [Google Scholar]
  15. Lyu, Y. Computer-Assisted Proofs [Source Code]. Available online: https://github.com/lyuyunfang/Covering_functional_of_crosspolytopes (accessed on 12 January 2025).
Figure 1. The covering configuration described in Lemma 11.
Figure 1. The covering configuration described in Lemma 11.
Mathematics 13 00546 g001
Figure 2. The cases in Lemma 12, when N = 3 .
Figure 2. The cases in Lemma 12, when N = 3 .
Mathematics 13 00546 g002
Table 1. Adjacency of orthants.
Table 1. Adjacency of orthants.
OrthantSignsAdjacent OrthantsOrthantSignsAdjacent Orthants
O 1 (+, +, +)2, 3, 4 O 5 (+, −, −)2, 3, 8
O 2 (+, +, −)1, 5, 6 O 6 (−, +, −)2, 4, 8
O 3 (+, −, +)1, 5, 7 O 7 (−, −, +)3, 4, 8
O 4 (−, +, +)1, 6, 7 O 8 (−, −, −)5, 6, 7
Table 2. The set ext T k , v k , and c k , for each k [ 8 ] .
Table 2. The set ext T k , v k , and c k , for each k [ 8 ] .
k ext T k v k c k
1 { ( 6 , 1 , 6 ) , ( 1 , 6 , 6 ) , ( 6 , 6 , 1 ) } ( 1 , 1 , 1 ) ( 4 , 4 , 5 )
2 { ( 6 , 1 , 6 ) , ( 1 , 6 , 6 ) , ( 6 , 6 , 1 ) } ( 1 , 1 , 1 ) ( 4 , 4 , 5 )
3 { ( 6 , 1 , 6 ) , ( 1 , 6 , 6 ) , ( 6 , 6 , 1 ) } ( 1 , 1 , 1 ) ( 4 , 4 , 5 )
4 { ( 6 , 1 , 6 ) , ( 1 , 6 , 6 ) , ( 6 , 6 , 1 ) } ( 1 , 1 , 1 ) ( 4 , 4 , 5 )
5 { ( 6 , 1 , 6 ) , ( 1 , 6 , 6 ) , ( 6 , 6 , 1 ) } ( 1 , 1 , 1 ) ( 4 , 4 , 5 )
6 { ( 6 , 1 , 6 ) , ( 1 , 6 , 6 ) , ( 6 , 6 , 1 ) } ( 1 , 1 , 1 ) ( 4 , 4 , 5 )
7 { ( 6 , 1 , 6 ) , ( 1 , 6 , 6 ) , ( 6 , 6 , 1 ) } ( 1 , 1 , 1 ) ( 4 , 4 , 5 )
8 { ( 6 , 1 , 6 ) , ( 1 , 6 , 6 ) , ( 6 , 6 , 1 ) } ( 1 , 1 , 1 ) ( 4 , 4 , 5 )
Table 3. The proof of (14).
Table 3. The proof of (14).
PolytopeVerticesCenter
S 1 H f 1 , 7 + { ( 6 , 6 , 1 ) , ( 6 , 1 , 6 ) , ( 1 , 6 , 6 ) , ( 1 , 3 , 3 ) , ( 3 , 3 , 1 ) , ( 3 , 1 , 3 ) } ( 5 , 5 , 4 )
S 1 H f 1 , 7 { ( 3 , 3 , 1 ) , ( 3 , 1 , 3 ) , ( 1 , 1 , 1 ) , ( 1 , 3 , 3 ) } ( 2 , 2 , 2 )
S 5 H f 5 , 7 + { ( 6 , 6 , 1 ) , ( 6 , 1 , 6 ) , ( 1 , 6 , 6 ) , ( 1 , 3 , 3 ) , ( 3 , 3 , 1 ) , ( 3 , 1 , 3 ) } ( 5 , 4 , 5 )
S 5 H f 5 , 7 { ( 3 , 3 , 1 ) , ( 3 , 1 , 3 ) , ( 1 , 1 , 1 ) , ( 1 , 3 , 3 ) } ( 2 , 2 , 2 )
S 6 H f 6 , 7 + { ( 6 , 6 , 1 ) , ( 6 , 1 , 6 ) , ( 1 , 6 , 6 ) , ( 1 , 3 , 3 ) , ( 3 , 3 , 1 ) , ( 3 , 1 , 3 ) } ( 5 , 5 , 5 )
S 6 H f 6 , 7 { ( 3 , 3 , 1 ) , ( 3 , 1 , 3 ) , ( 1 , 1 , 1 ) , ( 1 , 3 , 3 ) } ( 2 , 2 , 2 )
S 2 H f 1 , 9 H f 3 , 9 H f 4 , 9 { ( 2 , 6 , 5 ) , ( 5 , 6 , 2 ) , ( 6 , 5 , 2 ) , ( 6 , 2 , 5 ) , ( 5 , 5 , 1 ) ,
( 1 , 1 , 1 ) , ( 5 , 1 , 5 ) , ( 2 , 5 , 6 ) , ( 1 , 5 , 5 ) , ( 5 , 2 , 6 ) }
  ( 2 , 2 , 2 )
S 2 H f 1 , 9 + { ( 6 , 6 , 1 ) , ( 6 , 5 , 2 ) , ( 5 , 5 , 1 ) , ( 5 , 6 , 2 ) } ( 5 , 5 , 4 )
S 2 H f 3 , 9 + { ( 2 , 6 , 5 ) , ( 2 , 5 , 6 ) , ( 1 , 6 , 6 ) , ( 1 , 5 , 5 ) } ( 4 , 5 , 5 )
S 2 H f 4 , 9 + { ( 6 , 2 , 5 ) , ( 5 , 2 , 6 ) , ( 5 , 1 , 5 ) , ( 6 , 1 , 6 ) } ( 5 , 4 , 5 )
Table 4. Possible k such that p B k .
Table 4. Possible k such that p B k .
pCoordinateskpCoordinatesk
p 1 T 6 ( 6 , 4 , 3 ) 2 , 4 , 6 , 8 p 4 T 6 ( 5 , 6 , 2 ) 2 , 4 , 6 , 8
p 2 T 8 ( 3 , 4 , 6 ) 5 , 6 , 7 , 8 p 5 S 1 ( 1 , 3 , 3 ) 1 , 2 , 3 , 4
p 3 T 5 ( 5 , 6 , 2 ) 2 , 3 , 5 , 8 p 6 T 3 ( 6 , 5 , 2 ) 1 , 3 , 5 , 7
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Lyu, Y.; Chen, F.; Wu, S. Homothetic Covering of Crosspolytopes. Mathematics 2025, 13, 546. https://doi.org/10.3390/math13040546

AMA Style

Lyu Y, Chen F, Wu S. Homothetic Covering of Crosspolytopes. Mathematics. 2025; 13(4):546. https://doi.org/10.3390/math13040546

Chicago/Turabian Style

Lyu, Yunfang, Feifei Chen, and Senlin Wu. 2025. "Homothetic Covering of Crosspolytopes" Mathematics 13, no. 4: 546. https://doi.org/10.3390/math13040546

APA Style

Lyu, Y., Chen, F., & Wu, S. (2025). Homothetic Covering of Crosspolytopes. Mathematics, 13(4), 546. https://doi.org/10.3390/math13040546

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop