Next Article in Journal
Generalised S-System-Type Equation: Sensitivity of the Deterministic and Stochastic Models for Bone Mechanotransduction
Next Article in Special Issue
Boundedness of Some Paraproducts on Spaces of Homogeneous Type
Previous Article in Journal
Fuzzy Ensemble of Multi-Criteria Decision Making Methods for Heating Energy Transition in Danish Households
Previous Article in Special Issue
A Survey on Function Spaces of John–Nirenberg Type
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Weighted Estimates for Iterated Commutators of Riesz Potential on Homogeneous Groups

Department of Applied Mathematics, School of Mathematics and Physics, University of Science and Technology Beijing, Beijing 100083, China
*
Author to whom correspondence should be addressed.
Mathematics 2021, 9(19), 2421; https://doi.org/10.3390/math9192421
Submission received: 24 August 2021 / Revised: 22 September 2021 / Accepted: 23 September 2021 / Published: 29 September 2021
(This article belongs to the Special Issue Recent Developments of Function Spaces and Their Applications I)

Abstract

:
In this paper, we study the two weight commutators theorem of Riesz potential on an arbitrary homogeneous group H of dimension N. Moreover, in accordance with the results in the Euclidean space, we acquire the quantitative weighted bound on homogeneous group.

1. Introduction and Main Results

Suppose H is a nilpotent Lie group, which has the multiplication, inverse, expansion and norm configurations ( x , y ) x y , x x 1 , ( t , x ) t x , x ρ ( x ) for x , y H , t > 0 , respectively, then we call H being a homogeneous group (see [1] or [2]). The multiplication and inverse operations are polynomials and t-action is an automorphism of the group structure, where t is of the form
t ( x 1 , , x n ) = ( t β 1 x 1 , , t β n x n )
for some constants 0 < β 1 β 2 β n . Besides, ρ ( x ) : = max 1 j n { | x j | 1 / β j } is a norm linked to the expansion configuration. We call the value N = j = 1 n β j the dimensionality of H . In addition to the Euclidean structure, H is equipped with a homogeneous nilpotent Lie group structure, where Lebesgue measure is a bi-invariant Haar measure, the identity is the origin 0 , x 1 = x and multiplication x y , x , y H , satisfies
(1) ( a x ) ( b x ) = a x + b x , x H , a , b R ;
(2) t ( x y ) = ( t x ) ( t y ) , x , y H , t > 0 ;
(3) if z = x y , then z k = P k ( x , y ) , where P 1 ( x , y ) = x 1 + y 1 and P k ( x , y ) = x k + y k + P k ( x , y ) for k 2 with a polynomial P k ( x , y ) depending only on x 1 , , x k 1 , y 1 , , y k 1 .
Finally, the Heisenberg group on R 3 is an example of a homogeneous group. If we define the multiplication
( x , y , u ) ( x , y , u ) = ( x + x , y + y , u + u + ( x y y x ) / 2 ) ,
( x , y , u ) ( x , y , u ) R 3 , the R 3 with this group law is the Heisenberg group H 1 ; a dilation is defined by t ( x , y , u ) = ( t x , t y , t 2 u ) , that is the parameters β 1 = 1 , β 2 = 1 , β 3 = 2 .
Definition 1.
Let w ( x ) is a function on H , which is non-negative locally integrable. For 1 < p < , we call that w is an A p weight, denoted by w A p , if
[ w ] A p : = sup B 1 | B | B w ( x ) d x ( 1 | B | B ( 1 w ) 1 p 1 d x ) p 1 < ,
The supremum here is taken over of all balls B H . We call that the quantity [ w ] A p is the A p constant of w. For p = 1 , if M ( w ) ( x ) c w ( x ) for a . e . x H , then we say that w is an A 1 weight, denoted by w A 1 , where M represents the Hardy-Littlewood maximal function. In addition, let A : = 1 p A p , then we have
[ w ] A : = sup B 1 | B | B w d x exp ( 1 | B | B l o g 1 w ) d x < .
Definition 2.
Let x H , and w ( x ) be a non-negative locally integrable function. For 1 < p < q < , w A p , q if
[ w ] A p , q : = sup B 1 | B | B w q 1 | B | B w p q p < ,
where p is the conjugate exponent of p, that is 1 p + 1 p = 1 .
Definition 3.
Suppose w A . Let b L l o c 1 ( H ) , then b ( x ) B M O w ( H ) if
b B M O w ( H ) : = sup B 1 w ( B ) B | b ( x ) b B | d x < ,
where b B : = 1 | B | B b ( x ) d x and the supremum is taken over of all balls B H .
We now review the definition of Riesz potential on homogeneous group. For 0 < α < N,
I α f ( x ) : = H f ( y ) ρ ( x y 1 ) N α d y ,
and the corresponding associated maximal function M α by
M α f ( x ) = sup x B 1 | B | 1 α N B | f ( y ) | d y .
The reason why we study the weighted estimates for these operators is because they have a wide range of applications in partial differential equations, Sobolev embeddings or quantum mechanics (see [3] or [4]).
Muckenhoupt and Wheeden [5] are the first scholars to study the Riesz potential. When H is an isotropic Euclidean space, Muckenhoupt and Wheeden [5] show that I α is bounded from L p ( w p ) to L q ( w q ) for 1 < p < n α , 1 q = 1 p α n , w A p , q . Moreover, the sharp constant in this inequality was given in [6]:
I α L p ( w p ) L q ( w q ) C [ w ] A p , q ( 1 α n ) max ( 1 , p q ) .
Definition 4.
Suppose b L l o c 1 ( H ) , f L p ( H ) . Let [ b , I α ] be the commutator defined by
[ b , I α ] f ( x ) : = b ( x ) I α ( f ) ( x ) I α ( b f ) ( x ) .
The iterative commutators ( I α ) b m , m N , are defined naturally by
( I α ) b m f ( x ) : = [ b , ( I α ) b m 1 ] f ( x ) , ( I α ) b 1 f ( x ) : = [ b , I α ] f ( x ) .
In 2016, Holmes, Rahm and Spencer [7] prove that
[ b , I α ] : L w p p ( R n ) L λ q q ( R n ) b B M O μ ( R n ) ,
where 1 < p < n α , 1 q = 1 p α n , w , λ A p , q , μ = w λ . Later, the quantitative estimates for iterated commutators of fractional integrals was obtained by N. Accomazzo, J. C. Martínez-Perales and I. P. Rivera-Ríos [8].
In 2013, Sato [9] gave the estimates for singular integrals on homogeneous groups. In [10], X. T. Duong, H. Q. Li and J. Li established the Bloom-type two weight estimates for the commutator of Riesz transform on stratified Lie groups. Moreover, Z. Fan and J. Li [11] obtained the quantitative weighted estimates for rough singular integrals on homogeneous groups.
Motivated by the above estimates, we investigate the quantitative weighted estimation for the higher order commutators of fractional integral operators on homogeneous groups.
In this paper, our main result is the follow theorem.
Theorem 1.
Let 0 < α < N and 1 < p < N α , q defined by 1 q + α N = 1 p , and m is a positive integer. Assume that μ , λ A p , q and that ν = μ λ .
  • If b B M O ν 1 / m ( H ) , then
    ( I α ) b m f L λ q q ( H ) C m , N , α , p b B M O ν 1 / m ( H ) m κ m f L μ p p ( H ) ,
    where
    κ m = k = 0 m m k [ λ ] A p , q k m [ μ ] A p , q m k m ( 1 α N ) max { 1 , p q } A ( m , k ) B ( m , k )
    and
    A ( m , k ) [ λ q ] A q m + k + 1 2 [ μ q ] A q m k 1 2 m k m max { 1 , 1 q 1 } ,
    B ( m , k ) [ λ p ] A p k 1 2 [ μ p ] A p m k 1 2 k m max { 1 , 1 p 1 } .
  • For every b L l o c 1 ( H ) , if ( I α ) b m is bounded from L μ p p ( H ) to L λ q q ( H ) , then b B M O ν 1 / m ( H ) with
    b B M O ν 1 / m ( H ) m ( I α ) b m L μ p p ( H ) L λ q q ( H ) .

2. Domination of the Iterated Commutators by Sparse Operators

2.1. A System of Dyadic Cubes

We define a left-unchanged analogous-distance d on H by d ( x , y ) = ρ ( x 1 y ) , which signifies that there has a constant A 0 1 such that for any x , y , z H ,
d ( x , y ) A 0 [ d ( x , z ) + d ( z , y ) ] .
Next, let B ( x , r ) : = { y H : d ( x , y ) < r } be the open ball which is centered on x H and r > 0 is the radius.
Let A k be k-th denumerable index set. A denumerable class D : = k Z D k , D k : = { Q β k : β A k } , of Borel sets Q β k H is known as a set of dyadic cubes with arguments δ ( 0 , 1 ) and 0 < a 1 A 1 < if it has the characteristics below:
(1) H = β A k Q β k (disjoint union) for all k Z ;
(2) If k , then either Q γ Q β k or Q β k Q γ = ;
(3) For arbitrary ( k , β ) and for any k , there is a exclusive γ such that Q β k Q γ ;
(4) For arbitrary ( k , β ) there exists no more that M (a settled geometric constant) γ such that Q γ k + 1 Q β k , and Q β k = Q D k + 1 , Q Q β k Q ;
(5) B ( x β k , a 1 δ k ) Q β k B ( x β k , A 1 δ k ) = : B ( Q β k ) ;
(6) If k and Q γ Q β k , then B ( Q γ ) B ( Q β k ) . The set Q β k is called a dyadic cube of generation k with centre x β k Q β k and side length ( Q β k ) = δ k .
From the natures of the dyadic system above, for any Q β k , Q γ k + 1 and Q γ k + 1 Q β k , we get that there is a constant A ˜ 0 > 0 such that:
| Q γ k + 1 | | Q β k | A ˜ 0 | Q γ k + 1 | .

2.2. Adjacent Systems of Dyadic Cubes

Let { D t : t = 1 , 2 , , T } be a limited set of the dyadic families, then we call that it is a collection of neighbor systems of dyadic cubes with arguments δ ( 0 , 1 ) , 0 < a 1 A 1 < and 1 C a d j < if it has the following two characteristics:
(1) For any t { 1 , 2 , , T } , D t is a system of dyadic cubes with arguments δ ( 0 , 1 ) and 0 < a 1 A 1 < ;
(2) For any ball B ( x , r ) H with δ k + 3 < r δ k + 2 , k Z , there have t { 1 , 2 , , T } and Q D t of generation k which is centered on t x β k such that d ( x , t x β k ) < 2 A 0 δ k and
B ( x , r ) Q B ( x , C a d j r ) .

2.3. Sparse Operators

We review the concept of sparse family given in [12] on ordinary spaces of homogeneous description in the sense of Coifman and Weiss [13], which is also suitable in the case of homogeneous groups.
Definition 5.
Let 0 < η < 1 , for every Q S , we call that the collection S D of dyadic cubes be a η-sparse, if there exists a measurable subset E Q Q such that | E Q | η | Q | and the sets { E Q } Q S have only limited overlap.
Definition 6.
Given a sparse family, the sparse operator A S is defined by
A S ( f ) ( x ) = Q S f Q χ Q ( x ) ,
where f Q = 1 | Q | Q f ( x ) d x .
In this subfraction, the primary target is to reveal the following quantitative edition of Lacey’s pointwise domination inequality.
Proposition 1.
Let 0 < α < N . Let m be a nonnegative integer. For every f C c ( H ) and b L l o c m ( H ) , there exits T dyadic systems D t , t = 1 , 2 , , T and η-sparse families S t D t such that for a . e . x H ,
| ( I α ) b m f | C N , m , α t = 1 T k = 0 m m k A α , S t m , k ( b , f ) ( x ) , a . e . x H ,
where for a sparse family S , A α , S m , k ( b , · ) is the sparse operator given by
A α , S m , k ( b , f ) ( x ) = Q S | b ( x ) b Q | m k | Q | α N f ( b b Q ) k Q χ Q ( x ) .
To show the Proposition 1, we need some auxiliary maximal operators. To begin with, let j 0 ˜ be the smallest integer such that
2 j 0 ˜ > max { 3 A 0 , 2 A 0 C a d j }
and let C j 0 ˜ : = 2 j 0 ˜ + 2 A 0 .
Next we define the grand maximal truncated operator M I α as follows:
M I α f ( x ) = sup x B ess sup ξ B | I α ( f χ H \ C j 0 ˜ B ) ( ξ ) | ,
where the first supremum is taken over of all balls B H satisfying x B . We can know that this operator is of vital importance in the following proof, Given a ball B 0 H , for x B 0 we also define a local edition of M I α by
M I α , B 0 f ( x ) = sup x B B 0 ess sup ξ B | I α ( f χ C j 0 ˜ B 0 \ C j 0 ˜ B ) ( ξ ) | .
Now, we claim that the following lemma is true.
Lemma 1.
Let 0 < α < N . The following pointwise estimates holds:
  • For a . e . x B 0 ,
    | I α ( f χ C j 0 ˜ B 0 ) ( x ) | M I α , B 0 f ( x ) .
  • There exists a constant C N , α > 0 such that for a . e . x H ,
    M I α f ( x ) C N , α M α f ( x ) + I α | f | ( x ) .
Using the results of Lemma 1, we then prove the Proposition 1.
Proof of Proposition 1.
In order to proof the Proposition 1, we refer to the thinking in [8] for this domination, which is adapted to our situation of homogeneous groups.
Firstly, we suppose that f is supported in a ball B 0 : = B ( x 0 , r ) H , next we disintegrate H which respect to this ball B 0 . We can do it as follows. We start define the annuli U j : = 2 j + 1 B 0 \ 2 j B 0 , j 0 and select the minimum integer j 0 such that
j 0 > j 0 ˜ and 2 j 0 > 4 A 0
Next, for any U j , we select the balls
{ B ˜ j , } 1 L j ,
centred in U j and with radius 2 j j 0 ˜ r to cover U j . From the doubling property [13], we obtain
sup j L j C A 0 , j 0 ˜ ,
where C A 0 , j 0 ˜ is an positive constant that only relates on A 0 and j 0 ˜ .
We now go over the characters of these B ˜ j , . Denote B ˜ j , : = B ( x j , , 2 j j 0 ˜ r ) , where j 0 ˜ is defines as in (4). Then we have C a d j B ˜ j , : = B ( x j , , C a d j 2 j j 0 ˜ r ) , which was shown in the proof of Theorem 3.7 in [12] that
C a d j B ˜ j , U j + j 0 = , j 0 and = 1 , 2 , , L j ;
and
C a d j B ˜ j , U j j 0 = , j j 0 and = 1 , 2 , , L j .
Now, because of the Equation (8) and (9), we see that each C a d j B ˜ j , , at most overlap with 2 j 0 + 1 annuli U j ’s. Moreover, for every j and , C j 0 ˜ B ˜ j , covers B 0 .
Next by observing the (2), there is an integer t 0 { 1 , 2 , , T } and Q 0 D t 0 such that B 0 Q 0 C a d j B 0 . Additionally, for this Q 0 , as in Section 2.1 the ball that includes Q 0 and has comparable measure to Q 0 is represented by B ( Q 0 ) . Consequently, B 0 is overwritten by B ( Q 0 ) and | B ( Q 0 ) | | B 0 | , where the implicit constant relates only to C a d j and A 1 .
Now we claim that there exists a 1 2 -sparse family F t 0 D t 0 ( Q 0 ) , the set of all dyadic cubes in t 0 -th dyadic system that are contained in Q 0 , such that for a . e . x B 0 ,
| ( I α ) b m ( f χ C j 0 ˜ B ( Q 0 ) ) ( x ) | C N , m , α k = 0 m m k B α , F t 0 m , k ( b , f ) ( x ) ,
where
B α , F t 0 m , k ( b , f ) ( x ) = Q F t 0 | b ( x ) b R Q | m k | C j 0 ˜ B ( Q ) | α N f ( b b R Q ) k C j 0 ˜ B ( Q ) χ Q ( x ) .
Here, R Q is the dyadic cube in D t for some t { 1 , 2 , , T } such that C j 0 ˜ B ( Q ) R Q C a d j · C j 0 ˜ B ( Q ) , where B ( Q ) is defined as in Section 2.1, j 0 defined as in (5) and j 0 ˜ defined as in (4).
Assume that we have already proven the assertion (10). Let us take a partition of H as follows:
H = j = 0 2 j B 0 .
We next consider the annuli U j : = 2 j + 1 B 0 \ 2 j B 0 for j 0 and the covering { B ˜ j , } = 1 L j of U j as in (6). We note that for each B ˜ j , , there exist t j , { 1 , 2 , , T } and Q ˜ j , D t j , such that B ˜ j , Q ˜ j , C a d j B ˜ j , . Therefore, we acquire that for each such B ˜ j , , the enlargement C j 0 ˜ B ( Q ˜ j , l ) covers B 0 since C j 0 ˜ B ˜ j , covers B 0 .
Next, we utilize (10) to each B ˜ j , , then we acquire a 1 2 -sparse family F ˜ j , D t j , ( Q ˜ j , ) such that (10) can be established for a . e . x B ˜ j , .
Now, set F : = j , F ˜ j , . Then we observe that the balls C a d j B ˜ j , are overlapping not more than C A 0 , j 0 ˜ ( 2 j 0 + 1 ) times, where C A 0 , j 0 ˜ is the constant in (7). Then, we can obtain that F is a 1 2 C A 0 , j 0 ˜ ( 2 j 0 + 1 ) -sparse family and for a . e . c H ,
| ( I α ) b m ( f ) ( x ) | C N , m , α k = 0 m m k Q F | b ( x ) b R Q | m k | C j 0 ˜ B ( Q ) | α N f ( b b R Q ) k C j 0 ˜ B ( Q ) χ Q ( x ) .
Since C j 0 ˜ B ( Q ) R Q , and it is clear that | R Q | C ¯ | C j 0 ˜ B ( Q ) | ( C ¯ depends only on C a d j ), we obtain that f C j 0 ˜ B ( Q ) C ¯ f R Q . Now, we set S t : = { R Q D t : Q F } , t { 1 , 2 , , T } , then since the fact that F is 1 2 C A 0 , j 0 ˜ ( 2 j 0 + 1 ) -sparse, we can acquire that each family S t is 1 2 C A 0 , j 0 ˜ ( 2 j 0 + 1 ) c ¯ -sparse.
Now, we let
η : = 1 2 C A 0 , j 0 ˜ ( 2 j 0 + 1 ) c ¯ ,
where c ¯ is a constant relating only on C ¯ , C j 0 ˜ . Then it follows that (3) holds, which finishes the proof. □
Proof of the Assertion (10).
To demonstrate the assertion it suffice to attest the following recursive computation: there exist the cubes P j D t 0 ( Q 0 ) that does not intersect each other such that j | P j | 1 2 | Q 0 | and for a . e . x B 0 ,
| ( I α ) b m ( f χ C j 0 ˜ B ( Q 0 ) ) ( x ) | χ Q 0 ( x ) C N , m , α k = 0 m m k | b ( x ) b R Q 0 | m k | C j 0 ˜ B ( Q 0 ) | α N f ( b b R Q 0 ) k C j 0 ˜ B ( Q 0 ) χ Q 0 ( x ) + j | ( I α ) b m ( f χ C j 0 ˜ B ( P j ) ) ( x ) | χ P j ( x ) .
Iterating this estimate, we acquire (10) with F t 0 being the union of all the families { P j k } , where { P j 0 } = { Q 0 } , { P j 1 } = { P j } as mentioned above, and { P j k } are the cubes acquired at the k-th stage of the iterative approach. Clearly F t 0 is a 1 2 -sparse family, since let
E P j k = P j k \ j P j k + 1 .
Now we prove the recursive estimate. For any countable family { P j } j of disjoint cubes P j D t 0 ( Q 0 ) , we have that
| ( I α ) b m ( f χ C j 0 ˜ B ( Q 0 ) ) ( x ) χ Q 0 ( x ) | ( I α ) b m ( f χ C j 0 ˜ B ( Q 0 ) ) ( x ) χ Q 0 \ j P j ( x ) + j | ( I α ) b m ( f χ C j 0 ˜ B ( Q 0 ) ) ( x ) χ P j ( x ) | ( I α ) b m ( f χ C j 0 ˜ B ( Q 0 ) ) ( x ) χ Q 0 \ j P j ( x ) + j | ( I α ) b m ( f χ C j 0 ˜ B ( Q 0 ) \ C j 0 ˜ B ( P j ) ) ( x ) χ P j ( x ) + j | ( I α ) b m ( f χ C j 0 ˜ B ( P j ) ) ( x ) χ P j ( x )
So we just have to reveal that we can opt for a family of pairwise disjoint cubes { P j } D t 0 ( Q 0 ) such that j | P j | 1 2 | Q 0 | and that for a . e . x B 0 ,
| ( I α ) b m ( f χ C j 0 ˜ B ( Q 0 ) ) ( x ) | χ Q 0 \ j P j ( x ) + j | ( I α ) b m ( f χ C j 0 ˜ B ( Q 0 ) \ C j 0 ˜ B ( P j ) ) ( x ) | χ P j ( x ) C N , m , α k = 0 m m k | b ( x ) b R Q 0 | m k | C j 0 ˜ B ( Q 0 ) | α N f ( b b R Q 0 ) k C j 0 ˜ B ( Q 0 ) χ Q 0 ( x ) .
Using that ( I α ) b m f = ( I α ) b c m f for any c R , and also that
( I α ) b c m f = k = 0 m ( 1 ) k m k I α ( ( b c ) k f ) ( b c ) m k ,
it follows that
| ( I α ) b m ( f χ C j 0 ˜ B ( Q 0 ) ) ( x ) | χ Q 0 \ j P j ( x ) + j | ( I α ) b m ( f χ C j 0 ˜ B ( Q 0 ) \ C j 0 ˜ B ( P j ) ) ( x ) | χ P j ( x ) k = 0 m m k | b ( x ) b R Q 0 | m k | I α ( ( b b R Q 0 ) k f χ C j 0 ˜ B ( Q 0 ) ) ( x ) | χ Q 0 \ j P j ( x ) + k = 0 m m k | b ( x ) b R Q 0 | m k j | I α ( ( b b R Q 0 ) k f χ C j 0 ˜ B ( Q 0 ) \ C j 0 ˜ B ( P j ) ) ( x ) | χ P j ( x ) = : W 1 + W 2 .
Now we define the set E = k = 0 m E k , where
E k = { x B 0 : M I α , B 0 ( ( b b R Q 0 ) k f ) ( x ) > C N , m , α | C j 0 ˜ B ( Q 0 ) | α N ( b b R Q 0 ) k f C j 0 ˜ B ( Q 0 ) } ,
with C N , m , α being a positive number to be chosen.
From [8], we can choose C N , m , α big enough (depending on C j 0 ˜ , C a d j , and A 1 ) such that
| E | 1 4 A 0 ˜ | B 0 | ,
where A 0 ˜ is defined in Section 2.1. We now utilize the Calderón-Zygmund decomposition to the function χ E on B 0 at the height λ : = 1 2 A 0 ˜ , to acquire pairwise disjoint cubes { P j } D t 0 ( Q 0 ) such that
1 2 A 0 ˜ | P j | | P j E | 1 2 | P j |
and | E \ j P j | = 0 . This implies that
j | P j | 1 2 | B 0 | and P j E c .
Fix some j. Since we have P j E c , we observe that
M I α , B 0 ( ( b b R Q 0 ) k f ) ( x ) C N , m , α | C j 0 ˜ B ( Q 0 ) | α N ( b b R Q 0 ) k f C j 0 ˜ B ( Q 0 ) ,
which allows us to control the summation in W 2 by considering the cube P j .
Now by (i) in Lemma 1, we know that
| I α ( ( b b R Q 0 ) k f χ C j 0 ˜ B ( Q 0 ) ) ( x ) | M I α , B 0 ( ( b b R Q 0 ) k f ) ( x ) , for a . e . x B 0 .
Since | E \ j P j | = 0 , we have that
M I α , B 0 ( ( b b R Q 0 ) k f ) ( x ) C N , m , α | C j 0 ˜ B ( Q 0 ) | α N ( b b R Q 0 ) k f C j 0 ˜ B ( Q 0 ) , for a . e . x B 0 \ j P j .
Consequently,
| I α ( ( b b R Q 0 ) k f χ C j 0 ˜ B ( Q 0 ) ) ( x ) | C N , m , α | C j 0 ˜ B ( Q 0 ) | α N ( b b R Q 0 ) k f C j 0 ˜ B ( Q 0 ) , for a . e . x B 0 \ j P j .
These estimates allow us to control the remaining terms in W 1 , so we are done. □
Proof of Lemma 1.
Now we give the proof process of Lemma 1.
The result in the Euclidean space case can be referred to as [8]. Now, we can adapt the proof in [8] to our setting of homogeneous groups.
(i) Let r is close enough to 0 such that B ( x , r ) B 0 . Then,
| I α ( f χ C j 0 ˜ B 0 ) ( x ) | | I α ( f χ C j 0 ˜ B ( x , r ) ) ( x ) | + | I α ( f χ C j 0 ˜ B 0 \ C j 0 ˜ B ( x , r ) ) ( x ) | | I α ( f χ C j 0 ˜ B ( x , r ) ) ( x ) | + M I α , B 0 f ( x ) ,
the estimate for the first term follows by standard computations involving a dyadic annuli-type decomposition of the B ( x , r ) .
| I α ( f χ C j 0 ˜ B ( x , r ) ) ( x ) | = | H f ( y ) χ C j 0 ˜ B ( x , r ) d ( x , y ) N α d y | B ( x , C j 0 ˜ r ) | f ( y ) | d ( x , y ) N α d y = i = 1 B ( x , C j 0 ˜ i r ) \ B ( x , C j 0 ˜ i 1 r ) | f ( y ) | d ( x , y ) N α d y i = 1 ( C j 0 ˜ i 1 r ) α N B ( x , C j 0 ˜ i r ) | f ( y ) | d y = i = 1 ( 1 C j 0 ˜ ) α N ( C j 0 ˜ i r ) α 1 ( C j 0 ˜ i r ) N B ( x , C j 0 ˜ i r ) | f ( y ) | d y C N , α , C j 0 ˜ r α M f ( x ) .
Then,
| I α ( f χ C j 0 ˜ B 0 ) ( x ) | C N , α , C j 0 ˜ r α M f ( x ) + M I α , B 0 f ( x ) ,
the estimate in (i) is settled letting r 0 in (11).
(ii) Let x , ξ B : = B ( x 0 , r ) . Let B x be the closed ball with radius 4 ( A 0 + C j 0 ˜ ) r , which centered at x. Then C j 0 ˜ B B x , and we acquire
| I α ( f χ H \ C j 0 ˜ B ) ( ξ ) | = | I α ( f χ H \ B x ) ( ξ ) + I α ( f χ B x \ C j 0 ˜ B ) ( ξ ) | | I α ( f χ H \ B x ) ( ξ ) I α ( f χ H \ B x ) ( x ) | + | I α ( f χ B x \ C j 0 ˜ B ) ( ξ ) | + | I α ( f χ H \ B x ) ( x ) |
For the first term, since ρ is homogeneous of degree α N , and by using the Proposition 1.7 in [1], we get
| I α ( f χ H \ B x ) ( ξ ) I α ( f χ H \ B x ) ( x ) | H \ B x | f ( y ) | | 1 d ( y , ξ ) N α 1 d ( x , y ) N α | d y C N , α H \ B x 2 r d ( x , y ) N α + 1 | f ( y ) | d y = C N , α i = 1 2 i B x \ 2 i 1 B x 2 r d ( x , y ) N α + 1 | f ( y ) | d y C N , α i = 1 2 r 2 i 1 | B x | 1 N N α + 1 2 i B x | f ( y ) | d y = C N , α i = 1 2 r 2 i 1 2 2 r ( A 0 + C j 0 ˜ ) N α + 1 2 i B x | f ( y ) | d y = C N , α i = 1 2 r 2 i + 1 r ( A 0 + C j 0 ˜ ) · 1 2 i + 1 r ( A 0 + C j 0 ˜ ) N α 2 i B x | f ( y ) | d y C N , α M α f ( x ) .
Next, for ξ B , y B x \ C j 0 ˜ B , we have d ( y , ξ ) > 2 j 0 ˜ r . Then we have
| I α ( f χ B x \ C j 0 ˜ B ) ( ξ ) | B x \ C j 0 ˜ B 1 d ( y , ξ ) N α | f ( y ) | d y 1 | 2 j 0 ˜ r | N α B x | f ( y ) | d y = C N , α 1 | 4 ( A 0 + C j 0 ˜ ) r | N α B x | f ( y ) | d y C N , α M α f ( x ) .
Finally, we observe that
| I α ( f χ H \ B x ) ( x ) | = | H \ B x f ( y ) d ( x , y ) N α d x | H | f ( y ) | d ( x , y ) N α d x = I α | f | ( x ) ,
which finishes the proof of (ii). □
Next, we review that the dyadic weighted B M O space associated with the system D t is defined as
B M O η , D t ( H ) : = { b L l o c 1 ( H ) : b B M O η , D t < } ,
where b B M O η , D t = sup Q D t 1 η ( Q ) Q | b ( x ) b Q | d x . Then according to the dyadic structure theorem studies in [14], one has
B M O η ( H ) = t = 1 T B M O η , D t ( H ) .
Now, to verify a function b is in B M O η ( H ) , it suffices to verify it belongs to each weighted dyadic B M O space B M O η , D t ( H ) . Given a dyadic cube Q D t with t = 1 , 2 , , T , and a measurable function f on H , we define the local mean oscillation of f on Q by
ω λ ( f ; Q ) = inf c R ( f c ) χ Q * ( λ | Q | ) , 0 < λ < 1 ,
where
( ( f c ) χ Q ) * ( λ | Q | ) = sup E Q , | E | = λ | Q | inf x E | ( f c ) ( x ) | .
With these notation and dyadic structure theorem above, following the same proof in [10], we also acquire that for any weight η A 2 , we have
b B M O η ( H ) C t = 1 T sup Q D t ω λ ( b ; Q ) | Q | η ( Q ) , 0 < λ 2 N + 1 ,
where C depends on η .
Proposition 2.
Suppose that H is a homogeneous group with dimension N , b L l o c 1 ( H ) . Then for any cube Q H , there exist measurable set F i Q with i = 1 , 2 , such that
ω 2 1 N + 2 ( b ; Q ) b ( x ) b ( y ) , ( x , y ) F 1 × F 2 .
Proof. 
We take ideas from N. Accomazzo, J. C. Martínez-Perales and I. P. Rivera-Ríos [8]. In [8], for any cube Q D t with t = 1 , 2 , , T , there exists a subset E Q with | E | = 1 2 N + 2 | Q | such that for every x E ,
ω 2 1 N + 2 ( b ; Q ) | b ( x ) m b ( Q ) | ,
where m b ( Q ) is a not necessarily unique number that satisfies
max | { x Q : b ( x ) > m b ( Q ) } | , | { x Q : b ( x ) < m b ( Q ) } | | Q | 2 .
Let E 1 Q with | E | = 1 2 | Q | and such that b ( x ) m b ( Q ) for every x E 1 . Further let E 2 = Q \ E 1 , then | E 2 | = 1 2 | Q | and for every x E 2 , b ( x ) m b ( Q ) .
We obtain that at least half of the set E is contained either in E 1 or in E 2 since Q is the disjoint union of E 1 and E 2 . Without loss of generality, we assume that half of E is in E 1 , then we let F 1 = E E 1 , F 2 = E 2 ( E E 1 ) c , we have
| F 1 | = | E | | E ( E E 1 ) C | | E | | E | 2 = | Q | 2 N + 3 ,
and
| F 2 | = | E 2 | | E 2 ( E E 1 ) | 1 2 | Q | 1 2 N + 3 | Q | = ( 1 2 1 2 N + 3 ) | Q | .
Then if x F 1 and y F 2 , we have that
ω 2 1 N + 2 ( b ; Q ) b ( x ) m b ( Q ) b ( x ) b ( y ) ,
which shows that Proposition 2 holds. □
Given a dyadic grid D , define the dyadic Riesz potential operator
I α D f ( x ) = Q D 1 | Q | 1 α N Q | f ( y ) | d y χ Q ( x ) .
Proposition 3.
Given 0 < α < N , then for any dyadic grid D ,
I α D f ( x ) I α f ( x ) .
Proof. 
The result in the Euclidean setting is from the Proposition 2.1 in [15]. Here, we can adapt the proof in [15] to our setting of spaces of homogeneous type. □

3. Proof of Theorem 1

To proof (i), we are following the ideas in [16] or [8].
Let D be a dyadic system in H and let S be a sparse family from D . We know
A α , S m , k ( b , f ) ( x ) = Q S | b ( x ) b Q | m k | Q | α N ( b b Q ) k f Q χ Q ( x ) ,
by duality, we have that
A α , S m , k ( b , f ) ( x ) L λ q q ( H ) sup g : g L λ q q ( H ) = 1 Q S Q | g ( x ) λ q | | b ( x ) b Q | m k d x | Q | α N × 1 | Q | Q | b ( x ) b Q | k | f ( x ) | d x .
By Lemma 3.5 in [12], there exists a sparse family S ˜ D such that S S ˜ and for every cube Q S ˜ , for a . e . x Q ,
| b ( x ) b Q | C N P S ˜ , P Q Ω ( b , P ) χ P ( x ) ,
where Ω ( b , P ) = 1 | P | P | b ( x ) b P | d x
Assume that b B M O η ( H ) with η to be chosen, then we have for a . e . x Q ,
| b ( x ) b Q | C N P S ˜ , P Q 1 η ( P ) P | b ( x ) b P | d x · η ( P ) | P | χ P ( x ) C N b B M O η ( H ) P S ˜ , P Q η ( P ) | P | χ P ( x ) .
Then, we further have
A α , S m , k ( b , f ) ( x ) L λ q q ( H ) C N b B M O η ( H ) m sup g : g L λ q q ( H ) = 1 Q S ( 1 | Q | Q | g ( x ) λ q | P S ˜ , P Q η ( P ) | P | χ P ( x ) m k d x ) × 1 | Q | Q P S ˜ , P Q η ( P ) | P | χ P ( x ) k | f ( x ) | d x · | Q | · | Q | α N .
Next, note that for each N , from [12], for an arbitrary function h, we have
Q | h ( x ) | ( Q S ˜ , P Q η ( P ) | P | χ P ( x ) ) d x Q A S ˜ , η ( | h | ) ( x ) d x ,
where A S ˜ , η ( | h | ) ( x ) : = A S ˜ ( | h | ) η , A S ˜ ( h ) : = Q S ˜ h Q χ Q and A S ˜ , η f stands for the -th iteration of A S ˜ , η .
Then we have
A α , S m , k ( b , f ) ( x ) L λ q q ( H ) C N b B M O η ( H ) m sup g : g L λ q q ( H ) = 1 Q S Q A S ˜ , η m k ( | g | λ q ) · 1 | Q | 1 α N Q A S ˜ , η k ( | f | ) C N b B M O η ( H ) m sup g : g L λ q q ( H ) = 1 H Q S 1 | Q | 1 α N Q A S ˜ , η k ( | f | ) χ Q ( x ) · A S ˜ , η m k ( | g | λ q ) = C N b B M O η ( H ) m sup g : g L λ q q ( H ) = 1 H I S α A S ˜ , η k ( | f | ) ( x ) A S ˜ , η m k ( | g | λ q ) ( x ) d x ,
where I S , η α f : = I S α ( f ) η and I S α f ( x ) = Q S 1 | Q | 1 α N Q | f | χ Q ( x ) .
From (13) and the boundedness of I α f , if p , q , α are as in the hypothesis of Theorem 1.1 and w A p , q , S D , then
I S α L w p p ( H ) L w q q ( H ) C N , p , q , α [ w ] A p , q ( 1 α N ) max { 1 , p q } .
Observe that A S ˜ is self-adjoint, then
H I S α A S ˜ , η k ( | f | ) A S ˜ , η m k ( | g | λ q ) = H A S ˜ A S ˜ , η m k 1 I S , η α A S ˜ , η K ( | f | ) | g | λ q .
By Hölder inequality, we have that
A α , S m , k ( b , f ) ( x ) L λ q q ( H ) C N b B M O η ( H ) m A S ˜ A S ˜ , η m k 1 I S , η α A S ˜ , η k ( | f | ) L λ q q ( H ) .
Applying that A S ˜ L w p ( H ) C N , p [ w ] A p max { 1 , 1 p 1 } (see, e.g., [17] ),
A S ˜ A S ˜ , η m k 1 I S , η α A S ˜ , η k ( | f | ) L λ q q ( H ) C N , p [ λ q ] A q max { 1 , 1 q 1 } A S ˜ A S ˜ , η m k 2 I S , η α A S ˜ , η k ( | f | ) L λ q η q q ( H ) C N , p [ λ q ] A q max { 1 , 1 q 1 } [ λ q η q ] A q max { 1 , 1 q 1 } A S ˜ , η m k 2 I S , η α A S ˜ , η k ( | f | ) L λ q η q q ( H ) C N , p i = 0 m k 1 [ λ q η i q ] A q max { 1 , 1 q 1 } I S , η α A S ˜ , η k ( | f | ) L λ q η ( m k 1 ) q q ( H ) .
Using (14), we have that
I S , η α A S ˜ , η k ( | f | ) L λ q η ( m k 1 ) q q ( H ) = I S α A S ˜ , η k ( | f | ) L λ q η ( m k ) q q ( H ) C N , p , α [ λ η m k ] A p , q ( 1 α N ) max { 1 , p q } A S ˜ , η k ( | f | ) L λ p η ( m k ) p p ( H ) ,
and applying again A S ˜ L w p ( H ) C N , p [ w ] A p max { 1 , 1 p 1 } ,
A S ˜ , η k ( | f | ) L λ p η ( m k ) p p ( H ) C N , p i = m k + 1 m [ λ p η i p ] A p max { 1 , 1 p 1 } f L λ p η m p p ( H ) ,
which, along with the previous estimate, yields
A α , S m , k ( b , f ) ( x ) L λ q q ( H ) C N , p , α b B M O η ( H ) m A ( m , k ) B ( m , k ) [ λ η m k ] ( 1 α N ) max { 1 , p q } f L λ p η m p p ( H ) ,
where
A ( m , k ) = i = 0 m k 1 [ λ q η i q ] A q max { 1 , 1 q 1 } ,
and
B ( m , k ) = i = m k + 1 [ λ p η i p ] A p max { 1 , 1 p 1 } .
Hence, setting η = ν 1 / m , where ν = ( μ λ ) 1 / p , it reading follows from Hölder’s inequality
[ λ s ν s i m ] A s [ λ s ] A s m i m [ μ s ] A s i m , s = p , q .
Thus, we acquire that
A ( m , k ) i = 0 m k 1 [ λ q ] A q m i m [ μ q ] A q i m max { 1 , 1 q 1 } [ λ q ] A q m + k + 1 2 [ μ q ] A q m k 1 2 m k m max { 1 , 1 q 1 } ,
and
B ( m , k ) i = m k + 1 m [ λ p ] A p m i m [ μ p ] A p i m max { 1 , 1 p 1 } [ λ p ] A p k 1 2 [ μ p ] A p m k 1 2 k m max { 1 , 1 p 1 } .
Combining all the preceding estimates obtains (i).
To proof (ii), we are going to follow ideas in [10]. Based on (12), it suffices to show that there exists a positive constant C such that for all dyadic cubes Q D t ,
ω 2 1 N + 2 ( b ; Q ) m C ν 1 / m ( Q ) | Q | m ( I α ) b m L μ p p ( H ) L λ q q ( H )
Using Proposition 2 and Hölder inequality implies that
ω 2 1 N + 2 ( b ; Q ) m | F 1 | | F 2 | F 1 F 2 ( b ( x ) b ( y ) ) m d x d y d i m a ( Q ) N α F 1 F 2 b ( x ) b ( y ) m d ( x , y ) N α d x d y = d i m a ( Q ) N α F 1 ( I α ) b m ( χ F 2 ) ( x ) d x C | Q | 1 α N Q λ q 1 q · H [ ( I α ) b m ( χ F 2 ) ] q λ q d x 1 q C | Q | 1 α N Q λ q 1 q · Q μ p 1 p ( I α ) b m L μ p p ( H ) L λ q q ( H ) = C | Q | 2 1 | Q | Q λ q 1 q · 1 | Q | Q μ p 1 p ( I α ) b m L μ p p ( H ) L λ q q ( H ) ,
where we used that 1 q + α N = 1 p .
Further, this yields
ω 2 1 N + 2 ( b ; Q ) m C 1 | Q | Q λ q 1 q · 1 | Q | Q μ p 1 p ( I α ) b m L μ p p ( H ) L λ q q ( H ) .
Then from [8], we have
1 | Q | Q μ p 1 p C 1 | Q | Q ν 1 / m m 1 | Q | Q λ p 1 p ,
so the
ω 2 1 N + 2 ( b ; Q ) m C 1 | Q | Q ν 1 / m m 1 | Q | Q λ q 1 q 1 | Q | Q λ p 1 p ( I α ) b m L μ p p ( H ) L λ q q ( H ) .
Now we observe that since q > p then by Hölder inequality,
1 | Q | Q λ p 1 p 1 | Q | Q λ q 1 q and 1 | Q | Q λ q 1 q 1 | Q | Q λ p 1 p ,
then
1 | Q | Q λ q 1 q 1 | Q | Q λ p 1 p 1 | Q | Q λ q 1 q 1 | Q | Q λ p q p 1 q .
Consequently, since λ A p , q , we finally get
ω 2 1 N + 2 ( b ; Q ) m C 1 | Q | Q ν 1 / m m ( I α ) b m L μ p p ( H ) L λ q q ( H ) .
Thus, (15) holds and hence, the proof of (ii) is complete.
Therefore, we complete the proof of Theorem 1.

Author Contributions

Writing—original draft preparation, D.C.; methodology, Y.C.; check, T.W. All authors have read and agreed to the published version of the manuscript.

Funding

The project was in part supported by: Yanping Chen’s National Natural Sciience Foundation of China (# 11871096, # 11471033).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

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

Conflicts of Interest

The authors declare no conflict of interest.

Abbreviations

BMOBounded Mean Oscillation

References

  1. Folland, G.B.; Stein, E.M. Hardy Spaces on Homogeneous Groups; Princetion University Press: Princeton, NJ, USA; Univerity of Tokyo: Tokyo, Japan, 1982. [Google Scholar]
  2. Stein, E.M. Harmonic Analysis: Real-Variable Methods, Orthogonality, and Oscillatory Integrals; Princeton University Press: Priceton, NJ, USA, 1993. [Google Scholar]
  3. Frazier, M.; Jawerth, B.; Weiss, G. Littlewood-Paley Theory and the Study of Function Spaces; American Mathematical Society: Providence, RI, USA, 1991; Volume 79. [Google Scholar] [CrossRef]
  4. Sawyer, E.; Wheeden, R.L. Weighted Inequalities for Fractional Integrals on Euclidean and Homogeneous Spaces. Am. J. Math. 1992, 114, 813–874. [Google Scholar] [CrossRef]
  5. Muckenhoupt, B.; Wheeden, R. Weighted Norm Inequalities for Fractional Integrals. Trans. Am. Math. Soc. 1974, 192, 261–274. [Google Scholar] [CrossRef]
  6. Lacey, M.T.; Moen, K.; Pérez, C.; Torres, R.H. Sharp Weighted Bounds for Fractional Integral Operators. J. Funct. Anal. 2010, 259, 1073–1097. [Google Scholar] [CrossRef] [Green Version]
  7. Holmes, I.; Rahm, R.; Spencer, S. Commutators with Fractional Integral Operators. Stud. Math. 2016, 233, 279–291. [Google Scholar] [CrossRef]
  8. Accomazzo, N.; Martínez-Perales, J.C.; Rivera-Ríos, I.P. On Bloom-Type Estimates for Iterated Commutators of Fractional Integrals. Indiana Univ. Math. J. 2020, 69, 1207–1230. [Google Scholar] [CrossRef]
  9. Sato, S. Estimate for Singular Integrals on Homegeneous Groups. J. Math. Anal. Appl. 2013, 400, 311–330. [Google Scholar] [CrossRef]
  10. Duong, X.T.; Li, H.Q.; Li, J.; Wick, B.D.; Wu, Q.Y. Lower Bound of Riesz Transform Kernels Revisited and Commutators on Stratified Lie Groups. arXiv 2018, arXiv:1803.01301. [Google Scholar]
  11. Fan, Z.; Li, J. Quantitative Weighted Estimates for Rough Singular Integrals on Homogeneous Groups. arXiv 2020, arXiv:2009.02433. [Google Scholar]
  12. Duong, X.T.; Gong, R.; Kuffner, M.S.; Li, J.; Wick, B.D.; Yang, D. Two Weight Commutators on Spaces of Homogeneous Type and Applications. J. Geom. Anal. 2021, 31, 980–1038. [Google Scholar] [CrossRef] [Green Version]
  13. Coifman, R.R.; Weiss, G. Analyse Harmonique Non-Commutative Sur Certains Espaces Homogènes: Étude De Certaines Intégrales Singulières. In Lecture Notes in Mathematics; Springer: Heidelberg/Berlin, Germany, 1971; Volume 242. [Google Scholar]
  14. Kairema, A.; Li, J.; Pereyra, M.C.; Ward, L.A. Haar Bases on Quasi-Metric Measure Spaces, and Dyadic Structure Theorems for Function Spaces on Product Spaces of Homogeneous Type. J. Funct. Anal. 2016, 271, 1793–1843. [Google Scholar] [CrossRef] [Green Version]
  15. Cruz-Uribe, D.; Moen, K. One and two weighted norm inequalities for Riesz potentials. Ill. J. Math. 2013, 57, 295–323. [Google Scholar]
  16. Lerner, A.K.; Ombrosi, S.; Rivera-Ríos, I.P. Commutators of Singular Integrals Revisited. Bull. Lond. Math. Soc. 2019, 51, 107–119. [Google Scholar] [CrossRef] [Green Version]
  17. Cruz-Uribe, D.; Martell, J.M.; Pérez, C. Sharp Weighted Estimates for Classical Operators. Adv. Math. 2012, 229, 408–441. [Google Scholar] [CrossRef] [Green Version]
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Chen, D.; Chen, Y.; Wang, T. Weighted Estimates for Iterated Commutators of Riesz Potential on Homogeneous Groups. Mathematics 2021, 9, 2421. https://doi.org/10.3390/math9192421

AMA Style

Chen D, Chen Y, Wang T. Weighted Estimates for Iterated Commutators of Riesz Potential on Homogeneous Groups. Mathematics. 2021; 9(19):2421. https://doi.org/10.3390/math9192421

Chicago/Turabian Style

Chen, Daimei, Yanping Chen, and Teng Wang. 2021. "Weighted Estimates for Iterated Commutators of Riesz Potential on Homogeneous Groups" Mathematics 9, no. 19: 2421. https://doi.org/10.3390/math9192421

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