R E S E A R C H
Open Access
A sharp inequality for multilinear singular
integral operators with non-smooth kernels
Guangze Gu
*and Mingjie Cai
*Correspondence: [email protected]
College of Mathematics and Econometrics, Hunan University, Changsha, 410082, P.R. China
Abstract
In this paper, we establish a sharp inequality for some multilinear singular integral operators with non-smooth kernels. As an application, we obtain the weighted Lp-norm inequality andLlogL-type inequality for the multilinear operators.
MSC: 42B20; 42B25
Keywords: multi-linear operator; singular integral operator with non-smooth kernel; sharp inequality; BMO;Ap-weight
1 Definitions and results
As the development of singular integral operators and their commutators, multilinear sin-gular integral operators have been well studied (see [–]). In this paper, we study some multilinear operator associated to the singular integral operators with non-smooth ker-nels as follows.
Definition A family of operatorsDt,t> , is said to be ‘approximations to the identity’ if, for everyt> ,Dtcan be represented by the kernelat(x,y) in the following sense:
Dt(f)(x) =
Rn
at(x,y)f(y)dy
for everyf ∈Lp(Rn) withp≥, andat(x,y) satisfies
at(x,y)≤ht(x,y) =Ct–n/s|x–y|/t,
wheresis a positive, bounded and decreasing function satisfying
lim
r→∞r n+
sr=
for some> .
Definition A linear operatorTis called a singular integral operator with non-smooth kernel ifTis bounded onL(Rn) and associated with a kernelK(x,y) such that
T(f)(x) =
RnK(x,y)f(y)dy
for every continuous function f with compact support, and for almost allx not in the support off.
() There exists an ‘approximation to the identity’{Bt,t> }such thatTBt has an asso-ciated kernelkt(x,y) and there existc,c> so that
|x–y|>ct/
K(x,y) –kt(x,y)dx≤c for ally∈Rn.
() There exists an ‘approximation to the identity’{At,t> }such thatAtThas an asso-ciated kernelKt(x,y) which satisfies
Kt(x,y)≤ct–n/ if|x–y| ≤ct/,
and
K(x,y) –Kt(x,y)≤ctδ/|x–y|–n–δ if|x–y| ≥ct/
for somec,c> ,δ> .
Letmjbe positive integers (j= , . . . ,l),m+· · ·+ml=m, and letbjbe functions onRn (j= , . . . ,l). Set, for ≤j≤m,
Rmj+(bj;x,y) =bj(x) –
|α|≤mj
α!D
α
bj(y)(x–y)α.
The multilinear operator associated toTis defined by
Tb(f)(x) =
Rn
l
j=Rmj+(bj;x,y)
|x–y|m K(x,y)f(y)dy.
Note that when m= ,Tb is just the multilinear commutator of T andb
j (see []). However, when m> ,Tb is a non-trivial generalization of the commutator. It is well known that multilinear operators are of great interest in harmonic analysis and have been widely studied by many authors (see [–]). Hu and Yang (see []) proved a variant sharp estimate for the multilinear singular integral operators. In [], Pérez and Trujillo-Gonzalez proved a sharp estimate for the multilinear commutator whenbj∈OscexpLrj(Rn) and noted that OscexpLrj ⊂BMO. The main purpose of this paper is to prove a sharp
function inequality for the multilinear singular integral operator with non-smooth kernel whenDαb
j∈BMO(Rn) for allαwith|α|=mj. As an application, we obtain anLp(p> ) norm inequality and an LlogL-type inequality for the multilinear operators. In [–], the boundedness of a singular integral operator with non-smooth kernel is obtained. In [], the boundedness of the commutator associated to the singular integral operator with non-smooth kernel is obtained. Our works are motivated by these papers.
First, let us introduce some notations. Throughout this paper,Qdenotes a cube ofRn with sides parallel to the axes. For any locally integrable functionf, the sharp function of
f is defined by
f#(x) =sup
Qx
|Q|
Q
where, and in what follows,fQ=|Q|–
Qf(x)dx. It is well known that (see [, ])
f#(x)≈sup
Qxc∈infC
|Q|
Q
f(y) –cdy.
We say thatf belongs toBMO(Rn) iff#belongs toL∞(Rn) andf
BMO=f#L∞. LetMbe a Hardy-Littlewood maximal operator defined by
M(f)(x) =sup
Qx
|Q|
Q
f(y)dy.
Fork∈N, we denote byMkthe operatorMiteratedktimes,i.e.,M(f)(x) =M(f)(x) and
Mk(f)(x) =MMk–(f)(x) whenk≥.
The sharp maximal functionMA(f) associated with the ‘approximations to the identity’
{At,t> }is defined by
M#A(f)(x) =sup
Qx
|Q|
Q
f(y) –AtQ(f)(y)dy,
wheretQ=l(Q)andl(Q) denotes the side length ofQ. For <r<∞, we denoteM#A(f)r by
M#A(f)r= MA#|f|r/r.
Letbe a Young function and˜ be the complementary associated to. For a func-tionf, we denote the-average by
f,Q=inf
λ> :
|Q|
Q
|f
(y)|
λ
dy≤
and the maximal function associated toby
M(f)(x) =sup
Qxf,Q.
The Young functions used in this paper are(t) =t( +logt)r and˜(t) =exp(t/r), the corresponding average and maximal functions are denoted by · L(logL)r,Q,ML(logL)r and · expL/r,Q,MexpL/r. Following [, , ], we know the generalized Hölder inequality
|Q|
Q
f(y)g(y)dy≤ f,Qg˜,Q
and the following inequality, forr,rj≥,j= , . . . ,lwith /r= /r+· · ·+ /rl, and any
x∈Rn,b∈BMO(Rn),
fL(logL)/r,Q≤ML(logL)/r(f)≤CML(logL)l(f)≤CMl+(f), b–bQexpLr,Q≤CbBMO,
|bk+Q–bQ| ≤CkbBMO.
We shall prove the following theorems.
Theorem If T is a singular integral operator with non-smooth kernel as given in Def-inition,let Dαb
j∈BMO(Rn)for allα with|α|=mjand j= , . . . ,l.Then there exists a
constant C> such that for any f∈C∞(Rn), <r< andx˜∈Rn,
M#ATb(f)r(x˜)≤C
l
j= |αj|=mj
Dαjb
jBMO
Ml+(f)(x˜).
Theorem If T is a singular integral operator with non-smooth kernel as given in Defini-tion,let Dαb
j∈BMO(Rn)for allαwith|α|=mjand j= , . . . ,l.Then Tbis bounded on
Lp(w)for any <p<∞and w∈A
p,that is,
Tb(f)Lp(w)≤C
l
j= |αj|=mj
Dαjb
jBMO
fLp(w).
Theorem If T is a singular integral operator with non-smooth kernel as given in Defini-tion,let w∈A,Dαbj∈BMO(Rn)for allαwith|α|=mjand j= , . . . ,l.Then there exists
a constant C> such that for allλ> ,
wx∈Rn:Tb(f)(x)>λ≤C
Rn |f(x)|
λ
+log+
|f
(x)|
λ
l
w(x)dx.
2 Proof of the theorem
To prove the theorems, we need the following lemma.
Lemma (see []) Let b be a function on Rnand Dαb∈Lq(Rn)for allαwith|α|=m and
some q>n.Then
Rm(b;x,y)≤C|x–y|m |α|=m
| ˜Q(x,y)|
˜ Q(x,y)
Dαb(z)qdz
/q ,
whereQ is the cube centered at x and having side length˜ √n|x–y|.
Lemma ([, p.]) Let <p<q<∞and for any function f ≥,we define that for
/r= /p– /q,
fWLq=sup λ>
λx∈Rn:f(x) >λ/q, Np,q(f) =sup
E f
χELp/χELr,
where thesupis taken for all measurable sets E with <|E|<∞.Then
fWLq≤Np,q(f)≤
q/(q–p)/pfWLq.
Lemma (see []) Let rj≥for j= , . . . ,l,we denote that/r= /r+· · ·+ /rl.Then
|Q|
Q
Lemma ([, ]) Let T be a singular integral operator with non-smooth kernel as given in Definition.Then T is bounded on Lp(Rn)for every <p<∞and bounded from L(Rn)
to WL(Rn).
Lemma (see [, ]) For anyγ > ,there exists a constant C> independent ofγ such that
x∈Rn:M(f)(x) >Dλ,MA#(f)(x)≤γ λ≤Cγx∈Rn:M(f)(x) >λ
forλ> ,where D is a fixed constant which only depends on n.Thus
M(f)Lp≤CM #
A(f)Lp for every f ∈Lp(Rn), <p<∞.
Lemma Let{At,t> }be an ‘approximation to the identity’ and b∈BMO(Rn).Then,for
every f ∈Lp(Rn),p> andx˜∈Rn,
sup
Q˜x
|Q|
Q
At
Q
(b–bQ)f(x)dx≤CbBMOM(f)(x˜),
where tQ=l(Q)and l(Q)denotes the side length of Q.
Proof We write, for any cube Q withx˜∈Q,
|Q|
Q
At
Q
(b–bQ)f
(x)dx≤ |Q|
Q
RnhtQ
(x,y)b(y) –bQ
f(y)dy dx ≤
|Q|
Q
Q
htQ(x,y)b(y) –bQ
f(y)dy dx
+ ∞
k=
|Q|
Q
k+Q\kQ
htQ(x,y)b(y) –bQ
f(y)dy dx
=I+I.
We have, by the generalized Hölder inequality,
I≤C
|Q||Q|
Q
Q
b(y) –bQ
f(y)dy dx ≤Cb–bQexpL,QfL(logL),Q
≤CbBMOM(f)(x˜).
ForI, notice forx∈Qandy∈k+Q\kQ, then|x–y| ≥k–tQandhtQ(x,y)≤C
s((k–)) |Q| , then
I≤C ∞
k=
s(k–)
|Q|
Q
k+Q
b(y) –bQ
f(y)dy dx
≤C
∞
k=
kns(k–)
|k+Q|
k+Q
b(y) –bQ
≤C
∞
k=
kns(k–)b–bQexpL,k+QfL(logL),k+Q
≤C
∞
k=
(k–)ns(k–)bBMOM(f)(x˜)
≤CbBMOM(f)(x˜),
where the last inequality follows from
∞
k=
(k–)ns(k–)≤C
∞
k=
–(k–)ε<∞
for some> . This completes the proof.
Proof of Theorem It suffices to prove forf ∈C∞(Rn) and some constantC
that the following inequality holds:
|Q|
Q
Tb(f)(x)r–AtQT
b(f)(x)rdx
/r
≤C
l
j= |αj|=mj
Dαjb
jBMO
Ml+(f)(x).
Without loss of generality, we may assumel= . Fix a cubeQ=Q(x,d) andx˜∈Q. LetQ˜ = √nQandb˜j(x) =bj(x) –
|α|=mj
α!(D
αb
j)Q˜xα, thenRmj(bj;x,y) =Rmj(b˜j;x,y) andD αb˜
j=
Dαb
j– (Dαbj)Q˜ for|α|=mj. We write, forf=fχQ˜ +fχRn\ ˜Q=f+f,
Tb(f)(x) =
Rn
j=Rmj+(b˜j;x,y)
|x–y|m K(x,y)f(y)dy=
Rn
j=Rmj(b˜j;x,y)
|x–y|m K(x,y)f(y)dy
–
|α|=m
α!
Rn
Rm(b˜;x,y)(x–y)αDαb˜(y)
|x–y|m K(x,y)f(y)dy
–
|α|=m
α!
Rn
Rm(b˜;x,y)(x–y)αDαb˜(y)
|x–y|m K(x,y)f(y)dy
+
|α|=m,|α|=m
α!α!
Rn
(x–y)α+αDαb˜
(y)Dαb˜(y)
|x–y|m K(x,y)f(y)dy
+
Rn
j=Rmj+(b˜j;x,y)
|x–y|m K(x,y)f(y)dy
=T
j=Rmj(b˜j;x,·) |x–·|m f
–T
|α|=m
α!
Rm(b˜;x,·)(x–·)αDαb˜
|x–·|m f
–T
|α|=m
α!
Rm(b˜;x,·)(x–·)αDαb˜
|x–·|m f
+T
|α|=m,|α|=m
α!α!
(x–·)α+αDαb˜ Dαb˜
|x–·|m f
+T
j=Rmj+(b˜j;x,·) |x–·|m f
and
AtQT
b(f)(x) =
Rn
j=Rmj(b˜j;x,y)
|x–y|m Kt(x,y)f(y)dy
–
|α|=m
α!
Rn
Rm(b˜;x,y)(x–y)αDαb˜(y)
|x–y|m Kt(x,y)f(y)dy
–
|α|=m
α!
Rn
Rm(b˜;x,y)(x–y)αDαb˜(y)
|x–y|m Kt(x,y)f(y)dy
+
|α|=m,|α|=m
α!α!
Rn
(x–y)α+αDαb˜
(y)Dαb˜(y)
|x–y|m Kt(x,y)f(y)dy
+
Rn
j=Rmj+(b˜j;x,y)
|x–y|m Kt(x,y)f(y)dy
=AtQT
j=Rmj(b˜j;x,·) |x–·|m f
–AtQT
|α|=m
α!
Rm(b˜;x,·)(x–·)αDαb˜
|x–·|m f
–AtQT
|α|=m
α!
Rm(b˜;x,·)(x–·)αDαb˜
|x–·|m f
+AtQT
|α|=m,|α|=m
α!α!
(x–·)α+αDαb˜ Dαb˜
|x–·|m f
+AtQT
j=Rmj+(b˜j;x,·) |x–·|m f
,
then
|Q|
Q
Tb(f)(x)r–AtQT
b(f)(x)r
dx /r ≤ |Q| Q
Tb(f)(x) –A tQT
b(f)(x)rdx
/r ≤ C |Q| Q T
j=Rmj(b˜j;x,·) |x–·|m f
rdx /r + C |Q| Q T
|α|=m
Rm(b˜;x,·)(x–·)αDαb˜
|x–·|m f
rdx /r + C |Q| Q T
|α|=m
Rm(b˜;x,·)(x–·)αDαb˜
|x–·|m f
rdx /r + C |Q|
Q
T
|α|=m,|α|=m
Q
(x–·)α+αDαb˜ Dαb˜
|x–·|m f
rdx /r + C |Q| Q AtQT
j=Rmj(b˜j;x,·) |x–·|m f
rdx
+ C |Q| Q AtQT
|α|=m
α!
Rm(b˜;x,·)(x–·)αDαb˜
|x–·|m f
rdx /r + C |Q|
Q
AtQT
|α|=m
α!
Rm(b˜;x,·)(x–·)αDαb˜
|x–·|m f
rdx /r + C |Q|
Q
AtQT
|α|=m,|α|=m
α!α!
(x–·)α+αDαb˜ Dαb˜
|x–·|m f
rdx /r + C |Q|
Q
(T–AtQT)
j=Rmj+(b˜j;x,·) |x–·|m f
rdx
/r
:=I+I+I+I+I+I+I+I+I.
Now, let us estimateI,I,I,I,I,I,I,IandI, respectively. First, forx∈Qandy∈ ˜Q, by Lemma , we get
Rm(b˜j;x,y)≤C|x–y|m |αj|=m
Dαjb
jBMO,
by Lemma and the weak type (, ) ofT(Lemma ), we obtain
I≤C
j= |αj|=mj
Dαjb
jBMO
|Q|
Rn
T(f)(x) r dx /r ≤C
j= |αj|=mj
Dαjb
jBMO
|Q|–T(f)χQLr |Q|/r–
≤C
j= |αj|=mj
Dαjb
jBMO
|Q|–T(f )WL
≤C
j= |αj|=mj
Dαjb
jBMO
| ˜Q|–f L
≤C
j= |αj|=mj
Dαjb
jBMO
M(f)(x˜).
ForI, we get, by Lemma and the generalized Hölder inequality,
I≤C
|α|=m
Dαb BMO
|α|=m
|Q| Rn T
Dαb˜ f
(x)rdx
/r
≤C
|α|=m
Dαb BMO
|α|=m
|Q|–T(D αb˜
f)χQLr |Q|/r–
≤C
|α|=m
Dαb BMO
|α|=m
|Q|–TDαb˜ fWL
≤C
|α|=m
Dαb BMO
|α|=m
≤C
|α|=m
Dαb BMO
|α|=m
Dαb –
Dαb
˜
QexpL,Q˜fL(logL),Q˜
≤C
j= |αj|=mj
Dαjb
jBMO
M(f)(x˜).
ForI, similar to the proof ofI, we get
I≤C
j= |α|=mj
DαbjBMO
M(f)(x˜).
Similarly, forI, takingr,r,r≥ such that /r= /r+ /r, we obtain, by Lemma and the generalized Hölder inequality,
I≤C
|α|=m,|α|=m
|Q|
Rn
TDαb˜ Dαb˜f
(x)rdx
/r
≤C
|α|=m,|α|=m
|Q|–T(Dαb˜Dαb˜f)χQLr |Q|/r–
≤C
|α|=m,|α|=m
|Q|–TDαb˜
Dαb˜fWL
≤C
|α|=m,|α|=m
|Q|–Dαb˜
Dαb˜fL
≤C
|α|=m,|α|=m
j=
Dαjb
j–
Dαjb
j
˜
QexpLrj,Q˜ · fL(logL)/r,Q˜
≤C
j= |α|=mj
DαbjBMO
M(f)(x˜).
ForI,I,IandI, by Lemma , we get
I+I+I+I≤C
j= |αj|=mj
Dαjb
jBMO
|Q|
Q
AtQT(f)(x)dx
+C
|α|=m
Dαb BMO
|α|=m
|Q|
Q
AtQT
Dαb˜ f
(x)dx
+C
|α|=m
Dαb BMO
|α|=m
|Q|
Q
AtQT
Dαb˜ f
(x)dx
+C
|α|=m,|α|=m
|Q|
Q
AtQT
Dαb˜ Dαb˜f
(x)dx
≤C
j= |α|=mj
DαbjBMO
ForI, we write
(T–AtQT)
j=Rmj+(b˜j;x,·) |x–·|m f
=
Rn
j=Rmj+(b˜j;x,y) |x–y|m
K(x,y) –Kt(x,y)
f(y)dy
=
Rn
j=Rmj(b˜j;x,y) |x–y|m
K(x,y) –Kt(x,y)f(y)dy
–
|α|=m
α!
Rn Dαb˜
(y)(x–y)αRm(b˜;x,y)
|x–y|m
K(x,y) –Kt(x,y)f(y)dy
–
|α|=m
α!
Rn Dαb˜
(y)(x–y)αRm(b˜;x,y)
|x–y|m
K(x,y) –Kt(x,y)
f(y)dy
+
|α|=m,|α|=m
α!α!
Rn Dαb˜
(y)Dαb˜(y)(x–y)α+α
|x–y|m
K(x,y) –Kt(x,y)
f(y)dy
=I()+I()+I()+I().
By Lemma and the following inequality (see [])
|bQ–bQ| ≤Clog
|Q|/|Q|
bBMO forQ⊂Q,
we know that forx∈Qandy∈k+Q˜\kQ˜,
Rm(b˜;x,y)≤C|x–y|m
|α|=m
Dαb
BMO+D
αb
˜ Q(x,y)–
Dαb
˜
Q
≤Ck|x–y|m
|α|=m
Dαb
BMO.
Note that|x–y| ≥d=t/and|x–y| ∼ |x
–y|forx∈Qandy∈Rn\ ˜Q. By the conditions onKandKt, we obtain
I()= ∞
k=
k+Q\˜ kQ˜
j=|Rmj(b˜j;x,y)|
|x–y|m K(x,y) –Kt(x,y)f(y)dy
≤C
j= |α|=mj
DαbjBMO
∞
k=
k+Q\˜ kQ˜k
dδ
|x–y|n+δ
f(y)dy
≤C
j= |α|=mj
DαbjBMO
∞
k=
k–δk
|kQ|˜
kQ˜
f(y)dy
≤C
j= |α|=mj
DαbjBMO
ForI(), we get, by the generalized Hölder inequality,
I()≤C
|α|=m
Dαb BMO
|α|=m ∞
k=
k+Q\˜ kQ˜ kdδ |x–y|n+δ
Dαb˜
(y)f(y)dy
≤C
|α|=m
Dαb BMO
×
|α|=m ∞
k=
k–δkDαb –
Dαb
˜
QexpL,kQ˜fL(logL),kQ˜
≤C
j= |α|=mj
Dαb
jBMO
M(f)(x˜).
Similarly,
I()≤C
j= |α|=mj
DαbjBMO
M(f)(x˜).
ForI(), taking r,r,r≥ such that /r= /r+ /r, by Lemma and the generalized Hölder inequality, we get
I()≤C
|α|=m,|α|=m ∞
k=
k+Q\˜ kQ˜ dδ |x–y|n+δ
Dαb˜
(y)Dαb˜(y)f(y)dy
≤C
|α|=m,|α|=m ∞
k=
j=
Dαjb
j–
Dαjb
j
˜
QexpLrj,kQ˜fL(logL)/r,kQ˜
≤C
j= |α|=mj
Dαb
jBMO
M(f)(x˜).
Thus
|I| ≤C
j= |α|=mj
DαbjBMO
M(f)(x˜).
This completes the proof of Theorem .
By Theorem and theLp(w)-boundedness ofMl+, we may obtain the conclusions of Theorem . By Theorem and [, ], we may obtain the conclusions of Theorem .
3 Applications
In this section we shall apply Theorems , and of the paper to the holomorphic func-tional calculus of linear elliptic operators. First, we review some definitions regarding the holomorphic functional calculus (see []). Given ≤θ<π, define
and its interior bySθ. SetSθ˜ =Sθ\ {}. A closed operatorLon some Banach spaceEis said
to be of typeθ if its spectrumσ(L)⊂Sθand if for everyν∈(θ,π], there exists a constant
Cνsuch that
|η|(ηI–L)–≤Cν, η∈ ˜/Sθ.
Forν∈(,π], let
H∞Sμ
=f :Sθ→C:f is holomorphic andfL∞<∞
,
wherefL∞=sup{|f(z)|:z∈Sμ} . Set
Sμ
=
g∈H∞Sμ
:∃s> ,∃c> such thatg(z)≤c |z|
s
+|z|s
.
IfLis of typeθandg∈H∞(Sμ), we defineg(L)∈L(E) by
g(L) = –(πi)–
(ηI–L)–g(η)dη,
where is the contour {ξ =re±iφ : r≥} parameterized clockwise around Sθ with
θ<φ<μ. If, in addition,Lis one-to-one and has a dense range, then, forf ∈H∞(Sμ),
f(L) = h(L)–(fh)(L),
whereh(z) =z( +z)–.Lis said to have a bounded holomorphic functional calculus on the sectorSμif
g(L)≤NgL∞
for someN> and for allg∈H∞(Sμ).
Now, letLbe a linear operator onL(Rn) withθ <π/ so that (–L) generates a holo-morphic semigroupe–zL, ≤ |arg(z)|<π/ –θ. Applying Theorem of [], we get the following.
Theorem Assume the following conditions are satisfied:
(i)The holomorphic semigroup e–zL, ≤ |arg(z)|<π/ –θ is represented by the kernels
az(x,y)which satisfy,for allν>θ,an upper bound
az(x,y)≤cνh|z|(x,y)
for x,y∈Rn,and≤ |arg(z)|<π/ –θ,where h
t(x,y) =Ct–n/s(|x–y|/t)and s is a positive,
bounded and decreasing function satisfying
lim
r→∞r n+
(ii)The operator L has a bounded holomorphic functional calculus in L(Rn),that is,for
allν>θand g∈H∞(Sμ),the operator g(L)satisfies
g(L)(f)L≤cνgL∞fL.
Then,for Dαb
j∈BMO(Rn)for allαwith|α|=mjand j= , . . . ,l,the multilinear operator
g(L)bassociated to g(L)and b
jsatisfies: (a) For <r< andx˜∈Rn,
M#Ag(L)b(f)r(x˜)≤C
l
j= |αj|=mj
Dαjb
jBMO
Ml+(f)(x˜);
(b) g(L)bis bounded onLp(w)for any <p<∞andw∈A
p,that is,
g(L)b(f)Lp(w)≤C
l
j= |αj|=mj
Dαjb
jBMO
fLp(w);
(c) There exists a constantC> such that for allλ> andw∈A,
wx∈Rn:g(L)b(f)(x)>λ≤C
Rn |f(x)|
λ
+log+
|f
(x)|
λ
l
w(x)dx.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
The authors completed the paper together. They also read and approved the final manuscript.
Acknowledgements
Project supported by Hunan Provincial Natural Science Foundation of China (12JJ6003) and Scientific Research Fund of Hunan Provincial Education Departments (12K017).
Received: 17 March 2013 Accepted: 12 August 2013 Published: 16 September 2013
References
1. Cohen, J: A sharp estimate for a multilinear singular integral onRn. Indiana Univ. Math. J.30, 693-702 (1981)
2. Cohen, J, Gosselin, J: On multilinear singular integral operators onRn. Stud. Math.72, 199-223 (1982)
3. Cohen, J, Gosselin, J: A BMO estimate for multilinear singular integral operators. Ill. J. Math.30, 445-465 (1986) 4. Coifman, R, Meyer, Y: Wavelets, Calderón-Zygmund and Multilinear Operators. Cambridge Studies in Advanced
Math., vol. 48. Cambridge University Press, Cambridge (1997)
5. Coifman, RR, Rochberg, R, Weiss, G: Factorization theorems for Hardy spaces in several variables. Ann. Math.103, 611-635 (1976)
6. Ding, Y, Lu, SZ: Weighted boundedness for a class rough multilinear operators. Acta Math. Sin.17, 517-526 (2001) 7. Pérez, C, Trujillo-Gonzalez, R: Sharp weighted estimates for multilinear commutators. J. Lond. Math. Soc.65, 672-692
(2002)
8. Hu, GE, Yang, DC: A variant sharp estimate for multilinear singular integral operators. Stud. Math.141, 25-42 (2000) 9. Duong, XT, McIntosh, A: Singular integral operators with non-smooth kernels on irregular domains. Rev. Mat.
Iberoam.15, 233-265 (1999)
10. Liu, LZ: Sharp function boundedness for vector-valued multilinear singular integral operators with non-smooth kernels. J. Contemp. Math. Anal.45, 185-196 (2010)
11. Liu, LZ: Multilinear singular integral operators on Triebel-Lizorkin and Lebesgue spaces. Bull. Malays. Math. Soc.35, 1075-1086 (2012)
12. Martell, JM: Sharp maximal functions associated with approximations of the identity in spaces of homogeneous type and applications. Stud. Math.161, 113-145 (2004)
14. Garcia-Cuerva, J, Rubio de Francia, JL: Weighted Norm Inequalities and Related Topics. North-Holland Mathematics Studies, vol. 116. North-Holland, Amsterdam (1985)
15. Stein, EM: Harmonic Analysis: Real Variable Methods, Orthogonality and Oscillatory Integrals. Princeton University Press, Princeton (1993)
16. Pérez, C: Endpoint estimate for commutators of singular integral operators. J. Funct. Anal.128, 163-185 (1995) 17. Pérez, C, Pradolini, G: Sharp weighted endpoint estimates for commutators of singular integral operators. Mich. Math.
J.49, 23-37 (2001)
doi:10.1186/1029-242X-2013-439