R E S E A R C H
Open Access
On Hilbert type inequalities
Chang-Jian Zhao
1*and Wing-Sum Cheung
2*Correspondence:
[email protected]; [email protected]
1Department of Mathematics, China
Jiliang University, Hangzhou 310018, P.R. China
Full list of author information is available at the end of the article
Abstract
In the present paper we establish new inequalities similar to the extensions of Hilbert’s double-series inequality and also give their integral analogues. Our results provide some new estimates to these types of inequalities.
MSC: 26D15
Keywords: Hilbert’s inequality; Pachpatte’s inequality; Hölder’s inequality
1 Introduction
In recent years several authors have given considerable attention to Hilbert’s double-series inequality together with its integral version, inverse version, and various generalizations (see [–]). In this paper, we establish multivariable sum inequalities for the extensions of Hilbert’s inequality and also obtain their integral forms. Our results provide some new estimates to these types of inequalities.
The well-known classical extension of Hilbert’s double-series theorem can be stated as follows [, p.].
Theorem A If p,p> are real numbers such that p+p ≥and <λ= –p –p =
q+
q ≤, where, as usual, qand qare the conjugate exponents of pand prespectively, then
∞
m=
∞
n= ambn (m+n)λ≤K
∞
m= apm
/p∞
n= bqn
/p
, (.)
where K=K(p,p)depends on pand ponly.
In , Pachpatte [] established a new inequality similar to inequality (.) as follows:
Theorem A Let p, q, a(s), b(t), a(), b(),∇a(s)and∇b(t)be as in [], then
m
s=
n
t=
|a(s)||b(t)|
qsp–+ptq–≤
pqm
(p–)/pn(q–)/q
m
s=
(m–s+ )∇a(s)p
/p
×
n
t=
(n–t+ )∇b(t)q
/q .
(.)
The integral analogue of inequality (.) is as follows [, p.].
Theorem B Let p, q, p, qandλbe as in Theorem A. If f ∈Lp(,∞)and g∈Lq(,∞), then
∞
∞
f(x)g(x)
(x+y)λ dx dy≤K
∞
fp(x)dx
/p ∞
gq(y)dy
/q
, (.)
where K=K(p,q)depends on p and q only.
In [], Pachpatte also established a similar version of inequality (.) as follows.
Theorem B Let p, q, f(s), g(t), f(), g(), f(s)and g(t)be as in [], then
x
y
|f(s)||g(t)|
qsp–+ptq–dt ds
≤
pqx
(p–)/py(q–)/q
x
(x–s)f(s)pds
/p y
(y–t)g(t)qdt
/q .
(.)
In the present paper we establish some new inequalities similar to Theorems A, A, B and B. Our results provide some new estimates to these types of inequalities.
2 Statement of results
Our main results are given in the following theorems.
Theorem . Let pi > be constants and p
i +
qi = . Let ai(si, . . . ,sni) be real-valued functions defined for sji= , , . . . ,mji, where mji (i,j= , , . . . ,n) are natural numbers.
For convenience, we write ai(, . . . , ) = and ai(,si, . . . ,sni) = ai(si, ,si, . . . ,sni) = · · ·=ai(si, . . . ,sn–,i, ) = . Define the operators ∇i by ∇iai(si, . . . ,sni) =ai(si, . . . ,sni) –
ai(si, . . . ,si–,i,sii– ,si+,i, . . . ,sni)for any function ai(si, . . . ,sni). Then
m
s=
· · · mn
sn=
m
s=
· · · mn
sn=
· · · mn
sn=
· · · mnn
snn=
n
i=|ai(si, . . . ,sni)| (ni=(si· · ·sni)/qi)
n i=/qi
≤M
n
i=
m
ni
sni=
· · · mi
si=
n
j=
(mji–sji+ )∇n· · · ∇ai(si, . . . ,sni) pi
/pi
,
(.)
where
M=M(mi, . . . ,mni) =
n– n
i=
/pi
n
i=/pi–n n
i=
(mi· · ·mni)/qi.
Remark . Letai(si, . . . ,sni) change toai(si) in Theorem . and in view ofai() = and ∇ai(si) =ai(si) –ai(si– ) for any functionai(si),i= , , . . . ,n, then
m
s=
m
s=
· · · mn
sn=
n i=|ai(si)| (ni=si/qi)
n
i=/qi ≤ ¯M
n
i=
mi
si=
(mi–si+ )∇ai(si) pi
/pi
, (.)
where
¯
M=M¯(m, . . . ,mn) =
n– n
i=
pi
n i=/pi–n
· n
i= m/qi
Remark . Taking forn= in Remark .. Ifp,p> satisfy p +p ≥ and <λ=
–p
–
p =
q +
q ≤, then inequality (.) reduces to
m
s=
m
s=
|a(s)||a(s)|
(qs+qs)λ
≤
(λqq)λ m/q
m
/q
m
s=
(m–s+ )∇a(s)
p
/p
(.)
×
m
s=
(m–s+ )∇a(s)
p
/p
,
which is an interesting variation of inequality (.).
On the other hand, ifλ= , then
p+
p =
q+
q = and sop=q,p=q. In this case
inequality (.) reduces to
m
s=
m
s=
|a(s)||a(s)| ps+qs
≤
pq
m(p–)/p
m
(q–)/q
m
s=
(m–s+ )∇a(s)
p
/p
×
m
s=
(m–s+ )∇a(s)
q
/q
.
This is just a similar version of inequality (.) in Theorem A.
Theorem . Let pi> be constants and pi+qi = . Let fi(τi, . . . ,τni)be real-valued nth
differentiable functions defined on[,xi)×· · ·×[,xni), where≤xji≤tji, tji∈(,∞)and
i,j= , , . . . ,n. Suppose
fi(xi, . . . ,xni) =
xi
· · ·
xni
∂n ∂τi· · ·∂τni
fi(τi, . . . ,τni)dτi· · ·dτni,
then
t
· · ·
tn
t
· · ·
tn
· · ·
tn
· · ·
tnn
n i=(
xi · · ·
xni |
∂n
∂τi···∂τnifi(τi, . . . ,τni)|
pidτ
i· · ·dτni)/pi (ni=(xi· · ·xni)/qi)
n i=/qi
dx· · ·dxndx· · ·dxn· · ·dxn· · ·dxnn (.)
≤N
n
i=
ti
· · ·
tni
n
j=
(tji–xji)
× ∂n
∂xi· · ·∂xni
fi(xi, . . . ,xni)
pidxi· · ·dxni
/pi
where
N=N(ti, . . . ,tni) =
n– n
i=
pi
n i=/pi–n
· n
i=
(ti· · ·tni)/qi.
Remark . Letfi(xi, . . . ,xni) change tofi(si) in Theorem . and in view offi() = ,i=
, , . . . ,n, then
x
. . .
xn
n i=|f(si)| (ni=si/qi)
n
i=/qidsn· · ·ds
≤ ¯N
n
i=
xi
(xi–si)fi(si) pi
dsi
/pi
,
(.)
where
¯
N=N¯(x, . . . ,xn) =
n– n
i=
pi
n i=/pi–n
· n
i= x/qi
i .
Remark . Taking forn= in Remark ., ifp,p> are such that p + p ≥ and
<λ= –p
–
p =
q+
q ≤, inequality (.) reduces to
x
x
|f(s)||f(s)|
(qs+qs)λ dsds
≤
(λqq)λ x/q
x
/q
x
(x–s)f(s)
p ds
/p
(.)
×
x
(x–s)f(s)pds
/p
,
which is an interesting variation of inequality (.).
On the other hand, ifλ= , then
p+
p =
q+
q = and sop=q,p=q. In this case
inequality (.) reduces to
x
x
|f(s)||f(s)| ps+qs
≤hh pq
x(p–)/p
x
(q–)/q
x
(x–s)f(s)pds
/p
×
x
(x–s)f(s)
q ds
/q
.
3 Proofs of results
Proof of Theorem . From the hypothesesai(, . . . , ) =ai(,si, . . . ,sni) =ai(si, ,si, . . . ,
sni) =· · ·=ai(si, . . . ,sn–,i, ) = , we have
ai(si, . . . ,sni)≤ sni
τni=
· · · si
τi=
∇n· · · ∇ai(τi, . . . ,τni). (.)
From the hypotheses of Theorem . and in view of Hölder’s inequality (see []) and inequality for mean [], we obtain
n
i=
ai(si, . . . ,sni)≤ n
i=
sni
τni=
· · · si
τi=
∇n· · · ∇ai(τi, . . . ,τni)
≤ n
i=
(si· · ·sni)/qi
sni
τni=
· · · si
τi=
∇n· · · ∇ai(τi, . . . ,τni)pi
/pi
≤(
n
i=(si· · ·sni)/qi) n
i=/qi
(n–ni=/pi)n– n
i=/pi
× n
i=
sni
τni=
· · · si
τi=
∇n· · · ∇ai(τi, . . . ,τni) pi
/pi
.
(.)
Dividing both sides of (.) by (ni=(si· · ·sni)/qi) n
i=/qi and then taking sums overs
ji from tomji (i,j= , , . . . ,n), respectively and then using again Hölder’s inequality, we obtain
m
s=
· · · mn
sn=
m
s=
· · · mn
sn=
· · · mn
sn=
· · · mnn
snn=
n
i=|∇n· · · ∇ai(si, . . . ,sni)| (ni=(si· · ·sni)/qi)
n i=/qi
≤
n– n
i=
/pi
n i=/pi–n
× n
i=
mni
sni=
· · · mi
si=
sni
τni=
· · · si
τi=
∇n· · · ∇ai(τi, . . . ,τni) pi
/pi
≤
n– n
i=
/pi
n i=/pi–n
× n
i=
(mi· · ·mni)/qi
mni
sni=
· · · mi
si=
sni
τni=
· · · si
τi=
∇n· · · ∇ai(τi, . . . ,τni)pi
/pi
=M
n
i=
mni
τni=
· · · mi
τi=
n
j=
(mji–τji+ )∇n· · · ∇ai(τi, . . . ,τni) pi
/pi
=M
n
i=
m
ni
sni=
· · · mi
si=
n
j=
(mji–sji+ )∇n· · · ∇ai(si, . . . ,sni) pi
/pi
.
Proof of Theorem . From the hypotheses of Theorem ., we have
fi(xi, . . . ,xni)≤
xi
· · ·
xni
∂n
∂τi· · ·∂τni
fi(τi, . . . ,τni)
dτi· · ·dτni. (.)
On the other hand, by using Hölder’s integral inequality (see []) and the following inequality for mean [],
n
i= λ/qi
i
/ni=/qi
≤n i=/qi
n
i=
λi/qi, λi> ,
we obtain
n
i=
fi(xi, . . . ,xni)
≤ n
i=
xi
· · · xni
∂τi· · ·∂n∂τnifi(τi, . . . ,τni)
dτi· · ·dτni
≤ n
i=
(xi· · ·xni)/qi
×
xi
· · ·
xni
∂n
∂τi· · ·∂τni
fi(τi, . . . ,τni)
pidτi· · ·dτni
/pi
≤(
n
i=(xi· · ·xni)/qi) n
i=/qi
(n–ni=/pi)n– n
i=/pi
× n
i=
xi
· · ·
xni
∂τi· · ·∂n∂τnifi(τi, . . . ,τni)
pidτi· · ·dτni
/pi
.
(.)
Dividing both sides of (.) by (ni=(xi· · ·xni)/qi) n
i=/qiand then integrating the result
inequality overxjifrom totji(i,j= , , . . . ,n), respectively and then using again Hölder’s integral inequality, we obtain
t
· · ·
tn
t
· · ·
tn
· · ·
tn
· · ·
tnn
n i=(
xi · · ·
xni
∂n
∂τi···∂τnifi(τi, . . . ,τni)
pidτi· · ·dτni)/pi (ni=(xi· · ·xni)/qi)
n i=/qi
dx· · ·dxndx· · ·dxn· · ·dxn· · ·dxnn
≤
n– n
i=
/pi
n i=/pi–n
× n
i=
ti
· · ·
tni
xi
· · ·
xni
∂τi· · ·∂n∂τnifi(τi, . . . ,τni)
pidτi· · ·dτni
/pi
≤
n– n
i=
/pi
n
i=/pi–n n
i=
(ti· · ·tni)/qi
×
ti
· · ·
tni
xi
· · ·
xni
∂τi· · ·∂n∂τnifi(τi, . . . ,τni)
pidτi· · ·dτni
dxi· · ·dxni
/pi
=N
n
i=
ti
· · · tni
n
j=
(tji–xji)
∂xi· · ·∂n∂xnifi(xi, . . . ,xni)
pidxi· · ·dxni
/pi
.
This concludes the proof.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
C-JZ and W-SC jointly contributed to the main results Theorems 2.1 and 2.2. All authors read and approved the final manuscript.
Author details
1Department of Mathematics, China Jiliang University, Hangzhou 310018, P.R. China.2Department of Mathematics, The
University of Hong Kong, Pokfulam Road, Hong Kong, P.R. China.
Acknowledgement
CJZ is supported by National Natural Science Foundation of China (10971205). WSC is partially supported by a HKU URG grant. The authors express their grateful thanks to the referees for their many very valuable suggestions and comments.
Received: 12 January 2012 Accepted: 6 June 2012 Published: 22 June 2012 References
1. Pachpatte, BG: On some new inequalities similar to Hilbert’s inequality. J. Math. Anal. Appl.226, 166-179 (1998) 2. Handley, GD, Koliha, JJ, Peˇcari´c, JE: New Hilbert-Pachpatte type integral inequalities. J. Math. Anal. Appl.257, 238-250
(2001)
3. Gao, MZ, Yang, BC: On the extended Hilbert’s inequality. Proc. Am. Math. Soc.126, 751-759 (1998) 4. Kuang, JC: On new extensions of Hilbert’s integral inequality. J. Math. Anal. Appl.235, 608-614 (1999) 5. Yang, BC: On new generalizations of Hilbert’s inequality. J. Math. Anal. Appl.248, 29-40 (2000)
6. Zhao, CJ: On inverses of disperse and continuous Pachpatte’s inequalities. Acta Math. Sin.46, 1111-1116 (2003) 7. Zhao, CJ: Generalizations on two new Hilbert type inequalities. J. Math. (Wuhan)20, 413-416 (2000)
8. Zhao, CJ, Debnath, L: Some new inverse type Hilbert integral inequalities. J. Math. Anal. Appl.262, 411-418 (2001) 9. Handley, GD, Koliha, JJ, Peˇcari´c, JE: A Hilbert type inequality. Tamkang J. Math.31, 311-315 (2000)
10. Hardy, GH, Littlewood, JE, Pólya, G: Inequalities. Cambridge University Press, Cambridge (1934)
11. Pachpatte, BG: Inequalities similar to certain extensions of Hilbert’s inequality. J. Math. Anal. Appl.243, 217-227 (2000)
doi:10.1186/1029-242X-2012-145
Cite this article as:Zhao and Cheung:On Hilbert type inequalities.Journal of Inequalities and Applications2012