Skip to main content

Two weighted inequalities for B-fractional integrals

Abstract

In this paper we prove a two weighted inequality for Riesz potentials \(I_{\alpha,\gamma} f\) (B-fractional integrals) associated with the Laplace-Bessel differential operator \(\Delta_{B}=\sum_{i=1}^{n} \frac{\partial^{2}}{\partial x_{i}^{2}} + \sum_{j=1}^{k} \frac{\gamma _{j}}{x_{j}}\frac{\partial}{\partial x_{j}}\). This result is an analog of Heinig’s result (Indiana Univ. Math. J. 33(4):573-582, 1984) for the B-fractional integral. Further, the Stein-Weiss inequality for B-fractional integrals is proved as an application of this result.

1 Introduction

Let \(\mathbb{R}_{k,+}^{n}=\{x=(x_{1},\ldots,x_{n})\in{\mathbb{R}^{n}}: x_{1}>0,x_{2}>0,\ldots,x_{k}>0 \}\), \(1\leq k\leq n\), and w be a weight function on \(\mathbb {R}_{k,+}^{n}\), i.e., w is a non-negative and measurable function on \(\mathbb {R}_{k,+}^{n}\). The weighted Lebesgue space \(L_{p,w,\gamma}\equiv L_{p,w,\gamma} ( \mathbb{R}^{n}_{k,+} ) \), \(1\leq p< \infty\), is the set of all classes of measurable functions f with finite norm

$$\|f\|_{p,w,\gamma}= \biggl( \int_{\mathbb{R}^{n}_{k,+}}\bigl\vert f ( x ) \bigr\vert ^{p}w(x) \bigl(x'\bigr)^{\gamma }\,dx \biggr) ^{\frac{1}{p} }, $$

where \((x')^{\gamma}=x_{1}^{\gamma_{1}}\cdot\ldots\cdot x_{k}^{\gamma_{k}}\) and \(\gamma= (\gamma_{1},\ldots,\gamma_{k} )\) is a multi-index consisting of fixed positive numbers such that \(|\gamma|=\gamma_{1} + \cdots + \gamma_{k}\).

If \(p=\infty\), we assume

$$L_{\infty,w, \gamma}\bigl(\mathbb{R}_{k,+}^{n} \bigr)=L_{\infty}\bigl(\mathbb {R}_{k,+}^{n}\bigr)=\bigl\{ f: \| f\| _{L_{\infty,w}}= \mathop{\operatorname{ess} \sup}_{x \in\mathbb{R}_{k,+}^{n}} \bigl\vert w(x)f(x)\bigr\vert < \infty \bigr\} . $$

The fractional integral operators play an important role in the theory of harmonic analysis, differentiation theory and PDE’s. Many mathematicians have dealt with the fractional integrals and related topics associated with the Laplace-Bessel differential operator \(\Delta_{B}=\sum_{i=1}^{n} \frac{\partial^{2}}{\partial x_{i}^{2}} + \sum_{j=1}^{k} \frac{\gamma _{j}}{x_{j}}\frac{\partial}{ \partial x_{j}}\) such as Aliev and Gadjiev [2], Guliyev [3], Gadjiev and Hajibayov [4], Guliyev et al. [5] and others. In this paper we consider fractional (B-fractional) integrals in the weighted Lebesgue space \(L_{p,w,\gamma} ( \mathbb{R}^{n}_{k,+} ) \) associated with the generalized shift operator defined by (see, for example [6, 7])

$$T^{y}f(x)=C_{k,\gamma} \int_{0}^{\pi}\cdots \int_{0}^{\pi}f \bigl( \bigl(x',y' \bigr)_{\alpha}, x''-y'' \bigr) \, d\nu(\alpha), $$

where

$$\begin{aligned}& C_{k, \gamma} =\pi^{-{\frac{k}{2}}}\prod_{i=1}^{k} \frac{\Gamma (\frac{\gamma_{i}+1}{2} )}{\Gamma (\frac{\gamma_{i}}{2} )},\qquad \bigl(x',y'\bigr)_{\alpha}=\bigl((x_{1},y_{1})_{\alpha_{1}},\ldots, (x_{k},y_{k})_{\alpha _{k}}\bigr), \\& (x_{i},y_{i})_{\alpha_{i}}= \sqrt{x_{i}^{2}-2x_{i}y_{i}\cos\alpha _{i}+y_{i}^{2}}, \qquad \bigl(x',x''\bigr)\in\mathbb{R}^{k}\times\mathbb{R}^{n-k},\quad \mbox{and} \\& d\nu(\alpha) = \prod_{i=1}^{k} \sin^{\gamma_{i}-1}\alpha_{i} \, d\alpha_{i},\quad 1\leq i\leq k, 1\le k \le n. \end{aligned}$$

It is well known that the generalized shift operator \(T^{y}\) is closely related to the Laplace-Bessel differential operator \(\Delta_{B}\). Furthermore, \(T^{y}\) generates the corresponding B-convolution

$$(f \otimes g) (x)= \int_{\mathbb{R}_{k,+}^{n}}f(y) \bigl(T^{y} g(x) \bigr) \bigl(y'\bigr)^{\gamma}\,dy. $$

The B-fractional integral (or B-Riesz potential) is defined by

$$I_{\alpha,\gamma} f(x)= \int_{\mathbb{R}^{n}_{k,+}}T^{y} \bigl(|x|^{\alpha-n-|\gamma|} \bigr)f(y) \bigl(y'\bigr)^{\gamma}\,dy, \quad 0< \alpha < n+|\gamma|. $$

The properties of the B-fractional integral has been examined extensively. We refer to [25, 810] and for more general case to [11, 12].

In the case \(w=1\), for the classical Riesz potential \(I_{\alpha}f\), the classical Hardy-Littlewood-Sobolev theorem [13] states that, if \(1 < p < \infty\) and \(\alpha p < n\), then \(I_{\alpha}f\) is an operator of strong type \((p, q)\), where \(\frac{1}{q}=\frac{1}{p}-\frac{\alpha}{n}\), and if \(p = 1\), then \(I_{\alpha}f\) is an operator of weak type \((1, q)\), where \(\frac{1}{q}=1-\frac{\alpha}{n}\).

In the following we give the Heinig’s result [1] for the boundedness of the classical Riesz potential \(I_{\alpha}f\) in weighted Lebesgue spaces which is a generalization of the Hardy-Littlewood-Sobolev theorem for \(I_{\alpha}f\).

Theorem A

([1])

Suppose u and v are defined on \(R^{n}\) and \(U=u^{\ast}\), \(\frac{1}{V}= (\frac{1}{v} )^{\ast}\). If \(1\leq p\leq q\leq\infty\), \(p<\infty\), and, for some r, \(1< r<\frac{n}{\alpha}\),

$$\sup_{s>0} \biggl( \int_{s}^{\infty} \bigl[U(t)t^{\alpha -n} \bigr]^{q}t^{n-1}\,dt \biggr)^{\frac{1}{q}} \biggl( \int_{0}^{s}V(t)^{-p'}t^{n-1} \,dt \biggr)^{\frac{1}{p'}}< \infty $$

and

$$\sup_{s>0} \biggl( \int_{0}^{s} \bigl[U(t)t^{\alpha-\frac{n}{r}} \bigr]^{q}t^{n-1}\,dt \biggr)^{\frac{1}{q}} \biggl( \int_{s}^{\infty} \bigl[V(t)t^{\frac{n}{r'}} \bigr]^{-p'}t^{n-1}\,dt \biggr)^{\frac{1}{p'}}< \infty, $$

then \(I_{\alpha}:L_{p,v}({\mathbb{R}^{n}})\rightarrow L_{q,u}({\mathbb {R}^{n}}) \) is bounded.

Our purpose in this paper is to give an analog of Heinig’s result for the B-fractional integral \(I_{\alpha,\gamma} f\). Further, the Stein-Weiss inequality for B-Riesz potential is proved as an application of this result. Note that the Stein-Weiss inequality for the classical Riesz potentials was given in [14]. For the B-fractional integrals, this inequality was proved in [9] and [10].

2 Preliminaries

Let \(1\leq p\leq\infty\). In the case \(w=1\), if f is in \(L_{p,\gamma } ( \mathbb{R}^{n}_{k,+} ) \) and φ is in \(L_{1,\gamma} ( \mathbb{R}^{n}_{k,+} ) \), then the function \(f\otimes\varphi\) belongs to \(L_{p,\gamma} ( \mathbb{R}^{n}_{k,+} ) \) and

$$\|f\otimes\varphi\|_{p,\gamma}\leq\|f\|_{p,\gamma}\| \varphi\| _{1,\gamma}. $$

Suppose f is a measurable function defined on \(\mathbb{R}^{n}_{k,+}\). For any measurable set \(E \subset\mathbb{R}_{k,+}^{n}\), let \(|E|_{\gamma}=\int_{E}(x')^{\gamma}\,dx\). The distribution function \(f _{\ast,\gamma}\) of the function f is given by

$$f_{\ast,\gamma}(s)=\bigl\vert \bigl\{ x:x \in\mathbb{R}^{n}_{k,+}, \bigl\vert f(x)\bigr\vert >s\bigr\} \bigr\vert _{\gamma}, \quad \text{for } s\geq0. $$

The distribution function \(f _{\ast,\gamma}\) is non-negative, non-increasing, and continuous from the right (see [15]). With the distribution function we associate the non-increasing rearrangement of f on \([0,\infty)\) defined by

$$f^{\ast,\gamma}(t)=\inf\bigl\{ s>0:f _{\ast,\gamma}(s) \leq t\bigr\} . $$

If \(f \in L_{p,\gamma} ( \mathbb{R}^{n}_{k,+} )\), \(1\leq p< \infty\), then

$$\biggl( \int_{\mathbb{R}^{n}_{k,+}}\bigl\vert f ( x ) \bigr\vert ^{p} \bigl(x'\bigr)^{\gamma}\,dx \biggr) ^{\frac{1}{p} }= \biggl(p \int_{0}^{\infty}s^{p-1}f _{\ast,\gamma}(s) \,ds \biggr)^{\frac{1}{p}}= \biggl( \int_{0}^{\infty}\bigl(f ^{\ast,\gamma}(t) \bigr)^{p}\,dt \biggr)^{\frac{1}{p}}. $$

In the following we give several inequalities which we will need in the proof of our main results.

Lemma 1

(Hardy inequalities [1618])

Suppose ξ and θ are non-negative locally integrable functions defined on \((0,\infty)\) and \(1< p\leq q<\infty\). Then there exists a constant \(C>0\) such that for all non-negative Lebesgue measurable function ψ on \((0,\infty)\), the inequality

$$ \biggl[ \int_{0}^{\infty} \biggl( \int_{0}^{t}\psi(s)\,ds \biggr)^{q}\xi (t)\,dt \biggr]^{\frac{1}{q}}\leq C \biggl( \int_{0}^{\infty} \bigl(\psi (t) \bigr)^{p} \theta(t)\,dt \biggr)^{\frac{1}{p}} $$
(1)

is satisfied if and only if

$$ \sup_{s>0} \biggl( \int_{s}^{\infty}\xi(t)\,dt \biggr)^{\frac{1}{q} } \biggl( \int_{0}^{s} \bigl(\theta(t) \bigr)^{1-p^{\prime}} \,dt \biggr)^{\frac{1}{p^{\prime}}}< \infty. $$
(2)

Similarly for the dual operator,

$$ \biggl[ \int_{0}^{\infty} \biggl( \int_{t}^{\infty}\psi(s)\,ds \biggr)^{q}\xi(t) \,dt \biggr]^{\frac{1}{q}}\leq C \biggl( \int_{0}^{\infty } \bigl(\psi(t) \bigr)^{p} \theta(t)\,dt \biggr)^{\frac{1}{p}} $$
(3)

is satisfied if and only if

$$ \sup_{s>0} \biggl( \int_{0}^{s}\xi(t)\,dt \biggr)^{\frac{1}{q} } \biggl( \int_{s}^{\infty} \bigl(\theta(t) \bigr)^{1-p^{\prime}} \,dt \biggr)^{\frac{1}{p^{\prime}}}< \infty. $$
(4)

Lemma 2

([1, 15, 17])

Let f and g be non-negative measurable functions on \(\mathbb {R}^{n}_{k,+}\). Then

$$ \int_{\mathbb{R}^{n}_{k,+}}f(x)g(x) \bigl(x'\bigr)^{\gamma} \,dx\leq \int _{0}^{\infty}f^{\ast,\gamma}(t)g^{\ast,\gamma}(t) \,dt $$
(5)

and

$$ \int_{0}^{\infty}f^{\ast,\gamma}(t)\frac{1}{ ( \frac {1}{g} )^{\ast,\gamma}(t)} \,dt \leq \int_{\mathbb {R}^{n}_{k,+}}f(x)g(x) \bigl(x'\bigr)^{\gamma} \,dx. $$
(6)

Lemma 3

([19])

Let \(1\leq p_{1}< p_{2}<\infty\) and \(1\leq q_{1}< q_{2}<\infty\). A sublinear operator T satisfies weak-type hypotheses \((p_{1},q_{1})\) and \((p_{2},q_{2})\) if and only if

$$ (Tf)^{\ast,\gamma}(t)\leq C \biggl(t^{-\frac{1}{q_{1}}} \int _{0}^{t^{\frac{\sigma_{1}}{\sigma_{2}}}}s^{\frac{1}{p_{1}}-1}f^{\ast ,\gamma}(s) \,ds+ t^{-\frac{1}{q_{2}}} \int_{t^{\frac{\sigma_{1}}{\sigma_{2}}}}^{\infty }s^{\frac{1}{p_{2}}-1}f^{\ast,\gamma}(s) \,ds \biggr), $$
(7)

where \(\sigma_{1}=\frac{1}{q_{1}}-\frac{1}{q_{2}}\) and \(\sigma_{2}=\frac {1}{p_{1}}-\frac{1}{p_{2}}\).

3 Two weighted inequalities for B-fractional integrals

In this section we prove an analog of Heinig’s result for the B-fractional integral \(I_{\alpha,\gamma} f\). Further, the Stein-Weiss inequality for B-Riesz potential is proved as an application of this result. In the following theorem we formulate analog of the Heinig’s result for the B-fractional integral \(I_{\alpha,\gamma} f\).

Theorem 1

Let \(0<\alpha<n+|\gamma|\), \(1< r<\frac{n+|\gamma|}{\alpha}\), \(1< p\leq q<\infty\). Suppose that u and v are non-negative locally integrable functions on \(\mathbb{R}^{n}_{k,+}\) with conditions

$$ \sup_{s>0} \biggl( \int_{s}^{\infty}u^{\ast,\gamma }(t)t^{-q(1-\frac{\alpha}{n+|\gamma|})}\,dt \biggr)^{\frac{1}{q}} \biggl( \int_{0}^{s} \biggl(\biggl(\frac{1}{v} \biggr)^{\ast,\gamma}(t) \biggr)^{p'-1}\,dt \biggr)^{\frac{1}{p'}}< \infty $$
(8)

and

$$ \sup_{s>0} \biggl( \int_{0}^{s}u^{\ast,\gamma}(t)t^{-q(\frac {1}{r}-\frac{\alpha}{n+|\gamma|})} \,dt \biggr)^{\frac{1}{q}} \biggl( \int_{s}^{\infty} \biggl(\biggl(\frac{1}{v} \biggr)^{\ast,\gamma }(t) \biggr)^{p'-1}t^{p'(\frac{1}{r}-1)}\,dt \biggr)^{\frac {1}{p'}}< \infty. $$
(9)

Then \(I_{\alpha,\gamma} \) is a bounded operator from \(L_{p,v,\gamma } ( \mathbb{R}^{n}_{k,+} ) \) to \(L_{q,u,\gamma} ( \mathbb{R}^{n}_{k,+} ) \), that is, there exists a constant \(C>0\) such that for any \(f\in L_{p,v,\gamma} ( \mathbb{R}^{n}_{k,+} )\),

$$\|I_{\alpha,\gamma} f\|_{q,u,\gamma}\leq C\|f\|_{p,v,\gamma}. $$

Proof

It is known that \(I_{\alpha,\gamma} f\) is an operator of weak type \(( 1, \frac{1}{1-\frac{\alpha}{n+|\gamma |}} )\) and is an operator of strong type \(( r, \frac {1}{\frac{1}{r}-\frac{\alpha}{n+|\gamma|}} )\), where \(1< r<\infty \). Refer to \(I_{\alpha,\gamma} f\), Lemma 3, taking

$$p_{1}=1,\qquad q_{1}=\frac{1}{1-\frac{\alpha}{n+|\gamma|}}, \qquad p_{2}=r,\qquad q_{2}=\frac{1}{\frac{1}{r}-\frac{\alpha}{n+|\gamma|}}. $$

Then

$$\sigma_{1}=\frac{1}{q_{1}}-\frac{1}{q_{2}}=1- \frac{1}{r}, \qquad \sigma _{2}=\frac{1}{p_{1}}- \frac{1}{p_{2}}=1-\frac{1}{r}, \qquad \frac{\sigma _{1}}{\sigma_{2}}=1, $$

and

$$\begin{aligned}& \biggl[ \int_{0}^{\infty} \bigl( (I_{\alpha,\gamma} f )^{\ast,\gamma}(t) \bigr)^{q}u^{\ast,\gamma} (t)\,dt \biggr]^{\frac{1}{q}} \\& \quad \leq C \biggl[ \int_{0}^{\infty} \biggl(t^{\frac{\alpha}{n+|\gamma |}-1} \int_{0}^{t}f^{\ast,\gamma} (s)\,ds + t^{\frac{\alpha }{n+|\gamma|}-\frac{1}{r}} \int_{t}^{\infty}s^{\frac{1}{r}-1}f^{\ast ,\gamma} (s) \,ds \biggr)^{q}u^{\ast,\gamma} (t)\,dt \biggr]^{\frac{1}{q}}. \end{aligned}$$

Applying the Minkowski inequality we obtain

$$\begin{aligned}& \biggl[ \int_{0}^{\infty} \bigl( (I_{\alpha,\gamma} f )^{\ast,\gamma}(t) \bigr)^{q}u^{\ast,\gamma} (t)\,dt \biggr]^{\frac{1}{q}} \\& \quad \leq C \biggl[ \int_{0}^{\infty}u^{\ast,\gamma} (t) t^{(\frac{\alpha }{n+|\gamma|}-1)q} \biggl( \int_{0}^{t}f^{\ast,\gamma} (s)\,ds \biggr)^{q} \,dt \biggr]^{\frac{1}{q}} \\& \qquad {}+C \biggl[ \int_{0}^{\infty}u^{\ast,\gamma} (t)t^{ (\frac {\alpha}{n+|\gamma|}-\frac{1}{r} )q} \biggl( \int_{t}^{\infty}s^{\frac{1}{r}-1}f^{\ast,\gamma} (s) \,ds \biggr)^{q} \,dt \biggr]^{\frac{1}{q}}. \end{aligned}$$
(10)

If we take the notation

$$\xi(t)= u^{\ast,\gamma} (t) t^{(\frac{\alpha}{n+|\gamma|}-1)q},\qquad \psi(t)=f^{\ast,\gamma}(t), \qquad \theta(t)=\frac{1}{ (\frac{1}{v} )^{\ast,\gamma} (t)}, $$

then we have (2) from (8) and applying (1)

$$\begin{aligned}& \biggl[ \int_{0}^{\infty}u^{\ast,\gamma} (t) t^{(\frac{\alpha }{n+|\gamma|}-1)q} \biggl( \int_{0}^{t}f^{\ast,\gamma} (s)\,ds \biggr)^{q} \,dt \biggr]^{\frac{1}{q}} \\& \quad \leq C \biggl( \int_{0}^{\infty}\frac{1}{ (\frac{1}{v} )^{\ast,\gamma} (t)} \bigl(f^{\ast,\gamma}(t) \bigr)^{p}\,dt \biggr)^{\frac{1}{p}}. \end{aligned}$$
(11)

Now if we take

$$\xi(t)= u^{\ast,\gamma} (t) t^{(\frac{\alpha}{n+|\gamma|}-\frac{1}{r})q},\qquad \psi(t)=t^{\frac{1}{r}-1}f^{\ast,\gamma}(t), \qquad \theta (t)=\frac{1}{ (\frac{1}{v} )^{\ast,\gamma} (t)}t^{p (\frac{1}{r}-1 )}, $$

then we have (4) from (9) and applying (3) we can assert that

$$\begin{aligned}& \biggl[ \int_{0}^{\infty}u^{\ast,\gamma} (t)t^{ (\frac{\alpha }{n+|\gamma|}-\frac{1}{r} )q} \biggl( \int_{t}^{\infty}s^{\frac{1}{r}-1}f^{\ast,\gamma} (s) \,ds \biggr)^{q} \,dt \biggr]^{\frac{1}{q}} \\& \quad \leq \biggl( \int_{0}^{\infty} \bigl(t^{\frac{1}{r}-1}f^{\ast,\gamma }(t) \bigr)^{p}\frac{1}{ (\frac{1}{v} )^{\ast,\gamma} (t)}t^{p(\frac{1}{r}-1)}\,dt \biggr)^{\frac{1}{p}} \\& \quad = \biggl( \int_{0}^{\infty}\frac{1}{ (\frac{1}{v} )^{\ast ,\gamma} (t)} \bigl(f^{\ast,\gamma}(t) \bigr)^{p}\,dt \biggr)^{\frac{1}{p}} . \end{aligned}$$
(12)

Combining (10), (11), (12) yields

$$ \biggl[ \int_{0}^{\infty} \bigl( (I_{\alpha,\gamma} f )^{\ast,\gamma}(t) \bigr)^{q}u^{\ast,\gamma} (t)\,dt \biggr]^{\frac{1}{q}} \leq C \biggl( \int_{0}^{\infty}\frac{1}{ (\frac{1}{v} )^{\ast,\gamma} (t)} \bigl(f^{\ast,\gamma}(t) \bigr)^{p}\,dt \biggr)^{\frac{1}{p}}. $$
(13)

Applying (5), (13), and (6) we have

$$\begin{aligned}& \biggl[ \int_{ \mathbb{R}^{n}_{k,+}} \bigl( \bigl(I_{\alpha,\gamma} f(x) \bigr) \bigr)^{q}u (x) \bigl(x'\bigr)^{\gamma}\,dx \biggr]^{\frac{1}{q}} \\& \quad \leq \biggl[ \int_{0}^{\infty} \bigl( (I_{\alpha,\gamma} f )^{\ast,\gamma}(t) \bigr)^{q}u^{\ast,\gamma} (t)\,dt \biggr]^{\frac{1}{q}} \\& \quad \leq C \biggl( \int_{0}^{\infty}\frac{1}{ (\frac{1}{v} )^{\ast,\gamma} (t)} \bigl(f^{\ast,\gamma}(t) \bigr)^{p}\,dt \biggr)^{\frac{1}{p}} \\& \quad \leq C \biggl[ \int_{ \mathbb{R}^{n}_{k,+}} \bigl( f(x) \bigr)^{p}v (x) \bigl(x'\bigr)^{\gamma}\,dx \biggr]^{\frac{1}{p}}. \end{aligned}$$

Thus the proof the theorem is completed. □

In the following theorem we prove the Stein-Weiss inequality for B-fractional integrals by using Theorem 1. Note that the Stein-Weiss inequality for classical Riesz potentials was given in [14]. For B-fractional integrals, this inequality was proved in [9] and [10].

Theorem 2

Let \(0<\alpha<n+|\gamma|\), \(1< p<\frac{n+|\gamma|}{\alpha}\), \(\beta<0\), \(0<\beta+\alpha p<(n+|\gamma|)(p-1)\), \(u(x)=|x|^{\beta }\), and \(v(x)=|x|^{\beta+\alpha p}\), for \(x\in\mathbb{R}^{n}_{k,+}\). Then \(I_{\alpha,\gamma} \) is a bounded operator from \(L_{p,v,\gamma } ( \mathbb{R}^{n}_{k,+} ) \) to \(L_{q,u,\gamma} ( \mathbb{R}^{n}_{k,+} ) \), that is, there exists a constant \(C>0\) such that, for any \(f\in L_{p,v,\gamma} ( \mathbb{R}^{n}_{k,+} )\),

$$\biggl( \int_{ \mathbb{R}^{n}_{k,+}}\bigl\vert I_{\alpha,\gamma} f(x) \bigr\vert ^{p} \vert x\vert ^{\beta}\bigl(x' \bigr)^{\gamma}\,dx \biggr)^{\frac{1}{p}}\leq C \biggl( \int_{ \mathbb{R}^{n}_{k,+}}\bigl\vert f(x) \bigr\vert ^{p} \vert x\vert ^{\beta+\alpha p} \bigl(x'\bigr)^{\gamma }\,dx \biggr)^{\frac{1}{p}}. $$

Proof

It is known that \(|B(0,r)|_{\gamma}=w(n,k,\gamma)r^{n+|\gamma|}\), where \(w(n,k,\gamma)=|B(0,1)|_{\gamma}\). Since \(\beta<0\) we have

$$\begin{aligned} u_{\ast,\gamma}(s) =&\bigl\vert \bigl\{ x \in\mathbb{R}^{n}_{k,+}: |x|^{\beta} >s\bigr\} \bigr\vert _{\gamma}\\ =&\bigl\vert \bigl\{ x:x \in\mathbb{R}^{n}_{k,+},|x|< s^{\frac{1}{\beta}} \bigr\} \bigr\vert _{\gamma}\\ =&\bigl\vert B\bigl(0,s^{\frac{1}{\beta}}\bigr) \bigr\vert _{\gamma}\\ =&w(n,k,\gamma)r^{\frac{n+|\gamma |}{\beta}} \end{aligned}$$

and

$$\begin{aligned} u^{\ast,\gamma}(t) =&\inf\bigl\{ s>0:u _{\ast,\gamma}(s) \leq t\bigr\} \\ =& \biggl( \frac{1}{w(n,k,\gamma)} \biggr)^{\frac{\beta}{n+|\gamma |}}t^{\frac{\beta}{n+|\gamma|}}. \end{aligned}$$

Since \(\beta+\alpha p>0\) we can write

$$\begin{aligned} \biggl(\frac{1}{v} \biggr)_{\ast,\gamma}(s)&=\bigl\vert \bigl\{ x:x \in\mathbb {R}^{n}_{k,+}|x|^{-(\beta+\alpha p)} >s\bigr\} \bigr\vert _{\gamma}\\ &=\bigl\vert \bigl\{ x:x \in\mathbb{R}^{n}_{k,+},|x|< s^{-\frac{1}{\beta+\alpha p}} \bigr\} \bigr\vert _{\gamma}\\ &=\bigl\vert B\bigl(0,s^{-\frac{1}{\beta+\alpha p}}\bigr) \bigr\vert _{\gamma}\\ &=w(n,k,\gamma)r^{-\frac{n+|\gamma|}{\beta+\alpha p}} \end{aligned}$$

and

$$\biggl(\frac{1}{v} \biggr)^{\ast,\gamma}(t)=\inf\biggl\{ s>0: \biggl( \frac{1}{v} \biggr) _{\ast,\gamma}(s) \leq t\biggr\} = \biggl( \frac{1}{w(n,k,\gamma)} \biggr)^{-\frac{\beta+\alpha p}{n+|\gamma|}}t^{-\frac{\beta+\alpha p}{n+|\gamma|}}. $$

Take \(p=q=r\) and examine (8) and (9). Since \(\beta+\alpha p<(n+|\gamma|)(p-1)\) we have \(\frac{\beta+\alpha p}{n+|\gamma|}-p<-1\) and \(-\frac{\beta+\alpha p}{n+|\gamma |}(p'-1)>-1\). Then

$$\begin{aligned}& \sup_{s>0} \biggl( \int_{s}^{\infty}u^{\ast,\gamma }(t)t^{-q(1-\frac{\alpha}{n+|\gamma|})}\,dt \biggr)^{\frac{1}{q}} \biggl( \int_{0}^{s} \biggl(\biggl(\frac{1}{v} \biggr)^{\ast,\gamma}(t) \biggr)^{p'-1}\,dt \biggr)^{\frac{1}{p'}} \\& \quad = \sup_{s>0} \biggl( \int_{s}^{\infty} \biggl(\frac {1}{w(n,k,\gamma)} \biggr)^{\frac{\beta}{n+|\gamma|}}t^{\frac {\beta}{n+|\gamma|}} t^{-p(1-\frac{\alpha}{n+|\gamma|})}\,dt \biggr)^{\frac{1}{p}} \\& \qquad {}\times \biggl( \int_{0}^{s} \biggl( \biggl(\frac{1}{w(n,k,\gamma )} \biggr)^{-\frac{\beta+\alpha p}{n+|\gamma|}}t^{-\frac{\beta +\alpha p}{n+|\gamma|}} \biggr)^{p'-1}\,dt \biggr)^{\frac{1}{p'}} \\& \quad = \biggl(\frac{1}{w(n,k,\gamma)} \biggr)^{\frac{-\alpha p}{(n+|\gamma|)p}} \sup _{s>0} \biggl(\frac{1}{\frac{\beta +\alpha p}{n+|\gamma|}-p+1}t^{\frac{\beta+\alpha p}{n+|\gamma |}-p+1} \bigg|_{s}^{\infty} \biggr)^{\frac{1}{p}} \\& \qquad {}\times \biggl(\frac{1}{-\frac{\beta+\alpha p}{n+|\gamma |}(p'-1)+1}t^{-\frac{\beta+\alpha p}{n+|\gamma|}(p'-1)+1} \bigg|_{0}^{s} \biggr)^{\frac{1}{p'}} \\& \quad = \biggl(\frac{1}{w(n,k,\gamma)} \biggr)^{\frac{-\alpha p}{(n+|\gamma|)p}} \biggl( \frac{-1}{\frac{\beta+\alpha p}{n+|\gamma|}-p+1} \biggr)^{\frac{1}{p}} \biggl(\frac{1}{-\frac{\beta +\alpha p}{n+|\gamma|}(p'-1)+1} \biggr)^{\frac{1}{p'}} \\& \qquad {}\times\sup_{s>0}s^{ (\frac{\beta+\alpha p}{n+|\gamma |}-p+1 )\frac{1}{p}+ (-\frac{\beta+\alpha p}{n+|\gamma |}(p'-1)+1 )\frac{1}{p'}} \\& \quad = \biggl(\frac{1}{w(n,k,\gamma)} \biggr)^{\frac{-\alpha p}{(n+|\gamma|)p}} \biggl( \frac{-1}{\frac{\beta+\alpha p}{n+|\gamma|}-p+1} \biggr)^{\frac{1}{p}} \biggl(\frac{1}{-\frac{\beta +\alpha p}{n+|\gamma|}(p'-1)+1} \biggr)^{\frac{1}{p'}}< \infty. \end{aligned}$$

Now examine (9). Since \(\beta+\alpha p>0\) we have \(\frac {\beta+\alpha p}{n+|\gamma|}-1>-1\) and \(-\frac{\beta+\alpha p}{n+|\gamma|}(p'-1)-1<-1\). Then

$$\begin{aligned}& \sup_{s>0} \biggl( \int_{0}^{s}u^{\ast,\gamma}(t)t^{-q(\frac {1}{r}-\frac{\alpha}{n+|\gamma|})} \,dt \biggr)^{\frac{1}{q}} \biggl( \int_{s}^{\infty} \biggl(\biggl(\frac{1}{v} \biggr)^{\ast,\gamma }(t) \biggr)^{p'-1}t^{p'(\frac{1}{r}-1)}\,dt \biggr)^{\frac{1}{p'}} \\& \quad = \sup_{s>0} \biggl( \int_{0}^{s} \biggl(\frac{1}{w(n,k,\gamma )} \biggr)^{\frac{\beta}{n+|\gamma|}} t^{\frac{\beta}{n+|\gamma|}}t^{-p(\frac{1}{p}-\frac{\alpha }{n+|\gamma|})}\,dt \biggr)^{\frac{1}{p}} \\& \qquad {}\times \biggl( \int_{s}^{\infty} \biggl( \biggl(\frac{1}{w(n,k,\gamma )} \biggr)^{-\frac{\beta+\alpha p}{n+|\gamma|}}t^{-\frac{\beta +\alpha p}{n+|\gamma|}} \biggr)^{p'-1}t^{p'(\frac{1}{p}-1)}\,dt \biggr)^{\frac{1}{p'}} \\& \quad = \biggl(\frac{1}{w(n,k,\gamma)} \biggr)^{\frac{\alpha}{n+|\gamma |}}\sup _{s>0} \biggl(\frac{n+|\gamma|}{\beta+\alpha p}t^{\frac{\beta+\alpha p}{n+|\gamma|}}\bigg|_{0}^{s} \biggr)^{\frac{1}{p}} \\& \qquad {}\times \biggl(\frac{n+|\gamma|}{(\beta+\alpha p)(1-p')}t^{-\frac {\beta+\alpha p}{n+|\gamma|}(p'-1)}\bigg|_{s}^{\infty} \biggr)^{\frac{1}{p'}} \\& \quad = \biggl(\frac{1}{w(n,k,\gamma)} \biggr)^{\frac{\alpha}{n+|\gamma |}}\frac{n+|\gamma|}{\beta+\alpha p} \biggl(\frac{1}{p'-1} \biggr)^{\frac{1}{p'}} \sup_{s>0}s^{\frac{\beta+\alpha p}{n+|\gamma|}\frac{1}{p}-\frac{\beta+\alpha p}{n+|\gamma|}\frac{p'-1}{p'}} \\& \quad = \biggl(\frac{1}{w(n,k,\gamma)} \biggr)^{\frac{\alpha}{n+|\gamma |}}\frac{n+|\gamma|}{\beta+\alpha p} \biggl(\frac{1}{p'-1} \biggr)^{\frac{1}{p'}}< \infty. \end{aligned}$$

Therefore (8) and (9) are satisfied and from Theorem 1, we have the result of the corollary. □

References

  1. Heinig, HP: Weighted norm inequalities for classes of operators. Indiana Univ. Math. J. 33(4), 573-582 (1984)

    Article  MathSciNet  MATH  Google Scholar 

  2. Aliev, IA, Gadjiev, AD: On classes of operators types, generated by a generalized shift. Rep. Enlarg. Sess. Semin. I. Vekua Appl. Math. 3(2), 21-24 (1988) (in Russian)

    Google Scholar 

  3. Guliyev, VS: On maximal function and fractional integral, associated with the Bessel differential operator. Math. Inequal. Appl. 6(2), 317-330 (2003)

    MathSciNet  Google Scholar 

  4. Gadjiev, AD, Hajibayov, MG: Inequalities for B-convolution operators. TWMS J. Pure Appl. Math. 1(1), 41-52 (2010)

    MathSciNet  MATH  Google Scholar 

  5. Guliyev, VS, Serbetci, A, Ekincioglu, I: Necessary and sufficient conditions for the boundedness of rough B-fractional integral operators in the Lorentz spaces. J. Math. Anal. Appl. 336, 425-437 (2007)

    Article  MathSciNet  MATH  Google Scholar 

  6. Kipriyanov, IA, Ivanov, LA: The obtaining of fundamental solutions for homogeneous equations with singularities with respect to several variables. In: Trudy Sem. S.L. Sobolev, vol. 1, pp. 55-77. Akad. Nauk SSSR Sibirsk. Otdel Inst. Mat., Novosibirsk (1983) (in Russian)

    Google Scholar 

  7. Kipriyanov, IA, Kononenko, VI: The fundamental solutions for B-elliptic equations. Differ. Uravn. 3(1), 114-129 (1967)

    MathSciNet  MATH  Google Scholar 

  8. Aykol, C, Guliyev, VS, Serbetci, A: The O’Neil inequality for the Hankel convolution operator and some applications. Eurasian Math. J. 4(3), 8-19 (2013)

    MathSciNet  MATH  Google Scholar 

  9. Gadjiev, AD, Guliyev, VS: The Stein-Weiss type inequality for fractional integrals, associated with the Laplace-Bessel differential operator. Fract. Calc. Appl. Anal. 11(1), 77-90 (2008)

    MathSciNet  MATH  Google Scholar 

  10. Gadjiev, AD, Guliyev, VS, Serbetci, A, Guliyev, EV: The Stein-Weiss type inequalities for the B-Riesz potentials. J. Math. Inequal. 5(1), 87-106 (2011)

    Article  MathSciNet  MATH  Google Scholar 

  11. Hajibayov, MG: Inequalities for convolutions of functions on commutative hypergroups. Azerb. J. Math. 4(1), 92-107 (2014)

    MathSciNet  MATH  Google Scholar 

  12. Hajibayov, MG: Generalized potentials on commutative hypergroups. Azerb. J. Math. 5(2), 37-46 (2015)

    MathSciNet  MATH  Google Scholar 

  13. Stein, EM: Singular Integrals and Differentiability Properties of Functions. Princeton Mathematical Series, vol. 30. Princeton University Press, Princeton (1970)

    MATH  Google Scholar 

  14. Stein, EM, Weiss, G: Fractional integrals on n-dimensional Euclidean space. J. Math. Mech. 7, 503-514 (1958)

    MathSciNet  MATH  Google Scholar 

  15. Bennett, C, Sharpley, R: Interpolation of Operators. Pure and Applied Math., vol. 129. Academic Press, Orlando (1988)

    Book  MATH  Google Scholar 

  16. Bradley, JS: Hardy inequalities with mixed norms. Can. Math. Bull. 21(4), 405-408 (1978)

    Article  MathSciNet  MATH  Google Scholar 

  17. Hardy, GH, Littlewood, JE, Polya, G: Inequalities, 2nd edn. Cambridge University Press, Cambridge (1952)

    MATH  Google Scholar 

  18. Tomaselli, G: A class of inequalities. Boll. Unione Mat. Ital. 2, 622-631 (1969)

    MathSciNet  MATH  Google Scholar 

  19. Calderon, AP: Spaces between \(L^{1}\) and \(L^{\infty }\) and the theorem of Marcinkiewicz. Stud. Math. 26, 273-299 (1966)

    MathSciNet  MATH  Google Scholar 

Download references

Acknowledgements

The authors would like to express their gratitude to the referees for their very valuable comments and suggestions. The research of second author was supported by the grant of Presidium of Azerbaijan National Academy of Sciences 2015.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Ahmet Eroglu.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

All authors read and approved the final manuscript.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Eroglu, A., Hajibayov, M.G. & Serbetci, A. Two weighted inequalities for B-fractional integrals. J Inequal Appl 2016, 168 (2016). https://doi.org/10.1186/s13660-016-1104-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13660-016-1104-2

MSC

Keywords