- Research
- Open Access
- Published:
On the general K-interpolation method for the sum and the intersection
Journal of Inequalities and Applications volume 2016, Article number: 307 (2016)
Abstract
Let \((A_{0}, A_{1})\) be a compatible couple of normed spaces. We study the interrelation of the general K-interpolation spaces of the couple \((A_{0}+A_{1}, A_{0} \cap A_{1})\) with those of the couples \((A_{0}, A_{1} )\), \((A_{0}+A_{1}, A_{0} )\), \((A_{0}+A_{1}, A_{1} )\), \((A_{0}, A_{0} \cap A_{1})\), and \((A_{1}, A_{0} \cap A_{1})\).
1 Introduction
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, i.e. we assume that both \(A_{0}\) and \(A_{1}\) are continuously embedded in a topological vector space \(\mathscr{A}\). The sum of \(A_{0}\) and \(A_{1}\), denoted by \(A_{0}+ A_{1}\), is the set of elements \(f \in \mathscr{A} \) that can be represented as \(f=f_{0}+f_{1}\) where \(f_{0} \in A_{0}\) and \(f_{1} \in A_{1}\). The norm on the sum space \(A_{0}+ A_{1}\) is given by
The norm on the intersection space \(A_{0} \cap A_{1}\) is given by
The Peetre’s K-functional is defined, for each \(f\in A_{0}+A_{1}\) and \(t>0\), by
Let Φ be a normed space of Lebesgue measurable functions, defined on \((0,\infty)\), with monotone norm: \(|g|\leq |h|\) implies \(\|g\|_{\Phi}\leq \|h\|_{\Phi}\). Further assume that
By definition, the general K-interpolation space \((A_{0},A_{1})_{\Phi}\) is a subspace of \(A_{0}+A_{1} \) having the following norm:
Here Φ is often termed the parameter of the K-interpolation method. We refer to [1] for a complete treatment of the general K-interpolation method.
Set
Let \((\theta,p)\in \Gamma\), then the classical scale of K-interpolation spaces \((A_{0}, A_{1})_{\theta,q}\) (see [2] or [3]) is obtained when Φ is taken to be the weighted Lebesgue space \(L_{q}(t^{-\theta})\) defined by the norm
The following identity was proved by Maligranda [4]:
where
and
Subsequently, Maligranda [5] considered the K-interpolation spaces \((A_{0}, A_{1})_{\varrho, p}\), which are obtained when Φ is given by
and extended the identity (1.2) by imposing certain monotonicity conditions on the parameter function ϱ. Another related identity, proved by Persson [6], states that
Recently, Haase [7] has completely described how the classical K-interpolation spaces for the couples \((A_{0}, A_{1} )\), \((A_{0}+A_{1}, A_{0} )\), \((A_{0}+A_{1}, A_{1} )\), \((A_{0}, A_{0} \cap A_{1})\), \((A_{1}, A_{0} \cap A_{1})\), and \(({A_{0}+A_{1}}, A_{0} \cap A_{1})\) interrelate. The assertions (1.5)-(1.12) in [7], Theorem 1.1, concern the spaces \((A_{0}+A_{1}, A_{0}\cap A_{1})_{\theta, p}\), and the goal of this paper is to extend these assertions by means of replacing the classical scale \((A_{0}, A_{1})_{\theta, p}\) by the general scale \((A_{0}, A_{1})_{\Phi}\).
The main ingredient of our proofs will be the estimate in Proposition 2.4 (see below) which relates the K-functional of the couple \((A_{0}+A_{1}, A_{0}\cap A_{1})\) with that of the original couple \((A_{0}, A_{1})\), whereas this estimate has not been used in [7]. Consequently, our arguments of the proofs are different from those in [7].
We will also apply our general results to the limiting K-interpolation spaces \((A_{0}, A_{1})_{0,p;K}\) and \((A_{0}, A_{1})_{1,p;K}\) recently introduced by Cobos, Fernández-Cabrera, and Silvestre [8]. Namely, if the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) are given by the norms
and
where \(1\leq p <\infty\), then \((A_{0}, A_{1})_{\Phi_{0}}= (A_{0}, A_{1})_{0,p;K}\) and \((A_{0}, A_{1})_{\Phi_{1}}= (A_{0}, A_{1})_{1,p;K}\). Note that, for limiting values \(\theta=0,1\), the space \((A_{0}, A_{1})_{\theta, p}\) is trivial (containing only zero element) when p is finite. The space \((A_{0}, A_{1})_{0,p;K}\) corresponds to the limiting value \(\theta=0\), and the space \((A_{0}, A_{1})_{1,p;K}\) corresponds to the limiting value \(\theta=1\). We will, for convenience, write \((A_{0}, A_{1})_{\{0\}, p}\) for \((A_{0}, A_{1})_{0,p;K}\), and \((A_{0}, A_{1})_{\{1\}, p}\) for \((A_{0}, A_{1})_{1,p;K}\).
The paper is organised as follows. In Section 2, we establish all necessary background material, whereas Section 3 contains the main results.
2 Background material
In the following we will use the notation \(A \lesssim B\) for non-negative quantities to mean that \(A\leq c B\) for some positive constant c which is independent of appropriate parameters involved in A and B. If \(A \lesssim B\) and \(B \lesssim A\), we will write \(A\approx B\). Moreover, we will use the symbol \(X \hookrightarrow Y\) to show that X is continuously embedded in Y.
The elementary but useful properties of the K-functional are collected in the following proposition.
Proposition 2.1
[3]
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces. Then \(K(t,f; A_{0},A_{1})\) is non-decreasing in t, and \(K(t,f;A_{0},A_{1})/t\) is non-increasing in t. Moreover, we have
In the next three propositions, we describe some formulas which relate the K-functional of the couples \((A_{0}+A_{1}, A_{1})\), \((A_{0},A_{0}\cap A_{1})\), and \((A_{0}+A_{1}, A_{0}\cap A_{1})\) with that of the original couple \((A_{0},A_{1})\).
Proposition 2.2
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and let \(f\in A_{0} + A_{1}\). Then
Proof
In view of (2.3), the proof follows immediately from the following relation:
which has been derived in [7], Lemma 2.1. □
For the proof of the next result, we refer to [7], Lemma 2.3.
Proposition 2.3
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and let \(f\in A_{0}\). Then
The next result is derived in [4], Theorem 3.
Proposition 2.4
Let \((A_{0},A_{1})\) be a compatible couple of normed space, and let \(f\in A_{0}+A_{1}\). Then
In our proofs, we will make use of the fact that, for a parameter space Φ, both \(\|s\chi_{(0,1)}(s)\|_{\Phi}\) and \(\|\chi_{(1,\infty)}\|_{\Phi}\) are finite. This fact is a simple consequence of (1.1). Moreover, in view of the monotonicity of the norm \(\|\cdot\|_{\Phi}\) and the fact that \(K(t,f;A_{0},A_{1})=\|f\|_{A_{0}+A_{1}}\), we have
We will make use of the next result, without explicitly mentioning it, in our proofs.
Proposition 2.5
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and assume that \(A_{1} \hookrightarrow A_{0}\). Then
Proof
It will suffice to derive
as the converse estimate is trivial. Using (2.6) and (2.1), we get
as our assumption \(A_{1} \hookrightarrow A_{0}\) implies that \(\|f\|_{A_{0}}\approx \|f\|_{A_{0}+A_{1}}\), so
Since \(K(t,f;A_{0},A_{1})/t\) is non-increasing in t, we obtain
which gives
Now (2.7) follows from (2.8) and (2.9). The proof is complete. □
3 Main results
Theorem 3.1
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces. Then, for an arbitrary parameter space Φ, we have with equivalent norms
Proof
Put \(B_{0}= (A_{0} + A_{1},A_{0})_{\Phi}\), \(B_{1}= (A_{0} + A_{1},A_{1})_{\Phi}\), and \(B=(A_{0} + A_{1},A_{0}\cap A_{1})_{\Phi}\). Let \(f \in A_{0}+A_{1}\). Then by Proposition 2.4
next making use of (2.3), we arrive at
Finally, appealing to Proposition 2.2, we get
which concludes the proof. □
Remark 3.2
The result of Theorem 3.1 generalizes the assertion (1.5) in [7], Theorem 1.1.
Theorem 3.3
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces. Then, for an arbitrary parameter space Φ, we have with equivalent norms
Proof
Put \(B_{0}=(A_{0}, A_{0}\cap A_{1})_{\Phi}\), \(B_{1}=(A_{1}, A_{0}\cap A_{1})_{\Phi}\) and \(B=(A_{0} + A_{1},A_{0}\cap A_{1})_{\Phi}\). Let \(f \in B_{0}+B_{1}\), and take an arbitrary decomposition \(f = f_{0} + f_{1}\) with \(f_{0} \in B_{0}\) and \(f_{1} \in B_{1}\). Then by (2.4), we have
now applying the simple fact that
we obtain
from which the estimate \(\|f\|_{B}\lesssim \|f\|_{B_{0}+ B_{1}}\) follows as the decomposition \(f = f_{0} + f_{1}\) is arbitrary. In order to establish the converse estimate, we take \(f \in B\) and note that there exists (by definition of the norm on \(A_{0}+A_{1}\)) a particular decomposition \(f=f_{0}+f_{1}\) with \(f_{0} \in A_{0}\) and \(f_{1} \in A_{1}\) such that
By Proposition 2.3,
since \(f_{0}=f-f_{1}\), we get by (2.4)
next we use (2.2) to obtain
and, using (3.1), we get
in accordance with (2.9), we deduce that
Analogously, we can obtain
Therefore, combining the previous two estimates, we find that
from which, in view of Proposition 2.4, it follows that
which completes the proof. □
Remark 3.4
The result of Theorem 3.3 generalizes the assertion (1.6) in [7], Theorem 1.1.
In order to formulate the further results, we need the following conditions on the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\):
- (C1):
-
\(\|\chi_{(0,1)}(s)g(s)\|_{\Phi_{0}}\lesssim \|\chi_{(0,1)}(s)g(s)\|_{\Phi_{1}}\).
- (C2):
-
\(\|\chi_{(0,1)}(s)g(s)\|_{\Phi_{1}}\lesssim \|\chi_{(0,1)}(s)g(s)\|_{\Phi_{0}}\).
- (C3):
-
\(\|\chi_{(1,\infty)}(s)g(s)\|_{\Phi_{1}}\lesssim\|\chi_{(1,\infty)}(s)g(s)\|_{\Phi_{0}}\).
- (C4):
-
\(\|\chi_{(1,\infty)}(s)g(s)\|_{\Phi_{0}}\approx \|\chi_{(0,1)}(s)sg(1/s)\|_{\Phi_{1}}\).
- (C5):
-
\(\|\chi_{(1,\infty)}(s)g(s)\|_{\Phi_{1}}\approx \|\chi_{(0,1)}(s)sg(1/s)\|_{\Phi_{0}}\).
Remark 3.5
Let \((\theta, p)\in \Gamma\), and assume that \(\Phi_{0}\) and \(\Phi_{1}\) are given by the norms
and
Then it is easy to see that (C1) and (C3) hold for \((\theta,p) \in \Gamma_{1}\), and (C2) holds for \((\theta,p) \in \Gamma_{2}\). The conditions (C4) and (C5) hold trivially for all \((\theta,p) \in \Gamma\).
Remark 3.6
Let \(1 \leq p < \infty\), and assume that \(\Phi_{0}\) and \(\Phi_{1}\) are given by (1.3) and (1.4). Then we note that (C1), (C3), (C4), and (C5) hold.
Theorem 3.7
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and assume that the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) satisfy (C1), (C3) and (C4). Then we have with equivalent norms
Proof
Put \(B_{0}=(A_{0},A_{1})_{\Phi_{0}}\), \(B_{1}=(A_{0},A_{1})_{\Phi_{1}}\) and \(B= (A_{0} + A_{1},A_{0}\cap A_{1})_{\Phi_{1}}\). Let \(f\in A_{0}+ A_{1}\). Then
which, in view of (C1) and (C3), reduces to
at this point we use (C4) to obtain
finally, applying Proposition 2.4, we conclude that
The proof is complete. □
Remark 3.8
Applying Theorem 3.7 to the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) given by (3.2) and (3.3), we get back the result (1.7) in [7], Theorem 1.1, for \((\theta,p)\in \Gamma_{1}\). Note that the case when \((\theta,p)\in \Gamma_{2}\) follows from the case when \((\theta,p)\in \Gamma_{1}\) by replacing θ by \(1- \theta\).
Corollary 3.9
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and let \(1\leq p<\infty\). Then we have with equivalent norms
Proof
The proof follows by applying Theorem 3.7 to the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) given by (1.3) and (1.4). □
Theorem 3.10
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and assume that the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) satisfy (C1), (C3), and (C5). Then we have with equivalent norms
Proof
Put \(B_{0}=(A_{0},A_{1})_{\Phi_{0}}\), \(B_{1}=(A_{0},A_{1})_{\Phi_{1}}\) and \(B= (A_{0} + A_{1},A_{0}\cap A_{1})_{\Phi_{0}}\). Let \(f \in B_{0} + B_{1}\), and write \(f = f_{0} + f_{1}\), where \(f_{0} \in B_{0}\) and \(f_{1}\in B_{1}\). Now by Proposition 2.4, we have
using (C5) gives
since \(f=f_{0}+f_{1}\), so by (2.4), we have
by (C1) and (C3), we arrive at
which gives
from which the estimate \(\|f\|_{B}\lesssim \|f\|_{B_{0}+ B_{1}}\) follows. To derive the other estimate, take \(f \in B\), and choose a particular decomposition \(f=f_{0}+f_{1}\), with \(f_{0} \in A_{0}\) and \(f_{1} \in A_{1}\), satisfying (3.1). Then
where we have used (2.1). Next proceeding in the same way as in the proof of Theorem 3.3, we obtain
Also, we can show that
which, in view of (C5), becomes
which, combined with (3.4), yields
which, in view of Proposition 2.4, gives
from which the desired estimate \(\|f\|_{B_{0}+B_{1}}\lesssim \|f\|_{B}\) follows. The proof of the theorem is finished. □
Remark 3.11
Theorem 3.10, applied to the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) given by (3.2) and (3.3), gives back (1.8) in [7], Theorem 1.1.
Corollary 3.12
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and let \(1\leq p<\infty\). Then we have with equivalent norms
Proof
Apply Theorem 3.10 to the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) given by (1.3) and (1.4). □
Theorem 3.13
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and assume that the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) satisfy (C1). Then we have with equivalent norms
Proof
Denote \(B_{0}=(A_{0},A_{0} \cap A_{1})_{\Phi_{0}}\), \(B_{1}=(A_{0} + A_{1},A_{0}\cap A_{1})_{\Phi_{1}}\), and \(B=(A_{0},A_{0}\cap A_{1})_{\Phi_{1}}\). Let \(f \in A_{0}\). The estimate \(\|f\|_{B_{0}}+\|f\|_{B_{1}}\lesssim \|f\|_{B}\) follows thanks to the condition (C1) and the following simple inequality:
To derive the converse estimate, we apply Proposition 2.3 to obtain
Next, since \(K(t,f;A_{0},A_{1})/t\) is non-increasing in t, observe that
noting \(K(1,f;A_{0},A_{0} \cap A_{1})=\|f\|_{A_{0}}\), we have
By Proposition 2.4, we also have
Finally, combining (3.6), (3.7), and (3.8), we obtain \(\|f\|_{B}\lesssim \|f\|_{B_{0}}+\|f\|_{B_{1}}\). The proof is finished. □
Remark 3.14
By applying Theorem 3.13 to the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) given by (3.2) and (3.3), we get back (1.9) in [7], Theorem 1.1, for \((\theta, p)\in \Gamma_{1}\).
Corollary 3.15
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and let \(1\leq p<\infty\). Then we have with equivalent norms
Proof
Apply Theorem 3.13 to the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) given by (1.3) and (1.4). □
Theorem 3.16
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and assume that the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) satisfy (C2). Then we have with equivalent norms
Proof
It will suffice to establish that \((A_{0},A_{0}\cap A_{1})_{\Phi_{0}}\hookrightarrow (A_{0} + A_{1},A_{0}\cap A_{1})_{\Phi_{1}}\). Let \(f\in (A_{0},A_{0}\cap A_{1})_{\Phi_{0}}\), then by (3.5) we have
consequently, in view of condition (C2), we obtain
which concludes the proof. □
Remark 3.17
For \((\theta, p)\in \Gamma_{2}\), the result (1.9) in [7], Theorem 1.1, follows from Theorem 3.16, applied to the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) given by (3.2) and (3.3).
Corollary 3.18
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and let \(1\leq p<\infty\). Then we have with equivalent norms
Proof
Apply Theorem 3.16 to the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) given by the norms
and
□
Theorem 3.19
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and assume that the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) satisfy (C 2). Then we have with equivalent norms
Proof
Put \(B_{0}=(A_{0}+A_{1}, A_{1})_{\Phi_{0}}\), \(B_{1}=(A_{0} + A_{1},A_{0}\cap A_{1})_{\Phi_{1}}\), and \(B=(A_{0}+A_{1}, A_{1})_{\Phi_{1}}\). Let \(f \in B_{0}+ B_{1}\), and take an arbitrary decomposition \(f = f_{0} + f_{1}\) with \(f_{0} \in B_{0}\) and \(f_{1} \in B_{1}\). Then by (2.4)
using condition (C2) and the fact that
we obtain
whence, since \(f=f_{0}+ f_{1}\) is an arbitrary decomposition, we get \(\|f\|_{B} \lesssim \|f\|_{B_{0}+B_{1}}\). For the converse estimate, let \(f \in B\), and choose a particular decomposition \(f=f_{0}+f_{1}\), with \(f_{0} \in A_{0}\) and \(f_{1} \in A_{1}\), satisfying (3.1). By Proposition 2.4,
using (2.1), we obtain
which, since \(f_{0}=f-f_{1}\), gives
now using (2.2), it follows that
Using (2.2) also gives
which, together with (3.11), leads to
whence, in view of (3.1), it follows that
according to 2.9, we arrive at
appealing to Proposition 2.2 yields
from which the desired estimate \(\|f\|_{B_{0}+ B_{1}} \lesssim \|f\|_{B}\) follows. The proof is complete. □
Remark 3.20
We recover (1.10) in [7], Theorem 1.1, for \((\theta, p)\in \Gamma_{2}\), by an application of Theorem 3.19 to the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) given by (3.2) and (3.3).
Corollary 3.21
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and let \(1\leq p<\infty\). Then we have with equivalent norms
Proof
Apply Theorem 3.19 to the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) given by (3.9) and (3.10). □
Theorem 3.22
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and assume that the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) satisfy (C1). Then we have with equivalent norms
Proof
It suffices to show that
Let \(f\in (A_{0} + A_{1}, A_{0}\cap A_{1})_{\Phi_{1}}\). Then, using condition (C1) and the elementary fact that
we have
which finishes the proof. □
Remark 3.23
Theorem 3.22, applied to the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) given by (3.2) and (3.3), gives back (1.10) in [7], Theorem 1.1, for \((\theta, p)\in \Gamma_{1}\).
Corollary 3.24
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and let \(1\leq p<\infty\). Then we have with equivalent norms
Proof
Apply Theorem 3.22 to the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) given by (1.3) and (1.4). □
Theorem 3.25
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and assume that the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) satisfy (C1). Then we have with equivalent norms
where
Proof
Set \(B_{0}=(A _{0}, A_{0}\cap A_{1})_{\Phi_{0}}\) and \(B_{1}=(A_{0} + A_{1}, A_{0} \cap A_{1})_{\Phi_{1}}\). Let \(f \in B_{0} +B_{1}\), and write \(f = f_{0} + f_{1}\) with \(f_{0} \in B_{0}\) and \(f_{1} \in B_{1}\). Making use of (2.4), we have
where
and
The condition (C1), along with the following simple inequality:
implies that
Next we observe that \(f_{0} \in A_{0}\) as \(B_{0} \subset A_{0}\). Therefore, we can apply (2.1) to arrive at
The proof of the estimate
is the same as that of (3.7). Finally, inserting estimates (3.13) and (3.14) in (3.12) and then using (3.15) and Proposition 2.4, we get
which gives the estimate \(\|f\|_{(A_{0},A_{1})_{\Psi}}\lesssim \|f\|_{B_{0}+B_{1}}\). In order to prove the other estimate, we take \(f\in (A_{0},A_{1})_{\Psi}\), and select a particular decomposition \(f = f_{0} + f_{1}\), with \(f_{0} \in A_{0}\) and \(f_{1} \in A_{1}\), satisfying condition (3.1). Then proceeding in the same way as in the proof of Theorem 3.3, we obtain
Also, we have
Therefore, these estimates, along with the definition of Ψ, imply that
whence we get \(\|f\|_{B_{0}+B_{1}}\lesssim \|f\|_{(A_{0},A_{1})_{\Psi}}\). The proof is finished. □
Remark 3.26
Take \(\Phi_{0}\) and \(\Phi_{1}\) to be given by (3.2) and (3.3), then we see that \(\Psi= \Phi_{0}\). Thus, we recover the result (1.11) in [7], Theorem 1.1, for \((\theta,p)\in \Gamma_{1}\). Since the case when \((\theta,p)\in \Gamma_{2}\) follows from the case when \((\theta,p)\in \Gamma_{1}\), Theorem 3.25 provides a generalization of the assertion (1.11) in [7], Theorem 1.1.
Corollary 3.27
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and let \(1\leq p<\infty\). Then we have with equivalent norms
Proof
Apply Theorem 3.25 to the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) given by (1.3) and (1.4). □
Theorem 3.28
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and assume that the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) satisfy (C2). Then we have with equivalent norms
where
Proof
Set \(B_{0}=(A_{0}+A_{1}, A_{0})_{\Phi_{0}}\) and \(B_{1}= (A_{0} + A_{1},A_{0}\cap A_{1})_{\Phi_{1}}\). Let \(f\in A_{0} + A_{1}\). Applying Proposition 2.2 to the compatible couple \((A_{1},A_{0})\), we get
using (2.3), we have
By Proposition 2.4,
combining this with (3.16) and making use of (C2), we arrive at
which completes the proof. □
Remark 3.29
Theorem 3.28 generalizes the result (1.12) in [7], Theorem 1.1.
Corollary 3.30
Let \((A_{0},A_{1})\) be a compatible couple of normed spaces, and let \(1\leq p<\infty\). Then we have with equivalent norms
Proof
Apply Theorem 3.28 to the parameter spaces \(\Phi_{0}\) and \(\Phi_{1}\) given by (3.9) and (3.10). □
References
Brudnyi, YA, Krugljak, NY: Interpolation Functors and Interpolation Spaces, vol. 1. North-Holland, Amsterdam (1991)
Bergh, J, Löfström, J: Interpolation Spaces. An Introduction. Springer, New York (1976)
Bennett, C, Sharpley, R: Interpolation of Operators. Academic Press, New York (1988)
Maligranda, L: The K-functional for symmetric spaces. In: Interpolation Spaces and Allied Topics in Analysis. Lecture Notes in Math., vol. 1070, pp. 170-182. Springer, Berlin (1984)
Maligranda, L: Interpolation between sum and intersection of Banach spaces. J. Approx. Theory 47, 42-53 (1986)
Persson, LE: Interpolation with a parameter function. Math. Scand. 59, 199-222 (1986)
Haase, M: Identification of some real interpolation spaces. Proc. Am. Math. Soc. 134, 2349-2358 (2006)
Cobos, F, Fernández-Cabrera, LM, Silvestre, P: New limiting real interpolation methods and their connection with the methods associated to the unit square. Math. Nachr. 286, 569-578 (2013)
Author information
Authors and Affiliations
Corresponding author
Additional information
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
All authors have equally contributed toward the article. 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.
About this article
Cite this article
Ahmed, I., Awais, M. On the general K-interpolation method for the sum and the intersection. J Inequal Appl 2016, 307 (2016). https://doi.org/10.1186/s13660-016-1248-0
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/s13660-016-1248-0
MSC
- 46B70
Keywords
- sum space
- intersection space
- K-functional
- the general K-interpolation method
- limiting K-interpolation methods