• 検索結果がありません。

Copias de espacios de Orlicz de sucesiones en los espacios de Interpolaci´on A

N/A
N/A
Protected

Academic year: 2022

シェア "Copias de espacios de Orlicz de sucesiones en los espacios de Interpolaci´on A"

Copied!
9
0
0

読み込み中.... (全文を見る)

全文

(1)

Copies of Orlicz sequences spaces in the interpolation spaces A ρ,Φ

Copias de espacios de Orlicz de sucesiones en los espacios de Interpolaci´on A

ρ,Φ

Ventura Echand´ıa ([email protected])

Escuela de Matem´aticas, Facultad de Ciencias, Universidad Central de Venezuela

Jorge Hern´andez ([email protected])

Dpto. T´ecnicas Cuantitativas, DAC, Universidad Centro-Occidental Lisandro Alvarado

Abstract

We prove, by using techniques similar to those in [3], that the inter- polation spaceAρ,Φcontains a copy of the Orlicz sequence spacehΦ. Hereρis a parameter function and Φ is an Orlicz function.

Key words and phrases:Orlicz spaces, Interpolation spaces, param- eter functions.

Resumen

En el presente trabajo, usando t´ecnicas an´alogas a las usadas en [3], demostramos que el espacio de Interpolaci´onAρ,Φ contiene una copia del espacio de Orlicz de sucesioneshΦ. ρdenotar´a una funci´on par´ame- tro y Φ una funci´on de Orlicz.

Palabras y frases clave:Espacios de Interpolaci´on, Espacios de Or- licz, funci´on Par´ametro

1 Introduction

In [3] it was proved that the classical Interpolation spaceAθ,pcontains a copy of`p. Here we are going to give a similar result for Orlicz spaces, for that we need some concepts.

Received 2006/02/22. Revised 2006/08/08. Accepted 2006/08/23.

MSC (2000): Primary 46E30.

This research has been financed by the project number 03-14-5452-2004 of the CDCH-UCV.

(2)

1.1 Orlicz spaces and parameter functions

Definition 1. An Orlicz function Φ is a an increasing, continuous , convex function on [0,∞) such that Φ(0) = 0. Φ is said to satisfy the ∆2-condition at zero if l´ım sup

t→0 Φ(2t)/Φ(t)<∞.

Definition 2. Let Φ be an Orlicz function, the space`Φof all scalar sequences n}n=1 such that

X

n=1

Φ µn|

µ

<∞ for some µ >0, provided with the norm

k{αn}n=1k`

Φ = ´ınf (

µ >0 : X

n=1

Φ µn|

µ

1 )

,

is a Banach space called anOrlicz sequence space.

The closed subspace hΦ of `Φ, consists of all scalar sequences n}n=1 such

that X

n=1

Φ µn|

µ

<∞, for all µ >0.

Remark 1. if Φ satisfy the ∆2-condition at zero, we have that the spaces`Φ

and hΦ coincide, so the result in this work generalize the one in [3] for the case Φ(t) = tpp, p >1.

We have the following result proved in [4]

Proposition 1. Let Φ be an Orlicz function. Then hΦ is a closed subspace of `Φ and the unit vectors{en}n=1 form a symetric basis ofhΦ.

In the following the next concept is very important.

Definition 3. A functionρis called a parameter function, orρ∈BK, if ρis a positive increasing continuous function on (0,∞),such that

Cρ= Z

0

m´ın(1,1 t)ρ(t)dt

t <∞, where ρ(t) = sup

s>0

ρ(st) ρ(t).

(3)

Definition 4. Given ρ∈BK and Φ an Orlicz function, we define the weighted Orlicz sequence space `ρ,Φ, as the space of all scalar sequences m}m∈Z such that

X

m∈Z

Φ

µ m| µρ(2m)

<∞ for some µ >0, equipped with the norm

°°m}m∈Z°

°`ρ,Φ= ´ınf (

µ >0 : X

m∈Z

Φ

µ m| µρ(2m)

1 )

.

1.2 Interpolation spaces

Definition 5. An Interpolation couple A = (A0, A1) consists of two Banach spacesA0andA1which are continuously embedded into a Haussdorff topological vector space V.

The spaceP ¡ A¢

=A0+A1 is endowed with the normK(1, a), where K(t, a) =K¡

t, a;A¢

= ´ınf

a=a0+a1

©ka0kA

0+tka1kA

1 :a0∈A0, a1∈A1

ª,

is the so called Peetre’s K−functional.

For 0< θ <1 and 1≤p <∞,theclassical Interpolation space Aθ,p, consists of those ain P ¡

A¢

, such that

kakθ,p=

 Z

0

(t−θK(t, a))pdt t

p

<∞.

In [2] we introduced the following function norm.

Definition 6. For ρ∈BK and Φ an Orlicz function, the function norm Fρ,Φon¡

(0,∞),dtt¢

is defined by

Fρ,Φ= ´ınf



r >0 : Z

0

Φ

·|u(t)|

rρ(t)

¸dt t 1



,

where uis a measurable function on (0,∞).

(4)

Using the function normFρ,Φ, we introduced in [2] for a Banach pairA, the interpolation space Aρ,Φ, as the space of all a P ¡

A¢

such that Fρ,Φ[K(t, a)]<∞,endowed with the normkakρ,Φ=Fρ,Φ[K(t, a)].

Since fora∈Aρ,Φ,withρ∈BK and Φ an Orlicz function, we have that 1

µK(2m, a) ρ(2m)

1 ρ(2)

ln(2)Φ

µK(2m, a) ρ(2m+1)

2Zm+1

2m

Φ

µK(t, a) ρ(t)

dt t

Φ µ

2K(2m, a) ρ(2m)

¶ , we obtain that, for alla∈Aρ,Φ,

°°{K(2m, a)}m∈Z°

°

`ρ,Φ 2kakρ,Φ 4ρ(2)°

°{K(2m, a)}m∈Z°

°

`ρ,Φ, (1) which gives a discretization of Aρ,Φ.

In this work we use this discretization to prove that the interpolation space Aρ,Φcontains a copy ofhΦ.

2 The main result.

Theorem 1. Let(A0, A1)be a Interpolation couple,ρ∈BK andΦan Orlicz function. We have that ifA0∩A1 is not closed inA0+A1 , then(A0, A1)ρ,Φ

contains a subspace isomorphic tohΦ.

Letε >0; we are going to construct a sequence{xn}n=1 in (A0, A1)ρ,Φ

and a sequence of integer {Nn}n=1, strictly increasing, which satisfies the following conditions

1. kxnkρ,Φ= 1 2. ´ınf

µ>0

( P

|m|>Nn

Φ

³K(2m,xn) µρ(2m)

´

1 )

=

°°

°{K(2m, xn)}|m|>Nn

°°

°`ρ,Φ

2n+2ε

3. ´ınf

µ>0

( P

|m|≤Nn

Φ

³K(2m,xn+1) µρ(2m)

´

1 )

=

°°

°{K(2m, xn+1)}|m|≤Nn

°°

°`ρ,Φ

2n+2ε . For the purpose, supposse we have defined x1, x2, ..., xn, N1, ..., Nn−1, which satisfies the above conditions. Since{K(2m, xn)} ∈`ρ,Φ, i.e.,

´ınf (

µ >0 : X

m∈Z

Φ

µK(2m, xn) µρ(2m)

1 )

<∞,

(5)

there exists 0< µ0<∞,so that X

m∈Z

Φ

µK(2m, xn) µ0ρ(2m)

1;

thus there exists Nn> Nn−1,such that X

|m|>Nn

Φ

µK(2m, xn) µ0ρ(2m)

ε 2n+2

µ 1 µ0

. Therefore

X

|m|>Nn

Φ

µK(2m, xn)

ε 2n+2ρ(2m)

1;

and from this we deduce that ε 2n+2

°°

°{K(2m, xn)}|m|>Nn

°°

°`ρ,Φ

. By using (1) we can findk1, k2>0 so that

k1kxkΣ(A)

°°

°{K(2m, x)}|m|>N

n

°°

°`ρ,Φ

≤k2kxkΣ(A), for allx∈(A0, A1)ρ,Φ.

Let now xn+1(A0, A1)ρ,Φbe such that kxn+1kΣ(A) ε

k22n+2 and kxn+1kρ,Φ= 1, then we have that

°°

°{K(2m, xn+1)}|m|≤Nn

°°

°`ρ,Φ

≤k2kxn+1kΣ(A) ε 2n+2. We have thus contructed the required sequence.

Let us see now that for all sequencesn}n=1,such that all but finitely many are zero, we have that

µ 1

2

k{αn}n=1kh

Φ

°°

°°

° X

n=1

αnxn

°°

°°

°ρ,Φ

(1 +ε)k{αn}n=1kh

Φ (2) This would mean that{xn}n=1is equivalent to the basis{en}n=1 ofhΦ.

(6)

In order to prove the inequality (2) we need the following definitions:

Form∈ Zandx∈Σ(A), put

Hm(x) =K(2m, x);

Hmis an equivalent norm tok..kΣ(A), for eachm∈ Z.

Also we put form∈ Z,

Fm=¡ Σ¡

A¢ , Hm

¢,

i.e. Fmis the space Σ(A) provided with the normHm. Let nowF = (⊕m∈ZFm)`ρ,Φ, i.e.

F=n

{xm}m∈Z:xm∈Fm,k{Hm(xm)}k`ρ,Φ<∞o ,

provided with the norm

°°{xm}m∈Z°

°F =k{Hm(xm)}k`ρ,Φ.

Given n}n=1a scalar sequence such that all but finitely many are zero, we define X ={Xm}m∈Z, Y ={Ym}m∈Z, Zn ={Zmn}m∈Z ∈F,in the following way

1. For eachm∈Z, Xm=P

n=1αnxn

2. Ym=

½ α1x1 if |m| ≤N1

αnxn if Nn−1≤ |m| ≤Nn, n≥2

3. Zm1 = 0,if|m| ≤N1 and Zm1 =α1x1 if|m|> N1

4. Forn≥2,Zmn =

½ 0, if Nn−1≤ |m| ≤Nn

αnxn, otherwise.

We have then that

X =Y + X

n=1

Zn (3)

(7)

and that

kXkF = °

°{Hm(Xm)}m∈Z°

°`ρ,Φ

=

°°

°°

° (

Hm

à X

n=1

αnxn

!)

m∈Z

°°

°°

°`ρ,Φ

=

°°

°°

° (

K(2m, X

n=1

αnxn) )

m∈Z

°°

°°

°

`ρ,Φ

= ´ınf (

λ >0 : X

m∈Z

Φ

µK(2m,P

n=1αnxn) λρ(2m)

1 )

=

°°

°°

° X

n=1

αnxn

°°

°°

°ρ,Φ

.

Moreover, we have that

kZnkF ≤ |αn| ε

2n+1, for eachn≥1.

In fact , for n= 1,we have that X

m∈Z

Φ

µK(2m, Zm1)

1|ρ(2m)

= X

|m|≥N1

Φ

µK(2m, x1) ρ(2m)

< ε 23 < ε

22,

then °

°Z1°

°F ≤ |α1| ε 22. Ifn≥2, we have that

X

m∈Z

Φ

µK(2m, Zmn)

n|ρ(2m)

= X

|m|≤Nn−1

Φ

µK(2m, xn) ρ(2m)

+ X

|m|>Nn

Φ

µK(2m, xn) ρ(2m)

ε

2n+2 + ε

2n+2 = ε 2n+1.

i.e.

1 2n+1 ε

X

m∈Z

Φ

µK(2m, Zmn)

n|ρ(2m)

X

m∈Z

Φ

µ K(2m, Zmn)

n|2n+1ε ρ(2m)

, therefore

kZnkF ≤ |αn| ε 2n+1.

(8)

Using the H¨older inequality we get X

n=1

kZnkF ≤ε 2

X

n=1

n| 2n ≤ε

2

°°n}n=1°

°hΦ

°°

°°

½ 1 2n

¾

n=1

°°

°°

hΨ

ε 2

°°n}n=1°

°hΦ,

where Ψ is the complementary function of Φ.

Since we have that kYkF

X n=1

kZnkF ≤ kXkF ≤ kYkF + X n=1

kZnkF,

we obtain that kYkF−ε

2

°°n}n=1°

°hΦ≤ kXkF ≤ kYkF +ε 2

°°n}n=1°

°hΦ. (4) For n= 1 we have that

1 = °

°{K(2m, x1)}m∈Z°

°

°°

°{K(2m, x1)}|m|≤N

1

°°

°`ρ,Φ+

°°

°{K(2m, x1)}|m|>N

1

°°

°`ρ,Φ

°°

°{K(2m, x1)}|m|≤N

1

°°

°`ρ,Φ+ ε 22, i.e.

1 ε 22

°°

°{K(2m, x1)}|m|≤N

1

°°

°`ρ,Φ 1, and forn≥2 we have that

1 = °°{K(2m, xn)}m∈Z°°

`ρ,Φ

=

°°

°{K(2m, xn)}|m|≤Nn−1+{K(2m, xn)}Nn−1≤|m|≤Nn+{K(2m, xn)}|m|>Nn

°°

°`ρ,Φ

ε

2n+1+

°°

°{K(2m, xn)}Nn−1≤|m|≤Nn

°°

°`ρ,Φ+ ε 2n+2, i.e.

1

22 1 3ε 2n+2

°°

°{K(2m, xn)}N

n−1≤|m|≤Nn

°°

°`ρ,Φ1.

Now using the fact that

kYkF = °

°{Hm(Ym)}m∈Z°

°`ρ,Φ

=

°°

°©

K(2m, Ym)ª

m∈Z

°°

°`ρ,Φ

=

°°

°°

°

©K(2m, α1x1)ª

|m|≤N1+ X n=2

µ©

K(2m, αnxn)ª

Nn−1≤|m|≤Nn

¶°°°°

°

`ρ,Φ

°°

°©

K(2m, α1x1)ª

|m|≤N1

°°

°

`ρ,Φ

+

°°

°°

° X n=2

µ©

K(2m, αnxn)ª

Nn−1≤|m|≤Nn

¶°°°

°°

`ρ,Φ

=

°°

°©

1|K(2m, x1)ª

|m|≤N1

°°

°

`ρ,Φ

+

°°

°°

° X n=2

µ©

n|K(2m, xn)ª

Nn−1≤|m|≤Nn

¶°°°°°

`ρ,Φ

,

(9)

we get, by replacing in (4), that µ

13ε 2

¶°

°n}n=1°

°hΦ≤ kXkF

³ 1 +ε

2

´ °°n}n=1°

°hΦ,

which means µ

13ε 2

¶ °°

°°

° X n=1

αnen

°°

°°

°hΦ

°°

°°

° X n=1

αnxn

°°

°°

°ρ,Φ

(1 +ε)

°°

°°

° X n=1

αnen

°°

°°

°hΦ

, as desired.

References

[1] Beauzamy, B., Espaces d’Interpolation R´eel, Topologie et G´eometrie, Lecture Notes in Math., 666, Springer-Verlag, 1978.

[2] Echand´ıa V., Finol C., Maligranda L., Interpolation of some spaces of Orlicz type, Bull. Polish Acad. Sci. Math., 38, 125–134, 1990.

[3] Levy M., L’espace d’Interpolation r´eel(A0, A1)θ,pcontain `p,C. R. Acad. Sci.

Paris S´er. A 289, 675–677, 1979.

[4] Lindenstrauss J., Tzafriri L.,Classical Banach Spaces, Vol. I, Springler-Verlag, 1977.

[5] Peetre, J., A theory of interpolation of normed spaces, Notas de Matematica Brazil 39, 1–86, 1968.

参照

関連したドキュメント

El análisis de supervivencia tiene entre sus objetivos encontrar esta función, que describe el riesgo de cambio de estado en diferentes periodos de tiempo y representa una secuencia

En el art´ıculo [3] con Horacio Porta y Gustavo Corach, iniciamos en 1991 (pues este trabajo data aproximadamente de julio del 91) el estudio del espacio de los operadores sim´

Mediante la f´ ormula anterior, se va a calcular el efecto fill-in en L t para distintas mallas utilizando el algoritmo Go-Away con CM y comparando los resultados obtenidos con

El uso del teorema de la aplicaci´ on contractiva, aplicado al respectivo operador generado por las dicotom´ıas presentadas en este trabajo, genera soluciones acotadas para una clase

Para estimar la edad de inicio de la Enfermedad de Alzheimer (EA) familiar cuando las censuras son arbitrarias (censura a derecha, a izquierda o en intervalo), ¿cuál es el cambio en

La ecuaci´ on de Schr¨ odinger es una ecuaci´ on lineal de manera que el caos, en el mismo sentido que aparece en las leyes cl´ asicas, no puede hacer su aparici´ on en la mec´

Se presenta una s´ıntesis de las principales caracter´ısticas que se in- cluyen al realizar un an´alisis del proceso de Galton-Watson: el tiempo de extinci´on del proceso,

En particular, observaron que cuando el punto C (el cu´ al es un punto sobre el lugar geom´etrico) se mueve a lo largo del segmento DE, existe una parte del segmento en donde al