• 検索結果がありません。

FOURIER TRANSFORMS OF DINI-LIPSCHITZ FUNCTIONS ON

N/A
N/A
Protected

Academic year: 2022

シェア "FOURIER TRANSFORMS OF DINI-LIPSCHITZ FUNCTIONS ON"

Copied!
4
0
0

読み込み中.... (全文を見る)

全文

(1)

Internat. J. Math. & Math. Scl.

VOL. 15 NO. 3 (1992) 609-612 609

FOURIER TRANSFORMS OF DINI-LIPSCHITZ FUNCTIONS ON

VlLENKIN

GROUPS

M.S. YOUNIS

Department

of Mathematics Yamrouk University

Irbid, Jordan

(Received

November30,1987andin revisedform

January

17,

1991)

ABSTRACT. In [4]

we proved some theoremson the Fourier Transforms offunctions satisfying conditions related to the Dini-Lipschitz conditions on the n-dimensional Euclidean space

R

n and the torus group

T . In

this paperwe extendthose theorems forfunctions with Fourier series on Vilenkingroups.

KEY WORDS AND PHRASES.

Dini-Lipschitz Functions,Vilenkin Fourier Series.

1991

AMS SUBJECT CLASSIFICATION CODE.

42c,43.

1.

INTRODUCTION.

Let

f(z)

belong tothe

Lebesgue

spae

LP,1 <

p

_<

2 of functions onthe real line

R

oron the circle group

T

with its usual norm

H"

p. The pth integral modulus of continuity

wp(f,h)is

definedas

In [4] (Theorem 3.3)

weproved thatif

f(z) belongs

to

LP(R)

such that

(f,h) 0(/(Log ),

o

<

1

then theFourierTrsfo

longs

to

L(R)

where

pl(p +

op-

1) < p’= pl(p- 1), > 11

theprint workwe sh extend

ts rt

forctionson

LP(G)

whe

G

is ampt

metrizab]ermension

AH

oup, i.e., Vilen oup.

2.

DEFINITIONS AND NOTATIONS.

Here

we intruceme dtions d notations that

H u Ist

on.

Ts

is by donesince wesh

my

follow

Oewr [I]

d Quek d

Yap [2]

in

ts rt.

t G

a Vilenoup. Thenits du

G

is a

&screte

co.table to.ion oup.

It

is

w o

that one c intruon

G

atable bic t of

n&ghurhs {Gn}

ofthe identity

element

{e}

of

G

such that

G=GoDG

1,

DG2,...,

d =o

G={e}.

On the other

hd,

let

V

dote the latorin

G

of the

suboup Gn

in

G.

Thit is

(2)

610 M. S. OUNIS

known that

{e}=V oCv

I,

C,

and thatI 0V

n=

if all

V,

are finite, the inclusionisproper.

We

introduce thenumbers mo,ml, m2,...,mk such that

mo=l,mk+l=pkmk;

keN,

pkbeingaprime

>_

2. Then evetT

V,

hasm,asits measureand thequotientsubgroup

Vn/V,_

has

P,

forits measuxe.

DEFINITION

2.1.

For zG,

let

(n,z)

denotethe continuouscharacter of z,i.e.

(,,z)G.

The Fouriertransform

(,)

of

f(z)L’ (G)

isdefinedby

2(") I/()(.,)d

G where

(n,z)

isthe complex conjugate of

(n,z).

DEFINITION

2.2.

Let f(z)L

p

(G).

The pth modulus ofcontinuity

o;p(f.k)

isdefinedby

The Lipschitz and the Dini-Lipschitz

ciasses

in

LP(G)

are thosefor which

w),(f,k)=

0

(m -(*)

and

’o Log rail

-1respectively.

DEFINITION

2.3 if evexT

P/

is finiteask--)oo wesay that

G

has the boundedness property

(P).

3.

MAIN RESULTS.

Withthe previous definitions and notations in

hand,

we now provethe

analogue

ofTheorem 3.3 in

[4].

Thuswestatethefollowing

THEOREM

3.1.

Le f(z)eL)’(G),

1

<

p

s

2, such that

wp(f,k) 0(m(*/(Log mk)’),

o

<

a

_<

1.

(.1)

Then

(n) 1)()

for q

P/(p- 1) _> > max(p/(p+ap- 1), 1/7).

PROOF.

Since the Fourier transform of

f(z+h)-f(z)is

given by

f(n)(n,h)-1),

the

Hausdorff-Yotmg

theorem yields

(.)I

q

(.,h)-

1 q

< Mwp(/,k)

q

O(m-q/(Logm, k)’q).

G

Theboundedness property

(P)

for

C

gives

(see Oxmeweer [1], (2)).

ink+

-1

I?(-)1’ 0(m;’/(Log,.)q

ApplyingtheHolder’s inequalitywith

9 s

qfor the lastestimate onearrives at

mk+l

-1

f(.)l 0(.;/(Log mF ) (m- /’)

and thisleadstothe finalestimate

?(.)

0

( (m -- + /) (Log m-).

eG k 0

(3)

FOURIER TRANSFORMS OF DINI-LIPSCHITZ FUNCTIONS 611 If1

a + /V <

o and

7 <

-I theseriesis convergent sincemf

_>

2

f,

and this proves the theorem.

REMARK

3.2.

We

remark here that for special choice of a,7, and

P

like. a

I,

7 1,

P

2,the previous theorem gives special interestingcases. This isquiteobviousandweshallnot deal with it anyfurther.

However,

the special case

P

2 and o

<

c

<

1 is particularly important and deservesspecialconsideration.

4.

FUNCTIONS IN L2(G).

The origin of this sectionis a theorem proved in Titchmarsh

([3]

Theorem85,p.

117)

forfunctionsbelongingto

Lip(a,2)

onthe realline

R. For

further referencewe stateitas.

THEOREM

4.1.

Let

j’(z)Lip

(a,2)

on

R.

Then the conditions

to2(f ,h) O(h’)

O<a<l, h--,o and

[oX+xTl[[

2du

=0(X-2),

asX-,oo

areequivalent.

This theorem was studiedrather extensivelyin

[5]

and

[6]

for functions in

L2(R2),

and

L2(T 2)

respectively, where several conditions of theorderofmagnitudefor theFourier transforms j of

f

provedtobeequivalenttooneanother.

In [4] (Theorems

5.1,

5.2)

we proved an analogue of Theorem 4.1 for the Dini-Lipschitz functions in

L2(R). In

this section weshall prove Theorem 5.2 in

[4]

for functionsin

L2(G).

THEOREM

4.2.

Let f(z)

belongto

L2(G).

Thenthe conditions

w2(J’,k 0(hO/(Log h)7),

hGk

(4.1)

areequivalent.

Here

h

m -1.

PROOF.

That the first implication is tree follows from Theorem 3.1 where it is proved that

I](,,)I o(,(f,))q

n=mk

Taking p q 2 and substitutingfor h

rn

-1 we obtain

(4.2).

We also hint that anargument basedon the Parseval’s identity similar to that of Titchmarsh’s leads independently to thesame result.

To

prove theconverselet

(4.2)

hold. Then

mk+

-1

](n)

2

0(m-2/(Logrn’ 0(m’2+al/(Log rn& + 1) 27) (4.3)

SinceG has the boundedness property

(P);

hence every

Pk =mt + 1link

isfinite for all

keN,

the

sameistrueof

Log P

k. Thustherighthand sides of

(4.2)

and

(4.3)

are thesame. Thisappliesto

estimatesof the form

ink+ 1-1

and

[.(n)[

2

0(m-2a/(Log(m) 7) (4.2)

(4)

612 M. S. OUNIS

To

sumup,bythe Parseval’s identityoneobtains

[[f(z+h)-f(z)[2dz= _ [(n)[2[(n,h)-l[2

G G

oo

mk

+ --01 m’2o/(Log m)27

m/r o

This isequivalentto

(4.1)

uponsubstituting for h

m "1,

heGtand theproofiscomplete.

REMARK

4.2.

We

concludeby indicating that Theorem5.1 in

[4]

istrueforVilenkinFourier

series, since, it canbe deduced asa special caseofTheorem4.1. Wealso add that for 0<c

<

1, Theorems 3.1 and4.1 of thepresentpapercanbeprovedforhigherdifferences of

f(z)eLP(G).

The

statementsand theproofsarealmost straightforward andwillnotbegiven.

ACKNOWLEDGEMENT:

Thisresearchwassupported

by

agrant from Yarmouk University.

REFERENCES

1.

ONNEWEER, C.W.,

Absolute

Convergence

ofFourier Serieson Certain

Groups. II, ]ke

Math,

Joal,

Vol.41

(1974),

679-688.

2.

QUEK, T.S.

and

YAP, L.Y.H.,

Absolute

Convergence

of Vilenkin-Fourier. Series,

J.

Math.

Analysis A_v_vl. Vol.74

(1980)

1-14.

3.

TITCHMARSH, E.C.,

TheoryofFourier

Integrals,

2ndEd. OxfordUniv.

Px’ess,

1948.

4.

YOUNIS, M.S.,

Fourier Transforms ofDini-Lipschitz Functions, ]Jtt.xtAL

J.

Math.

&

Sci.Vol. 9

No.

2

(1986)

301-312.

5.

YOUNIS, M.S.,

FourierTransformsin

L

pspaces,

M.

Phil.Thesis, Chelsea Coll., London, 1970.

6.

YOUNIS, M.S.,

FourierTransformsofLipschitzFunctionsoncompact

Groups,

Ph.D. Thesis,

McMaster

Univ., Hamilton, Ontario,

Canada,

1974.

参照

関連したドキュメント

This note is devoted to the study of geometric properties and the re- lationships between a projective space and an exponential class, both nat- urally associated with the

Now we introduce the concept of n-Lipschitz mapping and prove that the n-Lipschitz map- ping satisfying the n-distance one preserving property is an n-isometry under some

Now we introduce the concept of n-Lipschitz mapping and prove that the n-Lipschitz mapping satisfying the n-distance one preserving property is an n- isometry under some

After Grothendieck’s proof that every vector bundle on P 1 decomposes as a direct sum of line bundles, Atiyah’s groundbreaking paper of 1957 provided an answer to the next case:

We introduce a new regularity condition, of a qualitative type, under which we prove a version of Littlewood’s theorem for tangential approach whose shape may vary from point to

Hardy, “A theorem concerning Fourier transforms,” Journal of the London Mathematical Society, vol.. Morgan, “A note on Fourier transforms,” Journal of the London Mathematical

In this paper, we propose a new and efficient method that is applicable for the computation of the Fourier transform of a function which may possess a singular point or slowly converge

In this paper, we are going to show that this is impossible and that H (without any shift) is the worst distributed net of all the digital (0, m, 2)-nets over Z 2 that are