• No results found

The Solutions of the Series-Like Iterative Equation with Variable Coefficients

N/A
N/A
Protected

Academic year: 2020

Share "The Solutions of the Series-Like Iterative Equation with Variable Coefficients"

Copied!
7
0
0

Loading.... (view fulltext now)

Full text

(1)

Volume 2009, Article ID 173028,7pages doi:10.1155/2009/173028

Research Article

The

C

1

Solutions of the Series-Like Iterative

Equation with Variable Coefficients

Yuzhen Mi, Xiaopei Li, and Ling Ma

Mathematics and Computational School, Zhanjiang Normal University, Zhanjiang, Guangdong 524048, China

Correspondence should be addressed to Xiaopei Li,[email protected]

Received 23 March 2009; Revised 11 June 2009; Accepted 6 July 2009

Recommended by Tomas Dom´ınguez Benavides

By constructing a structure operator quite different from that ofZhang and Baker2000and using the Schauder fixed point theory, the existence and uniqueness of theC1solutions of the series-like

iterative equations with variable coefficients are discussed.

Copyrightq2009 Yuzhen Mi et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

1. Introduction

An important form of iterative equations is the polynomial-like iterative equation

λ1fx λ2f2x · · ·λnfnx Fx, xI: a, b , 1.1

whereFis a given function,fis an unknown function,λi ∈ R1i1,2, . . . , n,andfk k 1,2, . . . , n is the kth iterate of f, that is, f0x x, fkx ffk−1x. The case of all constantλiswas considered in1–10 . In 2000, W. N. Zhang and J. A. Baker first discussed the continuous solutions of such an iterative equation with variable coefficients λi λix which are all continuous in intervala, b .In 2001, J. G. Si and X. P. Wang furthermore gave the continuously differentiable solution of such equation in the same conditions as in11 . In this paper, we continue the works of11,12 , and consider the series-like iterative equation with variable coefficients

i1

(2)

whereλix : I → 0,1 are given continuous functions and∞i1λix 1, λ1xc >

0∀xI, maxxIλix ci.We improve the methods given by the authors in11,12, and the conditions of11,12 are weakened by constructing a new structure operator.

2. Preliminaries

LetC0I,R {f:I R, f is continuous}, clearlyC0I,R, ·

c0is a Banach space, where

f c0maxxI|fx|, forfinC0I,R.

Let C1I,R {f : I → R, f is continuous and continuously differentiable}, then

C1I,Ris a Banach space with the norm ·

c1, where f c1 f c0 f c0, forfinC1I,R. Being a closed subset,C1I, Idefined by

C1I, I fC1I, R, fII,xI 2.1

is a complete space.

The following lemmas are useful, and the methods of proof are similar to those of paper4 , but the conditions are weaker than those of4 .

Lemma 2.1. Suppose thatϕC1I, Iand

ϕxM, xI, 2.2

ϕx1ϕx2M|x

1−x2|,x1, x2∈I, 2.3

whereMandMare positive constants.Then

ϕnx1ϕnx2M 2n−2 in−1

Mi

|x1−x2|, 2.4

for anyx1, x2inI, whereϕndenotesdϕn/dx.

Lemma 2.2. Suppose thatϕ1, ϕ2∈C1I, Isatisfy2.2.Then

ϕn1ϕn2 c0≤

n

i1

Mi−1

ϕ1−ϕ2 c0. 2.5

Lemma 2.3. Suppose thatϕ1, ϕ2∈C1I, Isatisfy2.2and2.3.Then

ϕk11−ϕk21

c0≤k

1Mkϕ1ϕ2c0

Qk1M

k

i1

ki1Mki−1

ϕ1−ϕ2 c0,

2.6

(3)

3. Main Results

For given constantsM1>0 andM2>0, let

AM1, M2

ϕC1I, I:ϕxM1,xI ,

ϕx1ϕx2M

2|x1−x2|,x1, x2∈I

.

3.1

Theorem 3.1existence. Given positive constantsM1, M2 andF ∈ AM1, M2,if there exists constantsN1≥1andN2>0, such that

P1c−∞i2ciN1i−1≥M1/N1,

P2c−∞i2ci2iji21N1jM2/N2,

then1.2has a solutionfinAN1, N2.

Proof. For convenience, letdmax{|a|,|b|}.

DefineK:AN1, N2 → C1I, Isuch thatK:fKf, where

Kft

i1

λixfit,x, tI. 3.2

Sincef ∈ AN1, N2, it is easy to see that|fit| ≤dfor alltI, and|λixfit| ≤d|λix|for allx, tI.It follows from∞i1λix 1 that∞i1λixfitis uniformly convergent. Then

Kftis continuous fortI. Also we have

a

i1

λixa

i1

λixfit

i1

λixbb, 3.3

thusKfC0I, I.

For anyf ∈ AN1, N2, we have

dtdλixfitλixffi−1tfi−1tciN1i. 3.4

By condition P1, we see that ∞i1ciN1i is convergent, therefore∞i1cifit

is uniformly convergent fortI, this implies thatKftis continuously differentiable fortI. Moreover

dtdKft

≤∞

i1

λixfit

i1

(4)

ByLemma 2.1,

dtdKft1

d

dt

Kft2≤ ∞

i1

λixfit1−fit2

≤∞

i1

ci

⎛ ⎝N22i−2

ji−1

N1j

|t1t2|:μ2|t1t2|.

3.6

ThusKf ∈ Aμ1, μ2.

DefineT :AN1, N2 → C1I, Ias follows:

Tft 1

λ1xFt

1

λ1xKft ft,t, xI, 3.7

wheref ∈ AN1, N2. BecauseKf,F,andf are continuously differentiable for alltI,Tf is continuously differentiable for alltI. By conditionsP1andP2, for anyt1, t2inI,we have

dtdTft≤ 1 λ1xF

t 1

λ1x

i2

λixfit≤ 1

cM1

1

c

i2

ciN1i

≤ 1

cM1

1

ccN1−M1 N1.

3.8

We furthermore have

dtdTft1− d dt

Tft2≤ 1

λ1xF

t1Ft2 1

λ1x

i2

ci

fit1−fit2

≤ 1

cM2|t1−t2|

1

c

i2

ciN2

⎛ ⎝2i−2

ji−1

N1j ⎞ ⎠|t1t2|

N2|x1−x2|.

3.9

ThusT :AN1, N2 → AN1, N2is a self-diffeomorphism.

Now we prove the continuity of T under the norm · c1. For arbitrary f1, f2 ∈

(5)

Tf1−Tf2 c0 max

tI

− 1

λ1xKf1t f1t 1

λ1xKf2tf2t

≤ 1

cmaxtI

i2

λixf1it

i2

λixf2it

≤ 1 c

i2

cif1if2ic0

≤ 1

c

i2

ci i

k1

N1k−1

f1f2

c0,

dtdTf1− d dtTf2

c0

max tI

− 1

λ1x

Kf1t

f1t 1

λ1x

Kf2t

f2t

≤ 1

cmaxtI

i2

λixf1it

i2

λixf2it

≤ 1

c

i2

ci

fi

1

fi 2 c0 ≤ 1 c

i2

ci

iN1i−1f1f2c0QiN2 i−1

k1

ikN1ik−2

f1f2

c0

.

3.10

Let

E1 1

c

i2

ci i

k1

N1k−1QiN2 i−1

k1

ikNi1k−2 , E2 1 c

i2

ciiN1i−1, Emax{E1, E2}.

3.11

Then we have

Tf1Tf2

c1Tf1−Tf2c0

Tf1

Tf2

c0≤E1f1−f2c0E2f 1−f2c0

Ef1−f2c0Ef1f2c0 Ef1−f2c1,

3.12

which gives continuity ofT.

It is easy to show thatAN1, N2is a compact convex subset ofC1I, I. By Schauder’s fixed point theorem, we assert that there is a mappingf ∈ AN1, N2such that

ft Tft 1

λ1xFt

1

λ1xKft ft,tI. 3.13

(6)

Theorem 3.2Uniqueness. Suppose that (P1) and (P2) are satisfied, also one supposes that

P3E <1,

then for arbitrary functionFinAM1, M2,1.2has a unique solutionf∈ AN1, N2.

Proof. The existence of 1.2 in AN1, N2 is given by Theorem 3.1, from the proof of

Theorem 3.1, we see thatAN1, N2is a closed subset ofC1I, I, by3.12andP3, we see

thatT :AN1, N2 → AN1, N2is a contraction. ThereforeT has a unique fixed pointfx inAN1, N2, that is,1.2has a unique solution inAN1, N2, this proves the theorem.

4. Example

Consider the equation

i1

λixfix 1 4x

2, xI: 1,1 , 4.1

whereλ1x 33/36 1/36cos2πx/2, λ2x 1/36 1/36sin2πx/2, λ3x 1/36,

λ4x λ5x · · · 0.It is easy to see that 0 ≤λix≤ 1,i1λix 1, c 33/36, c2 2/36, c31/36, c4c5· · ·0.

For anyx, yin−1,1 ,

Fx|0.5x| ≤0.5, FxFy≤ |0.5x|0.5y1, 4.2

thusF∈ A0.5,1.By conditionP1, we can chooseN11.1,and by conditionP1, we can chooseN21.5. Then byTheorem 3.1, there is a continuously differentiable solution of4.1 inA1.1,1.5.

Remark 4.1. HereFxis not monotone forx∈−1,1 , hence it cannot be concluded by11,

12 .

Acknowledgments

This work was supported by Guangdong Provincial Natural Science Foundation07301595 and Zhan-jiang Normal University Science Research ProjectL0804.

References

1 J. Z. Zhang and L. Yang, “Disscussion on iterative roots of continuous and piecewise monotone functions,”Acta Mathematica Sinica, vol. 26, no. 4, pp. 398–412, 1983Chinese.

2 W. N. Zhang, “Discussion on the iterated equationni1λifix Fx,”Chinese Science Bulletin, vol. 32, pp. 1441–1451, 1987Chinese.

3 W. N. Zhang, “Stability of the solution of the iterated equationni1λifix Fx,”Acta Mathematica

Scientia, vol. 8, no. 4, pp. 421–424, 1988.

4 W. N. Zhang, “Discussion on the differentiable solutions of the iterated equationni1λifix Fx,”

(7)

5 W. Zhang, “Solutions of equivariance for a polynomial-like iterative equation,”Proceedings of the Royal Society of Edinburgh. Section A, vol. 130, no. 5, pp. 1153–1163, 2000.

6 M. Kulczycki and J. Tabor, “Iterative functional equations in the class of Lipschitz functions,” Aequationes Mathematicae, vol. 64, no. 1-2, pp. 24–33, 2002.

7 X. P. Li, “A class of iterative equation on a Banach space,”Journal of Sichuan University. Natural Science Edition, vol. 41, no. 3, pp. 505–510, 2004Chinese.

8 W. Zhang, K. Nikodem, and B. Xu, “Convex solutions of polynomial-like iterative equations,”Journal of Mathematical Analysis and Applications, vol. 315, no. 1, pp. 29–40, 2006.

9 B. Xu and W. Zhang, “Decreasing solutions and convex solutions of the polynomial-like iterative equation,”Journal of Mathematical Analysis and Applications, vol. 329, no. 1, pp. 483–497, 2007.

10 X. P. Li and S. F. Deng, “Differentiability for the high dimensional polynomial-like iterative equation,” Acta Mathematica Scientia B, vol. 25, no. 1, pp. 130–136, 2005.

11 W. Zhang and J. A. Baker, “Continuous solutions of a polynomial-like iterative equation with variable coefficients,”Annales Polonici Mathematici, vol. 73, no. 1, pp. 29–36, 2000.

References

Related documents

The explanatory emphasis in each case is not the apparent subject of the question (Mary, the treat), but the whole of which the individual is a part (the movement of the ball,

In order to investigate the naturalized epistemology of science, the questionnaire of Frameworks for Naturalized Epistemology(FNE) was adopted. Itdescribes the

easily be extended by introducing a separate mask genera- tor model. The idea is to generate object shapes based on a bounding box. They can further be used as input to the cur-

The proposed KVM-MN builds an effective local-position-aware image representation using multi-head key-value memory representation as well as cap- tures inherent matching

Food allergy is one of the main public health problem worldwide, and its prevalence appears to be rising. There is a strong relationship between food allergy and asthma, and

Matrix tablets of diclofenac were formulated employing Olibanum resin in different proportions of drug and polymer and the tablets were evaluated for drug

We aim at using a new method for performance measurement by combining balanced scorecard, path analysis, cooperative game theory and evolutionary game theory from an

Results: For the G carriers, influences of the SCNs were observed mostly in the default mode network, of which the peak clusters anchored by the posterior cingulate cortex