Page 1

This article appeared in a journal published by Elsevier. The attached

copy is furnished to the author for internal non-commercial research

and education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling or

licensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of the

article (e.g. in Word or Tex form) to their personal website or

institutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies are

encouraged to visit:

http://www.elsevier.com/copyright

Page 2

Author's personal copy

Applied Mathematics Letters 25 (2012) 466–470

Contents lists available at SciVerse ScienceDirect

Applied Mathematics Letters

journal homepage: www.elsevier.com/locate/aml

Fuzzy real valued lacunary I-convergent sequences

Bipan Hazarika∗

Department of Mathematics, Rajiv Gandhi University, Itanagar-791 112, Arunachal Pradesh, India

a r t i c l ei n f o

Article history:

Received 20 June 2011

Received in revised form 10 September

2011

Accepted 19 September 2011

Keywords:

Ideal

I-convergent

I-Cauchy

Fuzzy number

Lacunary sequence

a b s t r a c t

In this article, we introduce the concept of lacunary I-convergent sequence of fuzzy real

numbers and study some basic properties.

© 2011 Elsevier Ltd. All rights reserved.

1. Introduction

The works on I-convergence of real valued sequences was initially studied by Kostyrko et al. [1]. Later on, it was further

studied by Šalàt et al. [2,3], Tripathy et al. [4] and many others.

Let S be a non-empty set. A non-empty family of sets I ⊆ P(S) (power set of S) is called an ideal on S if (i) for each A,

B ∈ I, we have A ∪ B ∈ I; (ii) for each A ∈ I and B ⊆ A, we have B ∈ I. Let S be a non-empty set. A family F ⊆ P(S) (power

set of S) is called a filter on S if (i) φ ̸∈ F; (ii) for each A, B ∈ F, we have A ∩ B ∈ F; (iii) for each A ∈ F and B ⊇ A, we have

B ∈ F. An ideal I is called non-trivial if I ̸= φ and S ̸∈ I. It is clear that I ⊆ P(S) is a non-trivial ideal if and only if the class

F = F(I) = {S − A : A ∈ I} is a filter on S. The filter F(I) is called the filter associated with the ideal I. A non-trivial ideal

I ⊆ P(S) is called an admissible ideal on S if and only if it contains all singletons, i.e., if it contains {{x} : x ∈ S}. A non-trivial

ideal I is maximal if there cannot exist any non-trivial ideal J ̸= I containing I as a subset (for details see [1]).

The concept of fuzzy sets was initially introduced by Zadeh [5]. Later on, sequences of fuzzy real numbers have been

discussed by Nanda [6], Nuray and Savas [7] and many others.

A lacunary sequence is an increasing integer sequence θ = (kr) such that k0= 0 and hr= kr− kr−1→ ∞ as r → ∞.

The intervals are determined by θ and it will be defined by Jr= (kr−1,kr] and the ratio

Freedman et al. [8] defined the space Nθin the following way. For any lacunary sequence θ = (kr),

k∈Ir

The space Nθis a BK space with the norm

‖(xk)‖θ= sup

r

k∈Ir

N0

kr

kr−1will be defined by φr.

Nθ=

(xk) : lim

r→∞h−1

r

−

|xk− L| = 0, for some L

.

h−1

r

−

|xk|.

θdenotes the subset of these sequences in Nθfor which θ = 0, (N0

θ, ‖ · ‖θ) is also a BK space.

∗Tel.: +91 3602278512; fax: +91 360 2277881.

E-mail address: bh_rgu@yahoo.co.in.

0893-9659/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.

doi:10.1016/j.aml.2011.09.037

Page 3

Author's personal copy

B. Hazarika / Applied Mathematics Letters 25 (2012) 466–470

467

Lemma 1. Every solid space is monotone.

Lemma 2 (Kostyrko et al. [1, Lemma 5.1]). If I ⊂ 2Nis a maximal ideal, then for each A ⊂ N we have either A ∈ I or N − A ∈ I.

2. Definitions and notations

A fuzzy real number X is a fuzzy set on R i.e. a mapping X: R → J (= [0,1]) associating each real number t with its grade

of membership X(t).

A fuzzy real number X is called convex if X(t) ≥ X(s) ∧ X(r) = min(X(s),X(r)), where s < t < r. If there exists t0∈ R

such that X(t0) = 1, then the fuzzy real number X is called normal.

The α-level set of a fuzzy real number X,0 < α ≤ 1 denoted by Xαis defined as Xα= {t ∈ R : X(t) ≥ α}.

A fuzzy real number X is said to be upper semi-continuous if for each ε > 0,X−1([0,a + ε]), for all a ∈ J is open in the

usual topology of R. The set of all upper semi-continuous, normal, convex fuzzy number is denoted by R(J).

Let D denote the set of all closed and bounded intervals X = [x1,x2] on the real line R. For X = [x1,x2] and Y = [y1,y2]

in D, we define X ≤ Y if and only if x1≤ y1and x2≤ y2. Define a metric d on D by

d(X,Y) = max{|x1− y1|,|x2− y2|}.

It is known that (D, d) is a complete metric space and ‘‘≤’’ is a partial order on D.

The absolute value |X| of X ∈ R(J) is defined as

|X|(t) =

Let¯d : R(J) × R(J) → R be defined by

¯d (X,Y) = sup

0≤α≤1

Then¯d defines a metric on R(J).

We define X ≤ Y if and only if Xα≤ Yα, for allα ∈ J. The additive identity and multiplicative identity in R(J) are denoted

by¯0 and¯1, respectively.

A sequence (Xk) of fuzzy real numbers is said to be convergent to a fuzzy real number X0if for every ε > 0, there exists

n0∈ N such that¯d(Xk,X0) < ε, for all k ≥ n0.

A sequence (Xk) of fuzzy real numbers is said to be I-convergent if there exists a fuzzy real number X0such that for each

ε > 0, the set

{k ∈ N :¯d(Xk,X0) ≥ ε} ∈ I.

We write I-lim Xk= X0.

If I = If(class of all finite subsets of N), then If- convergence coincides with the usual convergence.

LetEFdenote the sequencespace of fuzzy numbers. ThenEFis said tobe solid(ornormal) if(Yk) ∈ EF, whenever(Xk) ∈ EF

and |Yk| ≤ |Xk|, for all k ∈ N.

A sequence space EFis said to be symmetric if (Xk) ∈ E implies (Xπ(k)) ∈ E, where π is a permutation of N.

A sequence space EFis said to be monotoneif it contains the canonical preimages of its step space.

Throughout the article, we assume that I is an admissible ideal of N.

max{X(t),X(−t)},

0,

if t > 0;

if t < 0.

d(Xα,Yα).

3. Lacunary I-convergent sequence of fuzzy real numbers

Definition 1. Let θ = (kr) be a lacunary sequence. Then a sequence (Xk) of fuzzy real numbers is said to be lacunary I-

convergent if for every ε > 0 such that

k∈Ir

We write Iθ− limXk= X.

Definition 2. Let θ = (kr) be a lacunary sequence. Then a sequence (Xk) of fuzzy real numbers is said to be lacunary I-null

if for every ε > 0 such that

k∈Ir

We write Iθ− limXk=¯0.

r ∈ N : h−1

r

−

¯d(Xk,X) ≥ ε

∈ I.

r ∈ N : h−1

r

−

¯d(Xk,¯0) ≥ ε

∈ I.

Page 4

Author's personal copy

468

B. Hazarika / Applied Mathematics Letters 25 (2012) 466–470

Definition 3. Letθ = (kr)bealacunarysequence.Thenasequence(Xk)offuzzyrealnumbersissaidtobelacunaryI-Cauchy

if there exists a subsequence (X/

k∈Jr

r(r)) of (Xk) such that k′(r) ∈ Jr, for each r, limr→∞X/

k(r) = X′and for everyε > 0 such that

r ∈ N : h−1

r

−

¯d(Xk,Xk′(r)) ≥ ε

∈ I.

Definition 4. A lacunary sequence θ′= (k′(r)) is said to be a lacunary refinement of the lacunary sequence θ = (kr) if

(kr) ⊂ (k′(r)).

Throughout wF, ℓF

fuzzy real numbers, respectively.

∞, cIF

θand (cIF

0)θdenotes all, bounded, lacunary I-convergent, and lacunary I-null class of sequences of

Theorem 1. A sequence (Xk) of fuzzy real numbers is Iθ-convergent if and only if it is an Iθ-Cauchy sequence.

Proof. Let (Xk) be a sequence of fuzzy real numbers with Iθ− limXk= X.

Write H(i)= {r ∈ N : h−1

Hence for each i,H(i)⊇ H(i+1)and {r ∈ N : h−1

We choose k1such that r ≥ k1, then {r ∈ N : h−1

Next we choose k2> k1such that r ≥ k2, then {r ∈ N : h−1

For each r satisfying k1≤ r < k2, choose k′(r) ∈ Jrsuch that

k′(r)∈Jr

In general, we choose kp+1> kp, such that r > kp+1then

kp+1∈Jr

Then for all r satisfying kp≤ r < kp+1, such that

k′(r)∈Jr

Thus we get k′(r) ∈ Jr, for each r and limr→∞Xk′(r)= X.

Therefore, for every ε > 0, we have

k,k′∈Jr

r

∑

k∈Jr¯d(Xk,X) <

1

i}, for each i ∈ N.

r

∑

∑

k∈Jr¯d(Xk,X) <

k1∈Jr¯d(Xk1,X) < 1} ̸∈ I.

r

1

i} ̸∈ I.

r

∑

k2∈Jr¯d(Xk2,X) <

1

2} ̸∈ I.

r ∈ N : h−1

r

−

¯d(Xk′(r),X) < 1

̸∈ I.

r ∈ N : h−1

r

−

¯d(Xkp+1,X) <

1

p

̸∈ I.

r ∈ N : h−1

r

−

¯d(Xk′(r),X) <

1

p

̸∈ I.

r ∈ N : h−1

r

−

¯d(Xk,Xk′(r)) ≥ ε

⊆

r ∈ N : h−1

r

−

−

k∈Jr

¯d(Xk,X) ≥ε

2

∪

r ∈ N : h−1

r

k′(r)∈Jr

¯d(Xk′(r),X) ≥ε

2

.

i.e.

r ∈ N : h−1

Then (Xk) is an Iθ-Cauchy sequence.

Conversely, suppose (Xk) is an Iθ-Cauchy sequence. Then for every ε > 0, we have

k∈Jr

r

∑

k,k′∈Jr¯d(Xk,Xk′(r)) ≥ ε∈ I.

r ∈ N : h−1

r

−

¯d(Xk,X) ≥ ε

⊆

r ∈ N : h−1

r

−

−

k,k′∈Jr

¯d(Xk,Xk′(r)) ≥ε

2

∪

r ∈ N : h−1

r

k′(r)∈Jr

¯d(Xk′(r),X) ≥ε

2

.

It follows that (Xk) is a Iθ-convergent sequence.

?

Theorem 2. If θ′is a lacunary refinement of a lacunary sequence θ and (Xk) ∈ cIF

θ′, then (Xk) ∈ cIF

θ.

Page 5

Author's personal copy

B. Hazarika / Applied Mathematics Letters 25 (2012) 466–470

469

Proof. Suppose that for each Jrof θ contains the points (k/

r,t)η(r)

t=1of θ′such that

r,t= (k/

kr−1< k/

r,1< k/

r,2< ··· < k/

r,η(r)= kr,

where J/

r,t−1,k/

r,t].

Since kr⊆ (k′(r)), so r, η(r) ≥ 1.

Let(J∗

j)∞

j=1be the sequence of intervals (J/

r,t) ordered by increasing right end points. Since(Xk) ∈ cIF

θ′, then for eachε > 0,

j ∈ N : (h∗

For each ε > 0, we have

j)−1−

J∗

j⊂Jr

¯d(Xk,X) ≥ ε

∈ I.

Also since hr= kr− kr−1, so h/

r,t= k/

r,t− k/

r,t−1.

r ∈ N : h−1

r

−

k∈Jr

¯d(Xk,X) ≥ ε

⊆

r ∈ N : h−1

r

−

k∈Jr

j ∈ N : (h∗

j)−1−

J∗

k∈J∗

j⊂Jr

j

¯d(Xk,X)

≥ ε

.

Therefore {r ∈ N : h−1

Hence (Xk) ∈ cIF

Theorem 3. Let ψ be a set of all lacunary sequences.

(a) If ψ is closed under arbitrary union, then cIF

(b) If ψ is closed under arbitrary intersection, then cIF

(c) If ψ is closed under union and intersection, then cIF

r

∑

k∈Jr¯d(Xk,X) ≥ ε} ∈ I.

θ.

?

µ=

θ∈ψcIF

θ, where µ =

µ⊆ cIF

θ∈ψθ;

ν=

θ∈ψcIF

θ⊆ cIF

θ, where υ =

θ∈ψθ;

ν.

Proof. (a) By hypothesis, we have µ ∈ ψ which is a refinement of each θ ∈ ψ. Then from Theorem 2, we have if (Xk) ∈ cIF

implies that (Xk) ∈ cIF

Thus for each θ ∈ ψ, we get cIF

Hence cIF

(b) By part (a) and Theorem 2, we have cIF

(c) By part (a) and (b), we get cIF

?

µ

θ.

µ⊆ cIF

θ. The reverse inclusion is obvious.

µ=

θ∈ψcIF

θ.

ν=

θ∈ψcIF

θ.

µ⊆ cIF

θ⊆ cIF

ν.

Theorem 4. cIF

θ∩ ℓF

∞is a closed subset of ℓF

∞.

Proof of the theorem is easy, so omitted.

Theorem 5. Let θ = (kr) be a lacunary sequence. Then the spaces cIF

Proof. We shall give the proof of the theorem for (cIF

all k ∈ N.

Then for a given ε > 0, we have

k∈Jr

Again the set D =

monotone.

?

θand (cIF

0)θare normal and monotone, in general.

0)θand (Yk) be such that¯d(Yk,¯0) ≤¯d(Xk,¯0), for

0)θonly. Let (Xk) ∈ (cIF

B =

r ∈ N : h−1

r

−

¯d(Xk,¯0) ≥ ε

∑

∈ I.

r ∈ N : h−1

r

k∈Jr¯d(Yk,¯0) ≥ ε⊆ B.

Hence D ∈ I and so (Yk) ∈ (cIF

0)θ. Thus the space (cIF

0)θ is normal. Also from Lemma 1, it follows that (cIF

0)θ is

Theorem 6. Let θ = (kr) be a lacunary sequence. Then the spaces (cIF

Proof. We will give the proof for cIF

real numbers defined by

1 + t − 2k,

0,

otherwise.

0)θand cIF

θare symmetric, in general.

θonly. Suppose I is not maximal and I ̸= If. Let us consider a sequence X = (Xk) of fuzzy

Xk(t) =

if t ∈ [2k − 1,2k];

if t ∈ [2k,2k + 1];

1 − t + 2k,

for k ∈ A ⊂ I an infinite set.

Page 6

Author's personal copy

470

B. Hazarika / Applied Mathematics Letters 25 (2012) 466–470

Then (Xk) ∈ cIF

Consider a sequence Y = (Yk), a rearrangement of the sequence (Xk) defined as follows:

Xk,

Then (Yk) ̸∈ cIF

θ. Let K ⊆ N be such that K ̸∈ I and N − K ̸∈ I (the set K exists by Lemma 2, as I is not maximal).

Yk=

if k ∈ K,

otherwise.

¯1,

θ. Therefore the space cIF

θis not symmetric. This completes the proof of the theorem.

?

4. Conclusions

In this article, we have investigated the notion of lacunary convergence from I-convergence of sequences point of view.

Still there are a lot to be investigated on sequence spaces applying the notion of I-convergence. The workers will apply the

techniques used in this article for further investigations on I-convergence.

Acknowledgments

The author thanks the reviewers for the comments on the paper and several constructive comments that have improved

the presentation of the results.

References

[1] P. Kostyrko, T. Šalàt, W. Wilczyński, I-convergence, Real. Anal. Exchange 26 (2) (2000–2001) 669–686.

[2] T. Šalàt, B.C. Tripathy, M. Ziman, On some properties of I-convergence, Tatra Mt. Math. Publ. 28 (2004) 279–286.

[3] T. Šalàt, B.C. Tripathy, M. Ziman, On I-convergence field, Indian J. Pure Appl. Math. 17 (2005) 45–54.

[4] B.C. Tripathy, B. Hazarika, B. Choudhary, Lacunary I-convergent sequences, Kyungpook Math. J (in press).

[5] L.A. Zadeh, Fuzzy sets, Inf. Control 8 (1965) 338–353.

[6] S. Nanda, On sequences of fuzzy numbers, Fuzzy Sets Syst. 33 (1989) 123–126.

[7] F. Nuray, E. Savas, Statistical convergence of sequences of fuzzy real numbers, Math. Slovaca 45 (3) (1995) 269–273.

[8] A.R. Freedman, J.J. Sember, M. Raphael, Some Cesaro-type summability spaces, Proc. Lond. Math. Soc. 37 (1978) 508–520.