An ideal I is a hereditary and additive family of subsets of positive integers ℕ. In this paper, we will introduce the concept of generalized random n-normed space as an extension of random n-normed space. Also, we study the concept of lacunary mean (L)-ideal convergence and L-ideal Cauchy for sequences of complex numbers in the generalized random n-norm. We introduce IL-limit points and IL-cluster points. Furthermore, Cauchy and IL-Cauchy sequences in this construction are given. Finally, we find relations among these concepts.
1. Introduction
The sets of natural numbers and complex numbers will be denoted by ℕ and ℂ, respectively. Fast [1] and Steinhaus [2] independently introduced the notion of statistical convergence for sequences of real numbers, which is a generalization of the concept of convergence. The concept of statistical convergence is a very valuable functional tool for studying the convergence problems of numerical sequences through the concept of density. Afterward, several generalizations and applications of this concept have been presented by different authors (see [3–6]). Kostyrko et al. [7] presented a generalization of the concept of statistical convergence with the help of an ideal I of subsets of the set of natural numbers ℕ, and more is studied in [8–11]. This concept of ideal convergence plays a fundamental role not only in pure mathematics but also in other branches of science concerning mathematics, mainly in information theory, computer science, dynamical systems, geographic information systems, and population modelling. Menger [12] generalized the metric axioms by associating a distribution function with each pair of points of a set. This system is called a probabilistic metric space. By using the concept of Menger, Šerstnev [13] introduced the concept of probabilistic normed spaces. It provides an important area into which many essential results of linear normed spaces can be generalized; see [14]. Later, Alsina et al. [15] presented a new definition of probabilistic normed space which includes the definition of normed space which includes the definition of Šerstnev as a special case. The concept of ideal convergence for single and double sequences of real numbers in probabilistic normed space was introduced and studied by Mursaleen and Mohiuddine [16]. Mursaleen and Alotaibi [17] studied the notion of ideal convergence for single and double sequences in random 2-normed spaces, respectively. For more details and linked concept, we refer to [18–26]. In [27, 28], Gähler introduced a gorgeous theory of 2-normed and n-normed spaces in the 1960s; we have studied these subjects and constructed some sequence spaces defined by ideal convergence in n-normed spaces [29, 30]. Another important alternative of statistical convergence is the notion of lacunary statistical convergence introduced by Fridy and Orhan [31]. Recently, Mohiuddine and Aiyub [4] studied lacunary statistical convergence by introducing the concept Θ-statistical convergence in random 2-normed space. Their work can be considered as a particular generalization of the statistical convergence. In [32], Mursaleen and Mohiuddine generalized the idea of lacunary statistical convergence with respect to the intuitionistic fuzzy normed space, and Debnath [33] investigated lacunary ideal convergence in intuitionistic fuzzy normed linear spaces. Also, lacunary statistically convergent double sequences in probabilistic normed space were studied by Mohiuddine and Savaş in [34]. Jebril and Dutta [35] introduced the concept of random n-normed space. In this paper, we firstly give some basic definitions and properties of random n-normed space in Section 2. In Section 3, we define a new and interesting notion of generalized random n-normed spaces; convergent sequences in it are introduced and we provide some results on it. In Section 4, we study lacunary mean (L)-ideal convergence and L-ideal Cauchy for sequences of complex numbers in the generalized random n-norm. Finally, in Section 5, we introduce IL-limit points and IL-cluster points. Moreover, Cauchy and IL-Cauchy sequences in this framework are given, and we find relations among these concepts.
2. Definitions and Preliminaries
For the reader’s expediency, we restate some definitions and results that will be used in this paper.
The notion of statistical convergence depends on the density (asymptotic or natural) of subsets of ℕ.
Definition 1.
A subset E of ℕ is said to have natural density δ(E) if
(1)δ(E)=limn→∞1n|{k≤n:k∈E}|exists,
where |E| denotes the cardinality of the set E.
Definition 2.
A sequence (xk) is statistically convergent to ℓ if, for every ε>0,
(2)δ({k∈ℕ:|xk-ℓ|≥ε})=0.
In this case, ℓ is called the statistical limit of the sequence (xk).
Definition 3.
A nonempty family of sets I⊆2ℕ is said to be an ideal on ℕ if and only if
ϕ∈I,
for each A,B∈I one has A∪B∈I,
for each B∈I and A⊂B, implies A∈I.
Definition 4.
An ideal I is an admissible ideal if {x}∈I for each x∈ℕ.
Definition 5.
An ideal I⊆2ℕ is said to be nontrivial if I≠ϕ and ℕ∉I.
Definition 6.
A nonempty family of sets F⊆2ℕ is said to be a filter on ℕ if and only if
ϕ∉F,
for each A,B∈F one has A∩B∈F,
for each A∈F and B⊃A, implies B∈F.
For each ideal I, there is a filter F(I) corresponding to I; that is,F(I)={K⊆ℕ:ℕ-K∈I}.
Example 7.
If we take I=If={A⊆ℕ:Aisafinitesubset}, then If is a nontrivial admissible ideal of ℕ and the corresponding convergence coincides with the usual convergence.
Example 8.
If we get I=Iδ={A⊆ℕ:δ(A)=0}, where δ(A) denote the asymptotic density of the set A, then Iδ is a nontrivial admissible ideal of ℕ and the corresponding convergence coincides with the statistical convergence.
Definition 9.
A sequence x=(xk) is said to be I-convergent to a real number ℓ if
(3){k∈ℕ:|xk-ℓ|≥ε}∈Iforeveryε>0.
In this case, we write I-limxk=ℓ.
Definition 10.
By a lacunary sequence Θ=(ij), j=0,1,2,…, where i0=0, one will mean an increasing sequence of nonnegative integers with ij-ij-1→∞ as j→∞, hj=ij-ij-1. The intervals determined by Θ will be denoted by Λj=(ij-1,ij].
Definition 11.
A sequence x=(xk) is said to be lacunary (L)-statistically convergent to the number ℓ if, for every ε>0, one has
(4)limj→∞1hj|{k∈Λj:|xk-ℓ|≥ε}|=0.
The notion of lacunary ideal convergence of real sequences is introduced by Tripathy et al. [36], and Hazarika [37, 38] introduced the lacunary ideal convergent sequences of fuzzy real numbers and studied some properties.
Definition 12.
Let I⊂2ℕ be a nontrivial ideal. A sequence x=(xk) is said to be IL-summable to a number ℓ if, for every ε>0, the set
(5){j∈ℕ:1hj∑k∈Λj|xk-ℓ|≥ε}∈I.
Definition 13.
Let n∈ℕ and let X be a linear space over the field K of dimension d, where d≥n≥2 and K is the field of real or complex numbers. A real valued function ∥·,…,·∥ on Xn satisfies the following four conditions:
∥x1,x2,…,xn∥=0 if and only if x1,x2,…,xn are linearly dependent in X;
∥x1,x2,…,xn∥ is invariant under permutation;
∥αx1,x2,…,xn∥=|α|∥x1,x2,…,xn∥ for any α∈K;
∥x+x′,x2,…,xn∥≤∥x,x2,…,xn∥+∥x′,x2,…,xn∥ is called an n-norm on X, and the pair (X;∥·,…,·∥) is called an n-normed space over the field K.
Definition 14.
A probability distribution function is a function F that is nondecreasing, left continuous on (0,∞) such that F(0)=0 and F(∞)=1. The family of all probability distribution functions will be denoted by Δ+. The space Δ+ is partially ordered by the usual pointwise ordering of functions and has both a maximal element ε0 and a minimal element ε∞; these are given, respectively, by
(6)ε0(t)={0,t≤0,1,t>0,ε∞(t)={0,t<∞,1,t=∞.
There is a natural topology on Δ+ that is induced by the modified Lévy metric dL [39, 40]; that is,
(7)dL(F,G)=inf{h:both[F,G;h]and[G,F;h]hold}
for all F,G∈Δ+andh∈(0,1], where [F,G,h] denote the condition
(8)G(t)≤F(t+h)+h,fort∈(0,1h).
Convergence with respect to this metric is equivalent to weak convergence of distribution functions; that is, (Fn) in Δ+ converges weakly to F (written as Fn→ωF) if and only if Fn(t) converges to F(t) at every point of continuity of the limit function F. Therefore, one has
(9)Fn→ωFiffdL(Fn,F)⟶0,F(x)>1-xiffdL(F,ε0)<xforeveryx>0.
Moreover, the metric space (Δ+,dL) is compact.
Definition 15.
A binary operation ⋆:[0,1]×[0,1]→[0,1] is said to be a continuous t-norm if the following conditions are satisfied:
⋆ is associative and commutative,
⋆ is continuous,
a⋆1=a for all a∈[0,1],
a⋆b≤c⋆d whenever a≤c and b≤d for each a,b,c,d∈[0,1].
Definition 16.
A binary operation ⋄:[0,1]×[0,1]→[0,1] is said to be a continuous t-conorm if the following conditions are satisfied:
⋄ is associative and commutative,
⋄ is continuous,
a⋄0=a for all a∈[0,1],
a⋄b≤c⋄d whenever a≤c and b≤d for each a,b,c,d∈[0,1].
Definition 17.
Let X be a linear space of dimension greater than one, ⋆ a continuous t-norm, and ρ a mapping from X2 into D+. If the following conditions are satisfied:
ρx,y=ε0 if x and y are linearly dependent,
ρx,y=ρy,x for every x and y in X,
ραx,y(t)=ρx,y(t/|α|) for every t>0; α≠0 and x;y∈X,
ρx+y,z(t)≥ρx,z(t)⋆ρy,z(t),
then ρ is called a random 2-norm on X and (X;ρ;⋆) is called a random 2-normed space.
Definition 18.
Let X be a linear space of dimension greater than one over a real field, ⋆ a continuous t-norm, and ρ a mapping from Xn into D+. If the following conditions are satisfied:
ρx1,x2,…,xn=ε0⇔x1,x2,…,xn are linearly dependent,
ρx1,x2,…,xn is invariant under any permutation of x1,x2,…,xn,
ραx1,x2,…,xn(t)=ρx1,x2,…,xn(t/|α|) for every t>0; α≠0,
then ρ is called a random n-norm on X and (X;ρ;⋆) is called a random n-normed space.
3. Generalized Random n-Normed Space
Throughout the paper let I be an admissible ideal of ℕ. By generalizing Definition 18, we obtain a new notion of generalized random n-normed space as follows.
Definition 19.
The five-tuple (X,ρ,ϱ,⋆,⋄) is said to be generalized random n-normed linear space or in short GRnNLS if X is a linear space over the field of complex numbers ℂ, ⋆ is a continuous t-norm, ⋄ is a continuous t-conorm, and ρ, ϱ are two mappings on Xn×(0,∞) into D+×(0,∞) satisfying the following conditions for every x=(x1,x2,…,xn)∈Xn and for each s,t∈(0,∞):
ρx1,x2,…,xn+ϱx1,x2,…,xn≤ε0,
ρx1,x2,…,xn≥ε∞,
ρx1,x2,…,xn=ε0 if and only if x1,x2,…,xn are linearly dependent,
ραx1,x2,…,xn(t)=ρx1,x2,…,xn(t/|α|) for each α∉ℂ∖0,
ρx1,x2,…,xn(t) is invariant under any permutation of (x1,x2,…,xn).
In this case, (ρ,ϱ) is called generalized random n-norm on X and we denote it by (ρ,ϱ)n.
Example 20.
Let (X,∥·,…,·∥) be an n-normed linear space. Put a⋆b=min{a,b} and a⋄b=max{a,b} for all a,b∈[0,1], ρx1,x2,…,xn(t)=t/(t+∥x1,x2,…,xn∥), and ϱx1,x2,…,xn(t)=∥x1,x2,…,xn∥/(t+∥x1,x2,…,xn∥). Then, (X,ρ,ϱ,⋆,⋄) is GRnNLS.
Proof.
For all t,s∈(0,∞), we have the following.
Evidently, ρx1,x2,…,xn(t)+ϱx1,x2,…,xn(t)≤1.
Visibly, ρx1,x2,…,xn(t)≥0.
And
(10)ρx1,x2,…,xn(t)=1⟺tt+∥x1,x2,…,xn∥=1⟺t=t+∥x1,x2,…,xn∥⟺∥x1,x2,…,xn∥=0⟺x1,x2,…,xnarelinearlydependent.
While ∥x1,x2,…,xn∥ is invariant under any permutation of (x1,x2,…,xn), then ρx1,x2,…,xn(t) is invariant under any permutation of (x1,x2,…,xn).
Suppose that, without loss of generality,
(12)ρx1,x2,…,xn′(t)≤ρx1,x2,…,xn(s).⟹tt+∥x1,x2,…,xn′∥≤ss+∥x1,x2,…,xn∥⟹t(s+∥x1,x2,…,xn∥)≤s(t+∥x1,x2,…,xn′∥)⟹t(∥x1,x2,…,xn∥)≤s(∥x1,x2,…,xn′∥)⟹∥x1,x2,…,xn∥=st∥x1,x2,…,xn′∥.
As a result,
(13)∥x1,x2,…,xn∥+∥x1,x2,…,xn′∥≤st∥x1,x2,…,xn′∥+∥x1,x2,…,xn′∥=s+tt∥x1,x2,…,xn′∥.
However,
(14)∥x1,x2,…,xn+xn′∥≤∥x1,x2,…,xn∥+∥x1,x2,…,xn′∥≤s+tt∥x1,x2,…,xn′∥,⟹∥x1,x2,…,xn+xn′∥s+t≤∥x1,x2,…,xn′∥t⟹1+∥x1,x2,…,xn+xn′∥s+t≤1+∥x1,x2,…,xn′∥t⟹s+t+∥x1,x2,…,xn+xn′∥s+t≤t+∥x1,x2,…,xn′∥t⟹s+ts+t+∥x1,x2,…,xn+xn′∥≥tt+∥x1,x2,…,xn′∥⟹ρx1,x2,…,xn+xn′(s+t)≥min{ρx1,x2,…,xn(s),ρx1,x2,…,xn′(t)}.
Evidently, ρx1,x2,…,xn(·):(0,∞)→[0,1] is continuous.
ϱx1,x2,…,xn(t)≥0.
And
(15)ϱx1,x2,…,xn(t)=0⟺∥x1,x2,…,xn∥t+∥x1,x2,…,xn∥=0⟺∥x1,x2,…,xn∥=0⟺x1,x2,…,xnarelinearlydependent.
As ∥x1,x2,…,xn∥ is invariant under any permutation of (x1,x2,…,xn), then ϱx1,x2,…,xn(t) is invariant under any permutation of (x1,x2,…,xn).
Presume, without loss of generality, that
(17)ϱx1,x2,…,xn(s)≤ρx1,x2,…,xn′(t).⟹∥x1,x2,…,xn∥s+∥x1,x2,…,xn∥≤∥x1,x2,…,xn′∥t+∥x1,x2,…,xn′∥⟹∥x1,x2,…,xn∥(t+∥x1,x2,…,xn′∥)≤∥x1,x2,…,xn′∥(s+∥x1,x2,…,xn∥)⟹t∥x1,x2,…,xn∥≤s∥x1,x2,…,xn′∥.
Currently,
(18)∥x1,x2,…,xn+xn′∥s+t+∥x1,x2,…,xn+xn′∥-∥x1,x2,…,xn′∥t+∥x1,x2,…,xn′∥≤∥x1,x2,…,xn∥+∥x1,x2,…,xn′∥s+t+∥x1,x2,…,xn+xn′∥-∥x1,x2,…,xn′∥t+∥x1,x2,…,xn′∥=t∥x1,x2,…,xn∥-s∥x1,x2,…,xn′∥(s+t+∥x1,x2,…,xn+xn′∥)(t+∥x1,x2,…,xn′∥).
By (17),
(19)∥x1,x2,…,xn+xn′∥s+t+∥x1,x2,…,xn+xn′∥≤∥x1,x2,…,xn′∥t+∥x1,x2,…,xn′∥.
In the same way,
(20)∥x1,x2,…,xn+xn′∥s+t+∥x1,x2,…,xn+xn′∥≤∥x1,x2,…,xn′∥t+∥x1,x2,…,xn′∥.⟹ϱx1,x2,…,xn+xn′(s+t)≤max{ϱx1,x2,…,xn(s),ρx1,x2,…,xn′(t)}.
Clearly, ρx1,x2,…,xn(·):(0,∞)→[0,1] is continuous.
Remark 21.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS. Since ⋆ is a continuous t-norm and ⋄ is a continuous t-conorm, the system (r,t)-neighborhoods of θ (the null vector in X) with respect to t is
(21){B(θ,r,t):t>0,0<r<1},
where
(22)B(θ,r,t)={y∈X:ρy,x1,x2,…,xn-1(t)>1-r,ϱy,x1,x2,…,xn-1(t)<r,fort>0}
defined a first countable Hausdorff topology on X, called the (ρ,ϱ)n-topology. Hence, the (ρ,ϱ)n-topology can be completely specified by means of (ρ,ϱ)n-convergence of sequences.
Definition 22.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS, and let r∈(0,1) and x∈X. The set
(23)B(x,r,t)={y∈X:ρy-x,x1,x2,…,xn-1(t)>1-r,hhhϱy-x,x1,x2,…,xn-1(t)<r,fort>0}
is called open ball with center x and radius r with respect to t.
Definition 23.
Let (X,ρ,ϱ,*,⋄) be GRnNLS. A sequence x=(xk) in X is (ρ,ϱ)n-convergent to ℓ∈X with respect to the generalized random n-norm (ρ,ϱ)n if, for r∈(0,1) and every t>0, there exists k0 such that
(24)ρxk-ℓ,x1,x2,…,xn-1(t)≥1-r,ϱxk-ℓ,x1,x2,…,xn-1(t)≤rhhhhhhhhhh∀k≥k0.
In this case, one writes (ρ,ϱ)n-limx=ℓ.
Theorem 24.
Let (X,∥·,…,·∥) be an n-normed linear space. Put a⋆b=min{a,b} and a⋄b=max{a,b} for all a,b∈[0,1], ρx1,x2,…,xn(t)=t/(t+∥x1,x2,…,xn∥), and ϱx1,x2,…,xn(t)=∥x1,x2,…,xn∥/(t+∥x1,x2,…,xn∥). Then, for every sequence x=(xk) and nonzero x1,x2,…,xn-1∈X, one has
(25)limk→∞∥x1,x2,…,xn-1,xk-ℓ∥=0⟹(ρ,ϱ)n-limxk=ℓ.
Proof.
Assume that limk→∞∥x1,x2,…,xn-1,xk-ℓ∥=0. Then, for every ε>0 and for every x1,x2,…,xn-1∈X, there exists a positive integer k0 such that
(26)∥x1,x2,…,xn-1,xk-ℓ∥<εforeachk≥k0,
and, therefore, for any given t>0,
(27)t+∥x1,x2,…,xn-1,xk-ℓ∥t<t+εt
which is the same as
(28)tt+∥x1,x2,…,xn-1,xk-ℓ∥>tt+ε=1-εt+ε.
By letting r=ε/(t+ε)∈(0,1), we have
(29)ρxk-ℓ,x1,x2,…,xn-1(t)≥1-r∀k≥k0.
And since ρxk-ℓ,x1,x2,…,xn-1(t)=1-ϱxk-ℓ,x1,x2,…,xn-1(t), then we have
(30)ϱxk-ℓ,x1,x2,…,xn-1(t)≤r∀k≥k0.
This means (ρ,ϱ)n-limxk=ℓ.
4. IL(ρ,ϱ)n-Cauchy and Convergence in GRnNLSRemark 25.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS. Since ⋆ is a continuous t-norm and ⋄ is a continuous t-conorm, the system (r,t)nL-neighborhoods of θ with respect to t are
(31){BL(θ,r,t):t>0,0<r<1},
where
(32)BL(θ,r,t)={y∈X:1hj∑k∈Λjρyk,x1,x2,…,xn-1(t)>1-r,1hj∑k∈Λjϱyk,x1,x2,…,xn-1(t)<r,fort>0}
determines a first countable Hausdorff topology on X, called the (ρ,ϱ)nL-topology. Thus, the (ρ,ϱ)nL-topology can be completely specified by means of (ρ,ϱ)nL-convergence of sequences.
Definition 26.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS, and let r∈(0,1) and x∈X. The set
(33)BL(x,r,t)={y∈X:1hj∑k∈Λjρyk-x,x1,x2,…,xn-1(t)>1-r,h1hj∑k∈Λjϱyk-x,x1,x2,…,xn-1(t)<r,fort>0}
is called open ball with center x and radius r with respect to t.
Definition 27.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS. A sequence x=(xk) in X is L-convergent to ℓ∈X with respect to the generalized random n-norm (ρ,ϱ)n if, for ε∈(0,1) and every t>0, there exists j0 such that
(34)1hj∑k∈Λjρxk-ℓ,x1,x2,…,xn-1(t)≥1-ε,1hj∑k∈Λjϱxk-ℓ,x1,x2,…,xn-1(t)≤εhhhhhhhhhhhhhhh∀j≥j0.
In this case, one writes (ρ,ϱ)nL-limx=ℓ.
Definition 28.
Let I⊂2ℕ and let (X,ρ,ϱ,⋆,⋄) be GRnNLS. A sequence x=(xk) of elements in X is said to be IL-convergent to ℓ∈X with respect to the generalized random n-norm (ρ,ϱ)n if, for every ε∈(0,1) and t>0, the set
(35){j∈ℕ:1hj∑k∈Λjρxk-ℓ,x1,x2,…,xn-1(t)≤1-εor1hj∑k∈Λjϱxk-ℓ,x1,x2,…,xn-1(t)≥ε}∈I.
Then, one writes IL(ρ,ϱ)n-limx=ℓ.
Example 29.
Let (ℂ,∥·,…,·∥) be an n-normed linear space; take a⋆b=ab and a⋄b=min{a+b,1} for all a,b∈[0,1]. For all x∈ℂ and every t>0, consider
(36)ρx1,x2,…,xn(t)=tt+∥x1,x2,…,xn∥,ϱx1,x2,…,xn(t)=∥x1,x2,…,xn∥t+∥x1,x2,…,xn∥.
Then, (ℂ,ρ,ϱ,⋆,⋄) is GRnNLS. If we take I=Iδ, define a sequence x=(xk) as follows:
(37)xk={1,ifk=i8,i∈ℕ,0,otherwise.
Hence, for every ε∈(0,1) and t>0, we have
(38)δ({j∈ℕ:1hj∑k∈Λjρxk,x1,x2,…,xn-1(t)≤1-εor1hj∑k∈Λjϱxk,x1,x2,…,xn-1(t)≥ε})=0.
So IL(ρ,ϱ)n-limx=0.
Definition 30.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS. A sequence x=(xk) in X is said to be a Cauchy sequence with respect to the generalized random n-norm (ρ,ϱ)nL if, for every t>0 and ε∈(0,1), there exists j0∈ℕ satisfying
(39)1hj∑k∈Λjρxk-xm,x1,x2,…,xn-1(t)>1-ε,1hj∑k∈Λjϱxk-xm,x1,x2,…,xn-1(t)<εhhhhhhhhhhhhhh∀j,m≥k0.
Definition 31.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS. A sequence x=(xk) in X is said to be an IL-Cauchy sequence with respect to the generalized random n-norm (ρ,ϱ)nL if, for every t>0 and ε∈(0,1), there exists j0∈ℕ satisfying
(40){j∈ℕ:1hj∑k∈Λjρxk-xm,x1,x2,…,xn-1(t)>1-ε,1hj∑k∈Λjϱxk-xm,x1,x2,…,xn-1(t)<ε}∈F(I).
Theorem 32.
Let I⊂2ℕ, let (X,ρ,ϱ,⋆,⋄) be GRnNLS, and let x=(xk) be a sequence in X; then, for every ε>0 and t>0, one has the following:
IL(ρ,ϱ)n-limx=ℓ,
{j∈ℕ:(1/hj)∑k∈Λjρxk-ℓ,x1,x2,…,xn-1(t)≤1-ε}∈I and {j∈ℕ:(1/hj)∑k∈Λjϱxk-ℓ,x1,x2,…,xn-1(t)≥ε}∈I,
{j∈ℕ:(1/hj)∑k∈Λjρxk-ℓ,x1,x2,…,xn-1(t)≥1-ε and (1/hj)∑k∈Λjϱxk-ℓ,x1,x2,…,xn-1(t)≤ε}∈F(I),
{j∈ℕ:(1/hj)∑k∈Λjρxk-ℓ,x1,x2,…,xn-1(t)≥1-ε}∈F(I) and {j∈ℕ:(1/hj)∑k∈Λjϱxk-ℓ,x1,x2,…,xn-1(t)≤ε}∈F(I),
IL-limkρxk-ℓ,x1,x2,…,xn-1(t)=1 and IL-limkϱxk-ℓ,x1,x2,…,xn-1(t)=0.
The proof is easy, so it is omitted.
Theorem 33.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS and let x=(xk) be a sequence in X. If (ρ,ϱ)nL-limx exists, then it is unique.
Proof.
Suppose that (ρ,ϱ)nL-limx=ℓ1 and (ρ,ϱ)nL-limx=ℓ2 with ℓ1≠ℓ2. Give ε∈(0,1) and choose λ∈(0,1) such that (1-λ)⋆(1-λ)>1-ε and λ⋄λ<ε. Then, for each t>0, there exists j1∈ℕ such that
(41)1hj∑k∈Λjρxk-ℓ1,x1,x2,…,xn-1(t)>1-ε,1hj∑k∈Λjϱxk-ℓ1,x1,x2,…,xn-1(t)<εhhhhhhhhhhhhhhhh∀j≥j1.
Also, there exists j2∈ℕ such that
(42)1hj∑k∈Λjρxk-ℓ2,x1,x2,…,xn-1(t)>1-ε,1hj∑k∈Λjϱxk-ℓ2,x1,x2,…,xn-1(t)<εhhhhhhhhhhhhhhhh∀j≥j2.
Now, consider j0=max{j1,j2}. Then, for j≥j0, we find a s∈ℕ such that
(43)ρxs-ℓ1,x1,x2,…,xn-1(t2)>1hj∑k∈Λjρxk-ℓ1,x1,x2,…,xn-1(t2)≥1-λ,ρxs-ℓ2,x1,x2,…,xn-1(t2)>1hj∑k∈Λjρxk-ℓ2,x1,x2,…,xn-1(t2)≥1-λ.
Then, we get
(44)ρℓ1-ℓ2,x1,x2,…,xn-1(t)≥ρxs-ℓ1,x1,x2,…,xn-1(t2)⋆ρxs-ℓ2,x1,x2,…,xn-1(t2)>(1-λ)⋆(1-λ)>1-ε.
Since ε>0 is arbitrary, we have ρℓ1-ℓ2,x1,x2,…,xn-1(t)=1 for all t>0. By using a similar technique, it can be proved that ϱℓ1-ℓ2,x1,x2,…,xn-1(t)=0 for all t>0; hence, ℓ1=ℓ2.
Theorem 34.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS and let x=(xk) be a sequence in X. Then, one has
(45)(ρ,ϱ)nL-limx=ℓ⟹IL(ρ,ϱ)n-limx=ℓ.
Proof.
Let (ρ,ϱ)nL-limx=ℓ, and, then, for all t>0 and given ε∈(0,1), there exists j0∈ℕ such that
(46)1hj∑k∈Λjρxk-ℓ,x1,x2,…,xn-1(t)>1-ε,1hj∑k∈Λjϱxk-ℓ,x1,x2,…,xn-1(t)<εhhhhhhhhhhhhhhhh∀j≥j0.
Since I is an admissible ideal and
(47)G={j∈ℕ:1hj∑k∈Λjρxk-ℓ,x1,x2,…,xn-1(t)≤1-εor1hj∑k∈Λjϱxk-ℓ,x1,x2,…,xn-1(t)≥ε}⊆{1,2,3,…,j0-1},
we get G∈I. So IL(ρ,ϱ)n-limx=ℓ.
Theorem 35.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS and let x=(xk) be a sequence in X. If IL(ρ,ϱ)n-limx exists, then it is unique.
The proof follows by using Theorems 33 and 34.
Theorem 36.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS and let x=(xk) be a sequence in X. If (ρ,ϱ)nL-limx=ℓ, then there exists a subsequence (xmk) of x=(xk) such that (ρ,ϱ)n-limxmk=ℓ.
Proof.
Let (ρ,ϱ)nL-limx=ℓ. Then, for all t>0 and given ε∈(0,1), there exists j0∈ℕ such that
(48)1hj∑k∈Λjρxk-ℓ,x1,x2,…,xn-1(t)>1-ε,1hj∑k∈Λjϱxk-ℓ,x1,x2,…,xn-1(t)<εhhhhhhhhhhhhhhh∀j≥j0.
Observably, for each j≥j0, we can take an mk∈Λj such that
(49)ρxmk-L,x1,x2,…,xn-1(t)>1hj∑k∈Λjρxk-ℓ1,x1,x2,…,xn-1(t)>1-ε,ϱxmk-ℓ,x1,x2,…,xn-1(t)<1hj∑k∈Λjρxk-ℓ,x1,x2,…,xn-1(t)<ε.
It follows that (ρ,ϱ)n-limxmk=ℓ.
We create the following two results without proofs, since they can be easily recognized.
Theorem 37.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS. If a sequence x=(xk) in X is Cauchy sequence with respect to the generalized random n-norm (ρ,ϱ)nL, then it is IL-Cauchy sequence with respect to the same norm.
Theorem 38.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS. If a sequence x=(xk) in X is Cauchy sequence with respect to the generalized random n-norm (ρ,ϱ)nL, then there is a subsequence of x=(xk) which is ordinary Cauchy sequence with respect to the norm (ρ,ϱ)n.
5. IL-Limit Point, IL-Cluster Point, and IL-Cauchy Sequence in GRnNLSDefinition 39.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS, and if a sequence x=(xk) in X, then one has the following.
An element ℓ∈X is said to be IL-limit point of x=(xk) if there is a set
(50)𝕄={m1<m2<⋯<mk<⋯}⊂ℕwith𝕄′={j∈ℕ:mk∈Λj}∈F(I),(ρ,ϱ)nL-limxmk=ℓ.
An element ℓ∈X is said to be IL-cluster point of x=(xk) if, for every t>0 and ε∈(0,1), one has
(51){j∈ℕ:1hj∑k∈Λjρxk-ℓ,x1,x2,…,xn-1(t)>1-ε,1hj∑k∈Λjϱxk-ℓ,x1,x2,…,xn-1(t)<ε}∈F(I).
By ⋀(ρ,ϱ)nL(x) we denote the set of all IL-limit points and ⋁(ρ,ϱ)nL(x) the set of all IL-cluster points in X, respectively.
Definition 40.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS. A sequence x=(xk) in X is said to be IL*-Cauchy sequence with respect to the generalized random n-norm (ρ,ϱ)nL if
there exists a set 𝕄={m1<m2<⋯<mk,…}⊂ℕ such that 𝕄′={j∈ℕ:mk∈Λj}∈F(I);
the subsequence (xmk) of x=(xk) is a Cauchy sequence with respect to the generalized random n-norm (ρ,ϱ)nL.
Theorem 41.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS. For each sequence x=(xk) in X, one has
(52)⋀(ρ,ϱ)nL(x)⊂⋁(ρ,ϱ)nL(x).
Proof.
Let ℓ∈⋀(ρ,ϱ)nL(x); then there exists a set 𝕄⊂ℕ such that 𝕄′∈F(I), where 𝕄 and 𝕄′ are as in Definition 39, satisfies (ρ,ϱ)nL-limxmk=ℓ. Thus, for every t>0 and ε∈(0,1), there exists j0∈ℕ such that
(53)1hj∑k∈Λjρxmk-ℓ,x1,x2,…,xn-1(t)>1-ε,1hj∑k∈Λjϱxmk-ℓ,x1,x2,…,xn-1(t)<εhhhhhhhhhhhhhhhh∀j≥j0.
Thus, we have
(54)G={j∈ℕ:1hj∑k∈Λjρxmk-ℓ,x1,x2,…,xn-1(t)>1-ε,1hj∑k∈Λjρxmk-ℓ,x1,x2,…,xn-1(t)<ε}⊇𝕄′∖{m1,m2,…,mj0}.
Since I is an admissible ideal, we have
(55)𝕄′∖{m1,m2,…,mj0}∈F(I)andsoG∈F(I).Henceℓ∈⋁(ρ,ϱ)nL(x).
Theorem 42.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS. For each sequence x=(xk) in X, the set ⋀(ρ,ϱ)nL(x) is closed set in X with respect to the usual topology induced by the generalized random n-norm (ρ,ϱ)nL.
Proof.
Let y∈⋀(ρ,ϱ)nL(x)¯. Take t>0 and ε∈(0,1). Then, there exists ℓ0∈⋀(ρ,ϱ)nL(x)∩BL(y,ε,t). Choose δ>0 such that BL(ℓ0,δ,t)⊂BL(y,ε,t). We have
(56)G={j∈ℕ:1hj∑k∈Λjρxk-y,x1,x2,…,xn-1(t)>1-ε,1hj∑k∈Λjϱxk-y,x1,x2,…,xn-1(t)<ε}⊇{j∈ℕ:1hj∑k∈Λjρxk-ℓ0,x1,x2,…,xn-1(t)>1-δ,1hj∑k∈Λjϱxk-ℓ0,x1,x2,…,xn-1(t)<δ}=H.
Thus, H∈F(I) and so G∈F(I). Hence, y∈⋀(ρ,ϱ)nL(x).
Theorem 43.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS and x=(xk) in X. Then, the following statements are equivalent:
ℓ is a IL-limit point of x;
there exist two sequences y and z in X such that
(57)x=y+z,(ρ,ϱ)nL-limy=ℓ,{j∈ℕ:k∈Λj,zk≠θ}∈I,
where θ is the zero element in X.
Proof.
Let (1) hold; then, there exist sets 𝕄 and 𝕄′ as in Definition 39 such that
(58)𝕄′∉I,(ρ,ϱ)nL-limxmk=ℓ.
Define the sequences y and z as follows:
(59)yk={xk,ifk∈Λj;j∈𝕄′,ℓ,otherwise,zk={θ,ifk∈Λj;j∈𝕄′,xk-ℓ,otherwise.
Consider the case k∈Λj such that j∈ℕ-𝕄′. Then, for each ε∈(0,1) and t>0, we get
(60)ρyk-ℓ,x1,x2,…,xn-1(t)=1>1-ε,ϱyk-ℓ,x1,x2,…,xn-1(t)=0<ε.
Thus, in this case,
(61)1hj∑k∈Λjρyk-ℓ,x1,x2,…,xn-1(t)=1>1-ε,1hj∑k∈Λjϱyk-ℓ,x1,x2,…,xn-1(t)=0<ε.
For that, (ρ,ϱ)nL-limy=ℓ. Now, we have
(62){j∈ℕ:k∈Λj,Zk≠θ}⊂ℕ-𝕄′andso{j∈ℕ:k∈Λj,Zk≠θ}∈I.
Now, suppose that (2) holds. Let 𝕄′={j∈ℕ:k∈Λj,Zk=θ}. Then, obviously 𝕄′∈F(I) and so it is an infinite set. Form the set
(63)𝕄={m1<m2<⋯<mk<⋯}⊂ℕsuchthatmk∈Λj,zmk=θ.
Since xmk=ymk and (ρ,ϱ)nL-limy=ℓ, we find that (ρ,ϱ)nL-limxmk=ℓ. This completes the proof.
Theorem 44.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS and let x=(xk) be a sequence in X. Let I be a nontrivial ideal. If there is a IL(ρ,ϱ)n-convergent sequence y=(yk) in X such that {k∈ℕ:yk≠xk}∈I, then x is also IL(ρ,ϱ)n-convergent.
Proof.
Suppose that {k∈ℕ:yk≠xk}∈I and IL(ρ,ϱ)n-limy=ℓ. Then, for every ε∈(0,1) and t>0, the set
(64){j∈ℕ:1hj∑k∈Λjρyk-ℓ,x1,x2,…,xn-1(t)≤1-εor1hj∑k∈Λjϱyk-ℓ,x1,x2,…,xn-1(t)≥ε}∈I.
For every 0<ε<1 and t>0, we have
(65){j∈ℕ:1hj∑k∈Λjρxk-ℓ,x1,x2,…,xn-1(t)≤1-εor1hj∑k∈Λjϱxk-ℓ,x1,x2,…,xn-1(t)≥ε}⊆{k∈ℕ:yk≠xk}∪{j∈ℕ:1hj∑k∈Λjρyk-ℓ,x1,x2,…,xn-1(t)≤1-εorhhhhh1hj∑k∈Λjϱyk-ℓ,x1,x2,…,xn-1(t)≥ε}.
As both of the sets of the right-hand side are in I, we have that
(66){j∈ℕ:1hj∑k∈Λjρxk-ℓ,x1,x2,…,xn-1(t)≤1-εor1hj∑k∈Λjϱxk-ℓ,x1,x2,…,xn-1(t)≥ε}∈I.
And the proof of the theorem follows.
The proof of the following result can be easily reputable from the definitions.
Theorem 45.
Let (X,ρ,ϱ,⋆,⋄) be GRnNLS. If a sequence x=(xk) in X is IL*-Cauchy sequence with respect to the generalized random n-norm (ρ,ϱ)nL, then it is IL-Cauchy sequence also.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgments
Many thanks are due to the editor and the anonymous referee for careful reading of the paper and valuable suggestions which helped in improving an earlier version of it.
FastH.Sur la convergence statistique19512241244MR0048548ZBL0044.33605SteinhausH.Sur la convergence ordinaire et la convergence asymptotique195127374FridyJ. A.On statistical convergence198554301313MR816582ZBL0588.40001MohiuddineS. A.AiyubM.Lacunary statistical convergence in random 2-normed spaces201263581585MR2950573ŠalátT.On statistically convergent sequences of real numbers1980302139150MR587239ZBL0437.40003MaddoxI. J.Statistical convergence in a locally convex space1988104114114510.1017/S0305004100065312MR938459ZBL0674.40008KostyrkoP.ŠalátT.WilczyńskiW.I-convergence-convergence2000/01262669685MR1844385DemsK.On I-Cauchy sequences2004-2005301123128MR2126799MursaleenM.MohiuddineS. A.EdelyO. H. H.On the ideal convergence of double sequences in intuitionistic fuzzy normed spaces201059260361110.1016/j.camwa.2009.11.002MR2575550ZBL1189.40003CakalliH.HazarikaB.Ideal quasi-Cauchy sequences20122012, article 23410.1186/1029-242X-2012-234MR3017301ZBL06237673DasP.KostyrkoP.WilczyńskiW.MalikP.I and I* convergence of double sequences2008585605620MengerK.Statistical metrics194228535537MR000757610.1073/pnas.28.12.535ZBL0063.03886ŠerstnevA. N.1962122ŠerstnevA. N.On the concept of a stochastic normalized space1963149280283MR0154319AlsinaC.SchweizerB.SklarA.On the definition of a probabilistic normed space1993461-2919810.1007/BF01834000MR1220724ZBL0792.46062MursaleenM.MohiuddineS. A.On ideal convergence in probabilistic normed spaces2012621496210.2478/s12175-011-0071-9MR2886653ZBL1274.40034MursaleenM.AlotaibiA.On I-convergence in random 2-normed spaces201161693394010.2478/s12175-011-0059-5MR2859625GürdalM.AçıkI.On I-Cauchy sequences in 2-normed spaces200811234935410.7153/mia-11-26MR2410284YamancıU.GürdalM.On lacunary ideal convergence in random 2-normed space20132013810.1155/2013/868457868457MR3100669HazarikaB.On generalized difference ideal convergence in random 2-normed spaces201226612731282MR3099590ZBL06221229KumarV.Lafuerza-GuillénB.On ideal convergence of double sequences in probabilistic normed spaces20122881689170010.1007/s10114-012-9321-1MR2944833ZBL1259.54012KumarK.KumarV.On the I and I*-Cauchy sequences in probabilistic normed spaces2008214758LahiriB. K.DasP.I and I*-convergence in topological spaces20051302153160MohiuddineS. A.ŞevliH.CancanM.Statistical convergence in fuzzy 2-normed space2010124787798MR2649299ZBL1198.40002ŞahinerA.GürdalM.SaltanS.GunawanH.Ideal convergence in 2-normed spaces200711514771484MR2368664ZBL1134.46302SchweizerB.SklarA.Statistical metric spaces196010313334MR011515310.2140/pjm.1960.10.313ZBL0096.33203GählerS.Lineare 2-normierte Räume196428143MR0169021GählerS.Untersuchungen uber verallgemeinerte m-metrische Raume, I, II, III196940165189BakeryA. A.Elnour MohamedE. A.Alamin AhmedM.Some generalized difference sequence spaces defined by ideal convergence and Musielak-Orlicz function20132013910.1155/2013/972363972363MR3066298BakeryA. A.Generalized difference γ-sequence spaces defined by ideal convergence and the Musielak-Orlicz function201320131410.1155/2013/123798123798FridyJ. A.OrhanC.Lacunary statistical convergence199316014351MR122750210.2140/pjm.1993.160.43ZBL0794.60012MursaleenM.MohiuddineS. A.On lacunary statistical convergence with respect to the intuitionistic fuzzy normed space2009233214214910.1016/j.cam.2009.07.005MR2568512ZBL1183.46070DebnathP.Lacunary ideal convergence in intuitionistic fuzzy normed linear spaces201263370871510.1016/j.camwa.2011.11.034MR2871670ZBL1238.40003MohiuddineS. A.SavaşE.Lacunary statistically convergent double sequences in probabilistic normed spaces201258233133910.1007/s11565-012-0157-5MR2996463JebrilI. H.DuttaH.Generalization of n-Normed space201011819TripathyB. C.HazarikaB.ChoudharyB.Lacunary I-convergent sequences201252447348210.5666/KMJ.2012.52.4.473MR3006648HazarikaB.Lacunary I-convergent sequence of fuzzy real numbers2009102203206HazarikaB.Fuzzy real valued lacunary I-convergent sequences201225346647010.1016/j.aml.2011.09.037MR2856015SchweizerB.SklarA.1983New York, NY, USANorth-HollandNorth-Holland Series in Probability and Applied MathematicsMR790314SibleyD. A.A metric for weak convergence of distribution functions197113427430MR031408910.1216/RMJ-1971-1-3-427ZBL0245.60007