See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/353850888

Metric dimension of ideal-intersection graph of the ring

Article  in  AKCE International Journal of Graphs and Combinatorics · August 2021

DOI: 10.1080/09728600.2021.1962700

CITATIONS READS

0 4

3 authors, including:

Laxman Saha Kalishankar Tiwary

Balurghat College Raiganj University
23 PUBLICATIONS   89 CITATIONS    57 PUBLICATIONS   50 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

On fixed point theory for point valued and set valued mapping View project

Graph Coloring View project

All content following this page was uploaded by Laxman Saha on 13 August 2021.

The user has requested enhancement of the downloaded file.

 AKCE International Journal of Graphs and Combinatorics

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/uakc20

Metric dimension of ideal-intersection graph of

the ring

Laxman Saha, Mithun Basak & Kalishankar Tiwary

To cite this article: Laxman Saha, Mithun Basak & Kalishankar Tiwary (2021): Metric dimension of

ideal-intersection graph of the ring , AKCE International Journal of Graphs and Combinatorics,
DOI: 10.1080/09728600.2021.1962700

To link to this article: https://doi.org/10.1080/09728600.2021.1962700

© 2021 The Author(s). Published with

license by Taylor & Francis Group, LLC.

Published online: 12 Aug 2021.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=uakc20
AKCE INTERNATIONAL JOURNAL OF GRAPHS AND COMBINATORICS
https://doi.org/10.1080/09728600.2021.1962700

Metric dimension of ideal-intersection graph of the ring Zn

Laxman Sahaa, Mithun Basaka, and Kalishankar Tiwaryb
a
Department of Mathematics, Balurghat College, Balurghat, India; bDepartment of Mathematics, Raiganj University, Raiganj, India

ABSTRACT KEYWORDS
Metric Dimension of a simple connected graph is the minimum number of vertices those are used Ideal-intersection graph;
to identify each vertex of the graph uniquely using distance code. In this paper, we determine resolving set;
metric dimension of ideal-intersection graph for the ring Zn , where n being a positive integer. metric dimension

AMSC 2010
05C12; 05C05;
05C90; 05C76

1. Introduction Throughout this paper, G(R) denotes ideal-intersection

graph of the ring R.
The problem of placing a minimum number of stations
For a given positive integer n, define P(n) be the set of
(sonar/loran detecting devices) in a network so that the pos-
all primes that are used to decompose the integer n. For
ition of every vertex in the network can be uniquely
example, if n ¼ pa11 , pa22 , :::, par r , then PðnÞ ¼ fp1 , p2 , :::, pr g:
described in terms of graph theoretic distances. This moti-
Throughout this paper, we denote < m > be the ideal gen-
vates the study of graph theoretic parameter metric dimen-

in the ring Zn :
erated by m
sion of a connected graph G. The Metric dimension
problem was introduced by Slater [10] (independently Lemma 2.1. For any positive integer n ¼ pa11 , pa22 , :::, par r , every
Harary and Melter [7]) and further studied in [1–5, 8, 9]. b b b
ideal of the ring Zn is of the forms I ¼< p1 1 , p2 2 , :::, pr r >,
Application of resolving set and metric dimension arise in
where bi’s are integers with 0  bi  ai :
many various platforms such as Robot Navigation [8],
Digitization of Image [11], Network Optimization [3], Next lemma gives the number of vertices of the ideal
Mastermind game [5] and chemistry and Drug Design [4]. intersection graph GðZn Þ:
For an ordered subset S ¼ fw1 , w2 , :::, wk g  V(G) and a
Lemma 2.2. ForQ n ¼ pa11 , pa22 , :::, par r , the ideal-intersection
vertex v of G, the code of v with respect to S is a k-vector
graph GðZn Þ has 1ir ðai þ 1Þ  2 vertices.
given by codeS ðvÞ ¼ ðdðv, w1 Þ, dðv, w2 Þ, :::, dðv, wk ÞÞ: CodeS ðvÞ
is also called the metric representation of v with respect to
S. If codeS ðuÞ ¼ codeS ðvÞ implies that u ¼ v for all pairs u, v Proof. Lemma 2.1 gives that every ideal I of Zn is of the
b b b
of vertices of G, then S is called a resolving set for the graph form Q I ¼< p1 1 , p2 2 , :::, pr r > , where 0  bi  ai : There are
G. The metric dimension of graph G is the minimum cardin- total 1ir ðai þ 1Þ ideals of the ring Zn : Now if bi ¼ ai for
ality of a resolving set for G and it is denoted by bðGÞ: all i ¼ 1, 2, :::, r, then I ¼< 0 > and if bi ¼ 0 for all i ¼
Throughout this paper, codeS ðvÞ is simply denoted by 1, 2, :::, r, then I ¼ Zn : As vertices of GðZ Q n Þ are the nonzero
code-v: In this paper, we consider the more convenient proper ideals of Zn , so jVðGðZn ÞÞj ¼ 1ir ðai þ 1Þ  2:
intersection graphs namely, ideal intersection graph for the
b b b
ring Zn of integers modulo n and determine the metric Definition 2.2. For any ideal I ¼< p1 1 , p2 2 , :::, pr r > of the
dimension of GðZn Þ for all values of n. ring Zn with n ¼ pa11 , pa22 , :::, par r , define a set LðIÞ ¼ fpi :
bi ¼ ai g and called it by level set for I. The ideal I is called
i-th level ideal if cardinality of L(I) is i. Also we define
2. Preliminaries and definition L0 ðIÞ ¼ fpi : bi 6¼ ai g: It is clear to observe that for a fixed
ideal I, LðIÞ  PðnÞ and L0 ðIÞ is the complement of L(I) in
Definition 2.1. For a commutative ring R, ideal-intersection P(n), i.e., LðIÞ [ L0 ðIÞ ¼ PðnÞ with L0 ðIÞ \ LðIÞ ¼ /:
graph is a simple graph whose vertices are non-zero proper
ideals of the ring R and two vertices (ideals) are adjacent if Now we give some properties of the level set L(I) for an
their intersection is also a non-zero (proper) ideal of R. ideal I in the following.

CONTACT Laxman Saha [email protected] Department of Mathematics, Balurghat College, Balurghat 733101, India.
ß 2021 The Author(s). Published with license by Taylor & Francis Group, LLC.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.
2 L. SAHA ET AL.

Proposition 2.1. Let n ¼ pa11 , pa22 , :::, par r . Then for the ring set of all ith level ideals by Ai. Some ideal classes may con-
Zn , we have the following tain more than one elements. We call the ideal classes con-
taining exactly one element (vertex/ideal) by one element
(a) I ¼< 0 >¼< pa11 , pa22 , :::, par r >, if and only if LðIÞ ¼ ideal class, otherwise, many element ideal class.
PðnÞ, where PðnÞ ¼ fp1 , p2 , :::, pr g:
b b b
(b) I ¼< p1 1 , p2 2 , :::, pr r > with bi 6¼ ai for all i, if and Lemma 2.4. For the graph GðZn Þ, we have the following
only if LðIÞ ¼ /:
(a) The diameter of GðZn Þ is 2.
Lemma 2.3. For any two ideals I and J of the (b) Every distinct pair of classes preserve the distances, i.e.,
ring Zn , LðI \ JÞ ¼ LðIÞ [ LðJÞ: for two ideals I1 2 ½I and J1 2 ½J, dðI1 , J1 Þ ¼ 1 or 2
according as dðI, JÞ ¼ 1 and dðI, JÞ ¼ 2:
b b
Proof. By Lemma 2.1, we may assume I ¼< p1 1 , p2 2 , :::, pr r >
b (c) Each ideal class ½I forms a clique in GðZn Þ:
c c c maxfb1 , c1 g
and J ¼< p11 , p22 , :::, pr r > : Then I \ J ¼< p1 ,
maxfb2 , c2 g maxfbr , cr g
Proof. (a) Since every zero level element (if exists) is adja-
p2 , :::, pr > : Now from
Definition 2.2, LðI \ cent to all others elements of GðZn Þ and ðr  1Þ-th level ele-
JÞ ¼ fpi : maxfbi , ci g ¼ ai g ¼ fpi : either bi ¼ ai or ci ¼ ments are not adjacent to each other, so the result is true
ai g ¼ LðIÞ[ LðJÞ: w when GðZn Þ contains at least one zero level element. Now
Recall that for a positive integer n, P(n) denotes the set we assume that the graph GðZn Þ does not contains any zero
of all primes that are used to decompose n. From here to level element. From Corollary 2.1, dðI, JÞ  2 for any two
onward, we denote jPðnÞj by r. distinct ideals I and J with 1  jLðIÞj, jLðJÞj  r  2: Let I
and J be two arbitrary elements from ðr  1Þ th level with
Theorem 2.1. Two vertices I and J in GðZn Þ are non-adja- level sets L(I) and L(J) respectively. Since both L(I) and L(J)
cent if and only if jLðI \ JÞj ¼ r: contains ðr  1Þ elements from P(n) and jPðnÞj ¼
r, LðIÞ \ LðJÞ 6¼ /: Let pi 2 LðIÞ \ LðJÞ: Again there exists an
Proof. Since I and J are non zero proper ideals, their inter- one level element K with level set LðKÞ ¼ fpi g: Therefore by
section ideal is also proper ideal but it may be zero. Theorem 2.1, dðI, JÞ ¼ 2: Thus we obtain our result.
Definition 2.1 implies that two vertices I and J are non-adja- (b) From given conditions, LðI1 Þ ¼ LðIÞ and LðJ1 Þ ¼ LðJÞ:
cent if and only if < I \ J >¼< 0 > : Then by Proposition First we assume dðI, JÞ ¼ 1: Then by Corollary 2.1 and
2.1, < I \ J >¼< 0 > if and only if LðI \ JÞ ¼ PðnÞ and Lemma 2.3, I and J are adjacent if only if jLðI1 Þj þ jLðJ1 Þj 
hence jLðI \ JÞj ¼ jPðnÞj: w jLðI1 Þ \ LðJ1 Þj < r, which implies and implied by I1 and J1
are adjacent. Now if I and J are not adjacent, then dðI, JÞ ¼
Corollary 2.1. Two vertices I and J in GðZn Þ are adjacent if 2 as the diameter of GðZn Þ is 2. From Theorem 2.1,
and only if jLðI \ JÞj < r: dðI, JÞ ¼ 2 if and only if jLðI1 Þj þ jLðJ1 Þj  jLðI1 Þ \ LðJ1 Þj ¼
r, which implies and implied by dðI1 , J1 Þ ¼ 2:
Corollary 2.2. Every zero level ideal is adjacent to every ver- (c) If ½I is one element ideal class, then it is obvious.
tex of GðZn Þ: Again if ½I is a many element ideal class, then for two dis-
tinct I1 , I2 2 ½I, we have jLðI1 \ I2 Þj ¼ jLðI1 Þj þ jLðI2 Þj 
Proof. Let I be an zero level ideal and J 2 VðGðZn ÞÞ n fIg be jLðI1 Þ \ LðI2 Þj ¼ jLðI1 Þj ¼ jLðI2 Þj < r, which implies that I1
an arbitrary vertex. Then LðIÞ \ LðJÞ ¼ LðJÞ and hence and I2 are adjacent by Corollary 2.1. Thus ½I form a com-
jLðIÞ \ LðJÞj ¼ jLðJÞj  r  1: So I is adjacent to J in plete subgraph of GðZn Þ: w
GðZn Þ: w

Definition 2.5. [6] Let u be a vertex of a graph G. The open

Corollary 2.3. Two vertices (ideals) I and J with same level neighborhood of u is a set NðuÞ :¼ fv 2 VðGÞ :
set in GðZn Þ are adjacent. v is adjacent to ug, and the closed neighborhood of u is a
Now our aim is to find a partition of the vertex set by set N½u :¼ NðuÞ [ fug: Two distinct vertices u, v are adja-
defining an equivalence relation on VðGðZn ÞÞ: cent twins if N½u ¼ N½v, and non-adjacent twins if NðuÞ ¼
NðvÞ: For a graph G, a set T  VðGÞ is called a twin set of
Definition 2.3. We define a relation q on the set of vertices G if v, w are twins in G for every pair of distinct verti-
of GðZn Þ as follows: for two vertices I and J of GðZn Þ, IqJ if ces v, w 2 T:
and only if LðIÞ ¼ LðJÞ:
Clearly, the relation q defined above is an equivalence Lemma 2.5. [6] If u, v are twins in a connected graph G,
relation and so the vertex set VðGðZn ÞÞ can be partitioned then dðu, xÞ ¼ dðv, xÞ for every vertex x 2 VðGÞ n fu, vg:
into equivalence classes.
Lemma 2.6. [6] If T  VðGÞ is a twin set with cardinality
Definition 2.4. For a vertex I (ideal) of GðZn Þ, let us define jTj, then any resolving set must contains ðjTj  1Þ elements
the class of I by the set fJ 2 VðGðZn ÞÞ : JqIg and we denote of T.
it by ½I: For an ith level ideal I, we call the class ½I as ith
levelideal
 class. It is clear to observe that for each i, there Lemma 2.7. Each many element ideal class in GðZn Þ is a
are r
numbers of ith level ideal classes. We denote the twin set.
i
 AKCE INTERNATIONAL JOURNAL OF GRAPHS AND COMBINATORICS 3

Proof. Let ½I be a many element ideal class and I1 , I2 2 ½I Case-2: jB \ A1 j ¼ ðr  1Þ: Here exactly one element is
be any two elements. NðIÞ ¼ NðI1 Þ ¼ NðI2 Þ as dðJ, IÞ ¼ in B n A1 and this element must be in [r2 i¼1 Ai as B 
dðJ, I1 Þ ¼ dðJ, I2 Þ (by Lemma 2.4), where J is arbitrary. Thus i¼1 Ai : Say this unique element of B n A1 is in Ai, where
[r2
every elements of ½I have same neighborhood. Hence ½I is 2  i  ðr  2Þ: Then any two element of A2 with condition
a twin set. w that their level sets have non empty intersection with the
level set of the remaining element of B have same code as
they are at distance 1 from all elements of B.
3. Metric dimension of GðZp1 , p2 , p3 , ..., pr Þ From above it is clear that B must have intersection with
Ar1 : Since B 6¼ Ar1 then there exists at lest one element of
In this section, we determine the exact value of the metric Ar1 such that it is not in B. Suppose that 1  m < r
dimension of GðZp1 , p2 , p3 , :::, pr Þ: Recall that for n ¼ number of elements of Ar1 are not in B. These m elements
p1 , p2 , p3 , :::, pr , every ideal class is an one element ideal of B are from A1 , A2 , A3 , :::, Ar2 : Since 1  m < r and r 
class. The following result is due to Khuller et al. [8].
4, ð 2r Þ  m  2: Thus there are ð 2r Þ  m number of elements
Lemma 3.1. [8] Let G be a finite graph with diameter d and
of A2 having same codes because using Theorem 2.1 and
metric dimension b. Then G has maximum db þ b vertices. Corollary 2.1 we have
In the following theorem, we give the metric dimension (
2, vi 2 Ar1 or Ar2 ;
for GðZn Þ, when n ¼ p1 , p2 , p3 , :::, pr : dðu, vi Þ ¼
1, otherwise:
Theorem 3.1. For any integer r  4, the metric dimension of
GðZp1 , p2 , p3 , :::, pr Þ is r. for any u 2 A2 n B: This proves that B ¼ Ar1 and so Ar1
is the only basis for GðZp1 , p2 , :::, pr Þ: w

Proof. Let bðGðZn ÞÞ ¼ m and if possible, we assume m < r.

Then applying Lemma 3.1, the graph GðZn Þ can have at Theorem 3.1 gives the metric dimension for GðZp1 , p2 , p3 , :::, pr Þ,
most 2m þ m vertices as the diameter of GðZn Þ is 2. Lemma where pi are distinct primes and r  4: Now we determine
2.2 gives that GðZn Þ has 2r  2 vertices. Since r  4 and the metric dimension for GðZp1 p2 p3 Þ, where p1 , p2 , p3 are dis-
m < r, we have 2r  2 > 2r1 þ r  1  2m þ m: Thus we tinct primes.
arrive a contradiction. Therefore, bðGðZn ÞÞ  r: Now we
show Ar1 ¼ fI : LðIÞ ¼ PðnÞ n fpi g, 1  i  rg is a resolv- Theorem 3.3. For distinct primes p1, p2 and p3, the metric
dimension of ðGðZp1 p2 p3 Þ is 2.
ing set for GðZp1 , p2 , p3 , :::, pr Þ: Let I, J 2 VðGðZp1 , p2 , :::, pr ÞÞ n
Ar1 be two distinct ideals with level sets L(I) and L(J),
Proof. The vertex set of GðZp1 p2 p3 Þ is given by VðGðZp1 p2 p3 ÞÞ ¼
respectively. Without loss of generality, we may assume
f< p1 > , < p2 > , < p3 > , < p1 p2 > , < p2 p3 > , < p1 p3 >g:
jLðJÞj  jLðIÞj: Since I and J are distinct ideals, there exists
Here S ¼ f< p1 p2 > , < p1 p3 >g forms a resolving set for that
at least one pi 2 PðnÞ such that pi 2 LðIÞ but pi 62 LðJÞ even graph, as code– < p1 >¼ ð1, 1Þ, code– < p2 >¼ ð1, 2Þ, code– <
if it is the case jLðJÞj ¼ jLðIÞj: Let K 2 Ar1 with LðKÞ ¼ p3 >¼ ð2, 1Þ and code < p2 p3 >¼ ð2, 2Þ: Thus we say that
PðnÞ n fpi g: Now by Theorem 2.1 and Corollary 2.1, metric dimension is at most 2. Now the graph GðZp1 p2 p3 Þ con-
dðK, IÞ ¼ 1 and dðK, JÞ ¼ 2: Thus K resolves I and J. Hence tains three ideal of level 1 and three ideals of level 2. No 1-level
Ar1 form a resolving set. Therefore bðGðZp1 , p2 , :::, pr ÞÞ ¼ r: w ideal forms a resolving set as the codes of others two ideal of
In the following theorem, we show that the graph level 1 are same. Again the 2-level ideal f< p1 p2 >g is not a
GðZp1 , p2 , p3 , :::, pr Þ has unique metric basis, namely, Ar1 : resolving set as the codes of < p1 > and < p2 > are same. Due
to the similar reasons f< p1 p3 >g and f< p2 p3 >g can not
Theorem 3.2. For r  4, GðZp1 , p2 , p3 , :::, pr Þ has unique met- form a resolving set separately. Thus we say that
ric basis. bðGðZp1 p2 p3 ÞÞ2: Also in above we show that S¼f<p1 p2 >,
<p1 p3 >g is a resolving set. Therefore, bðGðZp1 p2 p3 ÞÞ¼2: w
Proof. Let B be an another metric basis other than Ar1 :
Since metric dimension of GðZp1 , p2 , :::, pr Þ is r, so jBj ¼ r: If Recall that n ¼ pa11 , pa22 , :::, par r where pi are prime and ai
B ¼ A1, then codes of each elements of A2 , A3 , :::, Ar2 are are positive integers. In the next section, we assume that
same with respect to A1 as every element of A1 is adjacent some ai are equal to 1 and others are greater than 1. For the
sake of problem, let us consider first consecutive s numbers
to any element of A2 , A3 , :::, Ar2 by Corollary 2.1. Again if
of ai’s are equal to one and rest are greater than one.
B  [r2 i¼2 Ai , then also codes of each elements of Ar1 are
same with respect to B as dðvi , vÞ ¼ dðvj , vÞ ¼ 2 where
vi , vj 2 Ar1 and v is an arbitrary element in [r2 i¼2 Ai : Now
4. Metric dimension of GðZp1 p2 ...ps pasþ1 . . . par r Þ
sþ1
let B  [r2 A : Then we consider the following two cases where ai  2
i¼1 i
depending on the number of elements in jB \ A1 j: For n ¼ p1 p2 :::ps pasþ1
sþ1
:::par r , the graph GðZn Þ consist of 2s  1
Case-1: jB \ A1 j  ðr  2Þ: In this case jA1 n Bj  2 and numbers of one element ideal classes and the remaining 2r  2s
the codes of all elements in A1 n B are same namely, numbers of many elements ideal classes.
ð1, 1, :::, 1Þ:
4 L. SAHA ET AL.

Example 4.1. For the graph GðZp1 p2 p23 p24 Þ, ½< p23 p24 >, On account of all of the above cases we have bðGðZn Þ 
½< p23 p24 p1 >, ½< p23 p24 p2 > are one element ideal classes 2s ðasþ1 þ 1Þðasþ2 þ 1Þ:::ðar þ 1Þ þ s  2r  1: w

whereas some many element ideal classes are ½< p3 > ¼

Theorem 3.2 gives a lower bound of the metric dimen-
f< p3 >, < p4 >, < p3 p4 >g; ½< p2 > ¼ f< p2 > , < p2 p3 p4 > ,
sion for GðZp1 p2 :::ps pasþ1 :::par r Þ: In the next theorem, we
< p2 p3 > , < p2 p4 >g; ½< p23 > ¼ f< p23 > , < p23 p4 >g; ½< p24 > ¼ sþ1

f< p24 > , < p3 p24 >g: give bðGðZp1 p2 :::ps pasþ1 :::par r ÞÞ:
sþ1

In the following theorem, we give a lower bound of met- Theorem 4.2. For n ¼ p1 p2 :::ps pasþ1
sþ1
:::par r , the metric dimen-
ric dimension for GðZp1 p2 :::ps pasþ1 :::par r Þ: sion of GðZn Þ is given by
sþ1
bðGðZn ÞÞ ¼ 2s ðasþ1 þ 1Þðasþ2 þ 1Þ:::ðar þ 1Þ þ s  2r  1:
Theorem 4.1. For n ¼ p1 p2 :::ps pasþ1
sþ1
:::par r , the metric dimen-
sion of GðZn Þ satisfies
bðGðZn ÞÞ  2s ðasþ1 þ 1Þðasþ2 þ 1Þ:::ðar þ 1Þ þ s  2r  1: Proof. From Theorem 3.2, bðGðZn ÞÞ  2s ðasþ1 þ 1Þðasþ2 þ
1Þ:::ðar þ 1Þ þ s  2r  1: For the reverse inequality, let B 
Proof. Let R be a minimal resolving set for GðZn Þ: In VðGðZn Þ be the collection of ideals exactly one from each
VðGðZn ÞÞ, there are 2r  2s many element ideal classes and equivalence class of q as defined in Definition 2.3. Also let
they form twin sets T1 , :::, T2r 2s due to Lemma 2.7. By C ¼ fI : LðIÞ ¼ PðnÞ n fpi g, i ¼ 1, :::, sg: We show that
Lemma 2.6, jR P \ Ti j  jTi j  1 for all i ¼ 1, 2, :::, 2r  2s : ðVðGðZn ÞÞ n BÞ [ C forms a resolving set for GðZn Þ: To
Therefore jRj  ðjTi j  1Þ ¼ 2s ðasþ1 þ 1Þðasþ2 þ 1Þ:::ðar þ prove this it is sufficient to show that elements of the set
1Þ  2r  1: We show that jRj  2s ðasþ1 þ 1Þðasþ2 þ B n C have distinct codes. Let I, J 2 B n C with level sets L(I)
1Þ:::ðar þ 1Þ þ s  2r  1: If possible, let jRj ¼ 2s ðasþ1 þ and L(J), respectively. Without loss of generality, we may
1Þðasþ2 þ 1Þ:::ðar þ 1Þ þ l  2r  1, where 0  l  s  1: Let assume jLðJÞj  jLðIÞj: Since I and J are distinct, there exists
B  R \ T  R be a set such that jB \ Ti j ¼ at least one pi 2 PðnÞ such that pi 2 LðIÞ but pi 62 LðJÞ: Let
PjTi j  1 for all K be an ideal with LðKÞ ¼ PðnÞ n pi : Then jLðKÞj ¼ r  1
i ¼ 1, 2, :::, 2r  2s : Thus B  R and jBj ¼ ðjTi j  1Þ: Let
S ¼ fI : I 2 T n B, LðIÞ ¼ fpi g, i ¼ 1, :::, sg and C ¼ fI : and K 2 ðVðGðZn ÞÞ n BÞ [ C: Now by Corollary 2.1,
LðIÞ ¼ PðnÞ n fpi g, i ¼ 1, :::, sg: Then C \ B ¼ / as the ele- dðK, IÞ ¼ 1 and by Theorem 2.1, dðK, JÞ ¼ 2: Thus K
ments of C are from the elements of one element ideal resolves I and J and hence ðVðGðZn ÞÞ n BÞ [ C forms a
classes which are at level ðr  1Þ whereas the elements of B resolving set. Clearly, the cardinality of ðVðGðZn ÞÞ n BÞ [ C
are from many element ideal classes. First we show that B is is 2s ðasþ1 þ 1Þðasþ2 þ 1Þ:::ðar þ 1Þ þ s  2r  1 and hence
a proper subset of R. If possible, let R ¼ B. Then two ideal the theorem. w

I, J 2 S with level sets LðIÞ ¼ fpi g, LðJÞ ¼ fpj g are not

resolved by any ideal K 2 B as dðI, KÞ ¼ dðJ, KÞ ¼ 1 where Example 4.2. For the graph GðZp1 p2 p23 Þ, ½< p1 > ¼ f< p1 > ,
K 2 B \ ð[r1 < p1 p3 >g, ½< p2 > ¼ f< p2 > , < p2 p3 >g, ½< p1 p2 > ¼
i¼1 Ai Þ due to Corollary 2.1. This contradicts that
R is a resolving set. Therefore, B is a proper subset of R and f< p1 p2 > , < p1 p2 p3 >g are many element ideal classes.
we may write R ¼ B [ A where A  VðGðZn ÞÞ n B with 1  bðGðZp1 p2 p23 ÞÞ ¼ 3 þ 2 ¼ 5: B ¼ f< p1 > , < p2 > , < p1 p2 > ,
jAj  s  1: Then the following possible cases arise. < p1 p23 > , < p2 p23 >g forms a minimal resolving set as the
Case-1: A  [r2 i¼0 Ai : If jA \ Sj ¼ s  1, then there is an code of the remaining elements with respect to B are
ideal I 2 S n A: Let LðIÞ ¼ fpi g for some i 2 f1, 2, :::, sg: ð1, 1, 1, 1, 2Þ, ð1, 1, 1, 2, 1Þ, ð1, 1, 2, 1, 1Þ, ð1, 1, 1, 1, 1Þ, ð1, 1, 1, 2,
Then dðI, KÞ ¼ dðJ, KÞ ¼ 1 for all K 2 A, where J 2 2Þ, respectively.
VðGðZn ÞÞ n B with LðJÞ ¼ /: This contradicts that R ¼
B [ A is a resolving set. Again if jA \ Sj < s  1, then there
is at least two ideals I, J 2 S which are not in A. Let LðIÞ ¼ 5. Metric dimension of GðZpa1 pa22 . . . par r Þ where ai  2
1
fpi g and LðIÞ ¼ fpj g for some i, j 2 f1, 2, :::, sg with i 6¼ j:
In this section we discuss the metric dimension of
Then by Corollary 2.1, dðI, KÞ ¼ dðJ, KÞ ¼ 1 for all K 2 A:
GðZpa1 , pa2 , pa3 , :::, par r Þ with condition that all ai’s are greater or
Also we have shown that the elements of S are not resolved 1 2 3

by B. Therefore we arrive at a contradiction that R ¼ B [ A equal to 2. Here vertex set is partitioned into 2r  1 ideal
is not a resolving set. classes and all of these ideal classes are many element
Case-2: A  Ar1 : If jA \ Cj ¼ s  1, then there is an ideal classes. These ideal classes are also form twin sets.
ideal I 2 C n A: Let LðIÞ ¼ PðnÞ n fpi g for some i 2 The following theorem gives the metric dimension of
f1, 2, :::, sg: Then for two ideals I1 , I2 2 B having level sets GðZpa1 , pa2 , :::, par r Þ, where ai  2:
1 2
distinct from fpi g, dðI1 , KÞ ¼ dðI2 , KÞ ¼ 1 for all K 2 A:This Q
Theorem 5.1. For n ¼ pa11 , pa22 , pa33 , :::, par r , bðGðZn ÞÞ ¼ ri¼1
contradicts that R ¼ B [ A is a resolving set. If jA \ Cj <
ðai þ 1Þ  2r  1, where all ai’s are greater or equal to 2.
s  1, there is at least two ideals I, J 2 C n A: Let LðIÞ ¼
PðnÞ n fpi g and LðIÞ ¼ PðnÞ n fpi g, where i, j 2 f1, 2, :::, sg: Proof. Let R be an arbitrary minimal resolving set for
Then for two ideals I1 , I2 2 B having level sets distinct from GðZn Þ: In GðZn Þ, 2r  1 ideal classes (all ideal classes ) are
fpi g and fpj g, dðI1 , KÞ ¼ dðI2 , KÞ ¼ 1 for all K 2 A due to many element ideal classes and each of these ideal classes
Corollary 2.1. This contradicts that R ¼ B [ A is a resolv- forms a twin set. Let T1 , :::, T2r 1 be these twin sets.
ing set. Applying Lemma 2.2, we have jR \ Ti j  jTi j  1: Therefore
 AKCE INTERNATIONAL JOURNAL OF GRAPHS AND COMBINATORICS 5

P Q
jRj  ðjTi j  1Þ ¼ jVðGðZn ÞÞj  ð2r  1Þ ¼ ri¼1 ðai þ 1Þ Acknowledgments
Q
2  ð2r  1Þ ¼ ri¼1 ðai þ 1Þ  2r  1 and hence bðGðZn ÞÞ 
Qr We would like to thank heartily to our esteemed reviewers for review-
i¼1 ðai þ 1Þ  2  1: For the reverse inequality, let B be a
r
ing this paper very consciously. Their valuable and precious comments
2r 1
subset of [i¼1 Ti such that jB \ Ti j ¼ jTi j  1 for each i 2 were of immense help to make the paper more presentable and worth-
f1, 2, :::, 2r  1g: We show that B forms a resolving set for while. The first author is thankful to the Science and Research Board
(SERB), DST, India for its financial support (Grant No. CRG/
GðZn Þ: Let I, J 2 VðGðZn ÞÞ n B: With out loss of generality, 2019/006909).
we may assume jLðJÞj  jLðJÞj: Since I and J are distinct,
there exists at least one pi 2 PðnÞ such that pi 2 LðIÞ but pi 6
2 LðJÞ: Again dðK, IÞ ¼ 1 and dðK, JÞ ¼ 2 for all K 2 B with Disclosure statement
LðKÞ ¼ PðnÞ n pi : Thus K resolves I Q and J. Therefore B No potential conflict of interest was reported by the author(s).
forms a resolving set with cardinality ri¼1 ðai þ 1Þ  2r  1
and hence the theorem. w
References
Example 5.1. The graph GðZp21 p22 p23 Þ has metric dimension [1] Basak, M., Saha, L., Das, G., Tiwary, K. (2020). Fault-tolerant
18. For this graph R ¼ f< p1 > , < p3 > , < p2 > , < p1 p2 > , metric dimension of circulant graphs Cn ð1, 2, 3Þ. Theor.
<p1 p3 >, <p2 p3 >, <p21 >,<p21 p2 >, <p21 p3 >, <p22 >,<p22 p1 >, Comput. Sci. 817: 66–79.
< p22 p3 > , < p23 > , < p23 p2 > , < p23 p1 > , < p21 p22 > , < p21 p23 > , [2] Basak, M., Saha, L., Tiwary, K. (2019). Metric dimension of
zero-divisor graph for the ring ðZn Þ. Int. J. Sci. Res. Math. Stat.
< p22 p23 >g is a resolving set with cardinality 18.
Sci. 6(1): 74–78.
[3] Beerliova, Z., Eberhard, F., Erlebach, T., Hall, A., Hoffmann,
M., Mihal’ak, M., Ram, L. S. (2006). Network discovery and
verification. IEEE J. Select. Areas Commun. 24(12):
6. Conclusion 2168–2181.
[4] Chartrand, G., Eroh, L., Johnson, M., Oellermann, O. (2000).
Every simple connected graph is isomorphic to some Resolvability in graphs and the metric dimension of a graph.
intersection graph. For the last few decades several math- Discrete Appl. Math. 105(1–3): 99–113.
ematician studied intersection graphs on various algebraic [5] Chvatal, V. (1983). Mastermind. Combinatorica. 3: 325–329.
[6] Hernando, C., Mora, M., Pelayo, I., Seara, C., Wood, D. (2010).
structures and so it is naturally more convenient to study Extremal graph theory for metric dimension and diameter.
the intersection graph G(F) when the members of F have Electron. J. Combin. 17(1): 1–28.
some algebraic structure. In this article, we consider the [7] Harary, F., Melter, R. (1976). On the metric dimension of a
intersection graphs G(F) when the set F is an ideal and graph. Ars Combin. 2: 191–195.
we determine the metric dimension of ideal intersection [8] Khuller, S., Raghavachari, B., Rosenfeld, A. (1996). Landmarks
graphs GðZn Þ: Using similar type of arguments as we in graphs. Discrete Appl. Math. 70(3): 217–229.
[9] Saha, L. (2021). Fault-tolerant metric dimension of cube of
have used in this article, the researchers may find the
paths. J. Phys. Conf. Ser. 1714(1): 012029.
metric dimension of ideal intersection graphs G(R) for [10] Slater, P. (1975). Leaves of trees. Cong. Numer. 14: 549–559.
any ring R which are isomorphic to some sub-ring [11] Tomescu, I., Melter, R. A. (1984). Metric bases in digital geom-
of ðZn Þ: etry. Comput. Vis. Graph. Image Process. 25(1): 113–121.

View publication stats