THE METRIC DIMENSION OF FRIENDSHIP GRAPH Fn , LOLLIPOP GRAPH L m,n AND PETERSEN GRAPH P n,m

  Bulletin of Mathematics

  ISSN Printed: 2087-5126; Online: 2355-8202

Vol. 08, No. 02 (2016), pp.117-124 http://jurnal.bull-math.org

THE METRIC DIMENSION OF FRIENDSHIP

GRAPH F n

  , LOLLIPOP GRAPH L m ,n

  AND PETERSEN GRAPH P n ,m

  

Mulyono and Wulandari

Abstract.

  Let u and v are vertices in connected graph G, the distance d(u, v)

is the length of the shortest path between the vertices u and v. For an ordered

subset W = {w 1 , w 2 , w 3 , . . . , w k } of vertices in a connected graph G and a vertex v

  

∈ V (G), a metric representation of v with respect to W is the k−vector r(v|W ) =

(d(v, w 1 ), d(v, w 2 ), . . . , d(v, w k )). The subset W is a resolving set if r(v|W ) for

every two vertices of G have distinct representations. The minimum cardinality of

resolving set for G is called the metric dimension of G and denoted by dim(G). This

paper is devoted to determine the metric dimension of friendship graph F n , lollipop graph L m,n and Petersen graph P n,m for m = 1. f5.

  1. INTRODUCTION Metric dimension is one of subjects in graph theory. The problem of studying the metric dimension was firstly introduced by Slater in 1975. Harary and Melter [6] proposed the same concept in their paper ’On the Metric Dimension of a Graph’. This paper focused on a concept namely metric representation that is a way to represent vertex location in a graph.

  Let u and v are vertices in connected graph G, the distance d(u, v) is the length of the shortest path between the vertices u and v. For an

  Received 16-08-2016, Accepted 21-09-2016. 2010 Mathematics Subject Classification : 05C20, 94C12 Key words and Phrases

  : resolving set, basis, metric dimension, friendship graph, lollipop graph, Petersen graph

  Mulyono & Wulandari – Metric Dimension

  1

  , w , w , . . . , w ordered subset W = {w

  2 3 k } of vertices in a connected graph

  G and a vertex v ∈ V (G), the metric representation of v with respect to W is the k−vector r(v|W ) = (d(v, w ), d(v, w ), . . . , d(v, w k )). The following

  1

  2 definition about metric dimension was proposed by Harary and Melter [6].

  Definition 1.1 The subset W is a resolving set if r(v|W ) for every two vertices of G have distinct representations. A resolving set of minimum cardinality for graph G is called a minimum resolving set or a basis for G.

  The metric dimension of G, denoted by dim(G), is the number of basis for G .

  The concept of metric dimension has proved to be useful in a variety of fields. Chartrand et al [2] applied the resolving set of metric dimension in chemistry to classify the chemical compound. Khuller et al [12] also applied in robotic navigation. Furthermore, Sebo et al [11] applied in combinatorial search and optimization.

  Moreover Hindayani [7] has studied the determining metric dimension of K r + mK s graph. The results are dim(K r + mK s ) = m + (r − 2) for m ≥ 2, s = 1, and dim(K r + mK s ) = (s − 1)m + (r − 1) for m, s ≥ 2. Another work related to the metric dimension is proposed by Permana [9] on the determining metric dimension for some trees in specific shape. He obtained dim(C ) = m(n − 1) for m ≥ 1, n ≥ 2, dim(F ) = m(n − 1)

  m,n m,n for m, n ≥ 2, and dim(B m,n ) = m(n − 2) for m ≥ 2, n ≥ 3 .

  In this paper we consider the metric dimension of friendship graph F n , lollipop graph L and Petersen graph P for m = 1.

  m,n n,m

  2. PRELIMINARIES , v , v , . . . A graph G consists of a set of objects V (G) = {v

  1

  2 3 } called

  vertices and other set E(G) = {e , e , e , . . . } whose elements are called

  1

  2

  3

  edges and graph is usually denoted as G = (V (G), E(G)) [13]. A graph at least has one vertex and perhaps has no edge. The number of vertices in G denoted by |V (G)| is often called the order of G , while the number of edges denoted by |E(G)| is its size [4].

  The edge e = (u, v) is denoted to join the vertices u and v. If e = (u, v) is an edge of a graph G, then u and v are adjacent vertices, while u and e are incident, as are v and e [3].

  For a connected graph G, we define the distance d(u, v) between two

  Mulyono & Wulandari – Metric Dimension

  d (u, v) ≥ 0 for all pairs u, v of vertices of G, and d(u, v) = 0 if and only if u = v [3].

  A friendship graph F n is a graph that can be constructed by coalescence n copies of the cycle graph C

  3 of length 3 with a common vertex. The friendship graph F n is also planar graph with 2n + 1 vertices and 3n edges.

  , v , v , , v The vertices set is V (F n ) = {c, v } and the edges set is E(F n ) =

  1

  2 2 2n {cv , cv , cv , , cv n } ∪ {v v , v v , . . . , v v , . . . , v v } for n ≥ 2 [1].

  1

  2

  3

  2

  1

  2

  3 4 2i−1 2i 2n−1 2n The following figure shows the friendship graph in common.

  Figure 1: Friendship graph F n A lollipop graph, denoted by L m,n (shown in Figure 2), is a graph which is constructed by appending a complete graph K m , m ≥ 3, to a pendant vertex of path graph P n . The vertices set denoted as V (L m,n ) =

  , u , u , . . . , u , v , v , v , . . . , v {u n m } [10].

  1

  2

  3

  1

  2

  3 Figure 2: Lollipop graph L m,n n −1

  A Petersen graph, denoted by P for n ≥ 3 and 1 ≤ m ≤ , is a

  n,m

  2

  , u , . . . , u , v , v , . . . , v 3−regular graph with 2n vertices V (P n,m ) = {u n n }

  1

  2

  1

  2

  and 3n edges E(P ) = {u u , u v , v v }∀i ∈ {1, 2, . . . , n}, where the

  n,m i i +1 i i i i +m

  subscripts are reduced by modulo n. The following figure shows the Petersen graph P n,

  1 .

  Mulyono & Wulandari – Metric Dimension

  Figure 3: Petersen graph P n,

  1

  3. RESULTS

  3.1 n

  The Metric Dimension of Friendship Graph F

  We begin by providing a stronger result what we indicated in the preceding section. Theorem 3.1

  For all integer n ≥ 2, dim(F n ) = n Proof 3.1

  , v , v , . . . , v We choose a subset W = {v }, and we must

  1

  3 5 2n−1

  show that dim(F ) = n for n ≥ 2. By definition 1.1, we got the representa-

  n

  tions of vertices in graph F n with respect to W are r (c|W ) = (1, 1, 1, . . . , 1, 1) r

  (v

  1 |W ) = (0, 2, 2, . . . , 2, 2)

  r (v |W ) = (1, 2, 2, . . . , 2, 2)

  2

  r (v |W ) = (2, 0, 2, . . . , 2, 2)

  3

  r (v |W ) = (2, 1, 2, . . . , 2, 2)

  4

  r (v |W ) = (2, 2, 0, . . . , 2, 2)

  5 .. ..

  . = . r

  (v 2n−1 |W ) = (2, 2, 2, . . . , 2, 0) r (v |W ) = (2, 2, 2, . . . , 2, 1)

  2n From above, the representations of vertices in graph F n are distinct.

  This impiles that W is resolving set, but it is not necessarily the lower bound. Thus the upper bound is dim(F n ) ≤ n.

  Now, we show that dim(F n ) ≥ n. Let W = {v , v , v , . . . , v }

  1

  3 5 2n−1

  is a resolving set which is |W | = n. Assume that W is another mini-

  1

  mum resolving set or we can denote |W | < n. If we choose an ordered set

  1 W

  , v ⊆ W − {v i }, i is odd, so that there are two vertices v i i ∈ F n such

  1

  • 1

  that r(v |W ) = r(v |W ) = (2, 2, 2, . . . , 2, 2). W is not a resolving set, a

  i i +1

  1 contradiction with assumption. Thus the lower bound is dim(F n ) ≥ n. Mulyono & Wulandari – Metric Dimension Thus, another strong result is showed in the following theorem.

  3.2 The Metric Dimension of Lollipop Graph L m,n

  , v

  , v

  3

  , . . . , v

  m −1

  } is a resolving set which is |W | = m−1. Assume that W

  1

  is another minimum resolving set or we can denote |W

  1

  | < m − 1. If we choose an ordered set W

  1

  ⊆ {v

  1

  2

  , v

  , v

  3

  , . . . , v

  m

  } − {v

  i

  , v

  j

  }, 1 ≤ i, j ≤ m, i 6= j, so that there are two vertices v i , v j ∈ L m,n such that r(v i |W ) = r(v j |W ) = (1, 1, 1, . . . , 1, 1). W

  1

  is not a resolving set, a contradiction with assumption. Thus the lower bound is dim(L m,n ) ≥ m − 1.

  From the above proving, we conclude that dim(F n ) = n. We continue to the other strong result.

  3.3 The Metric Dimension of Petersen Graph P n,m

  2

  1

  Theorem 3.2 For all integer m ≥ 3 and n ≥ 1, dim(L m,n ) = m − 1

  3

  Proof 3.2 We choose a subset W = {v

  1

  , v

  2

  , v

  3

  , . . . , v m

  −1

  }, and we must show that dim(L m,n ) = m − 1 for m ≥ 3, n ≥ 1. By definition 1.1, we got the representations of vertices in graph L m,n with respect to W are r (v

  1

  |W ) = (0, 1, 1, . . . , 1, 1) r (v

  2

  |W ) = (1, 0, 1, . . . , 1, 1) r (v

  |W ) = (1, 1, 0, . . . , 1, 1) .. . = ..

  Now, we show that dim(L m,n ) ≥ m−1. Let W = {v

  . r (v m

  −1

  |W ) = (1, 1, 1, . . . , 1, 0) r (v m |W ) = (1, 1, 1, . . . , 1, 1) r (u

  1

  |W ) = (1, 2, 2, . . . , 2, 2) r (u

  2

  |W ) = (2, 3, 3, . . . , 3, 3) r (u

  3

  |W ) = (3, 4, 4, . . . , 4, 4) .. . = ..

  . r (u n

  −1

  |W ) = (n − 1, n, n, . . . , n, n) r (v n |W ) = (n, n + 1, n + 1, . . . , n + 1, n + 1) From above, the representations of vertices in graph L m,n are distinct.

  This impiles that W is resolving set, but it is not necessarily the lower bound. Thus the upper bound is dim(L m,n ) ≤ m − 1.

  Theorem 3.3 For the Petersen graph P n,m we have

  Mulyono & Wulandari – Metric Dimension

  (i) dim (P n,m ) = 2 for m = 1, n is odd, n ≥ 3 dim (ii) (P n,m ) = 3 for m = 1, n is even, n ≥ 3

  Proof 3.3 Case (i) Showed that dim(P n,m ) = 2 for m = 1, n is odd, n ≥ 3

  n +1

  We choose a subset W = {u , u }, k = , and we must show that

  1

1 k

  2

  dim (P n,m ) = 2 for m = 1, n is odd, n ≥ 3 . By definition 1.1, we got the representations of vertices in graph P with respect to W are

  n,m

  r (u |W ) = (0, k − 1)

  1

  r (u |W ) = (1, k − 2)

  2

  r (u

  3 |W ) = (2, k − 3) .. ..

  . = . r

  (u k |W ) = (k − 1, k − 1) r (u |W ) = (k − 1, 1)

  k +1 .. ..

  . = . r

  (u n |W ) = (1, k − 1) r (v |W ) = (1, k)

  1

  r (v |W ) = (2, k − 1)

  2

  r (v

  3 |W ) = (3, k − 2) .. ..

  . = . r (v |W ) = (k, 1)

  k

  r (v k |W ) = (k, 2)

  • 1 .. ..

  . = . r

  (v n |W ) = (2, k) From above, the representations of vertices in graph P n,m are distinct. This impiles that W is resolving set with |W | = 2. Obtained the upper bound is dim(P n,m ) ≤ 2. For the Petersen graph P n,m , there is no resolving set that the cardinality is one. Thus the lower bound is dim(P n,m ) ≥ 2. Obtained that dim(P n,m ) ≤ 2 and dim(P n,m ) ≥ 2 , therefore dim(P n,m ) = 2.

  Case (ii) Showed that dim(P n,m ) = 3 for m = 1, n is even, n ≥ 4

  n

  • 2

  We choose a subset W = {u , u k , u n } with k = , and we must show

  1

  2

  that dim(P n,m ) = 3 for m = 1, n is even, n ≥ 4 . By definition 1.1, we got the representations of vertices in graph P n,m with respect to W are

  Mulyono & Wulandari – Metric Dimension

  r (u |W ) = (0, k − 1, 1)

  1

  r (u

  2 |W ) = (1, k − 2, 2)

  r (u |W ) = (2, k − 3, 3)

  3 .. ..

  . = . r

  (u k |W ) = (k − 1, 0, k − 2) r (u |W ) = (k − 2, 1, k − 3)

  k +1 .. ..

  . = . r

  (u n |W ) = (1, n − k, 0) r (v |W ) = (1, k, 2)

  1

  r (v |W ) = (2, k − 1, 3)

  2

  r (v |W ) = (3, k − 2, 4)

  3 .. ..

  . = . r

  (v k |W ) = (k, 1, k − 1) r (v |W ) = (k − 1, 2, k − 2)

  k +1 .. ..

  . = . r (v n |W ) = (2, k − 1, 1) From above, the representations of vertices in graph P n,m are distinct.

  This impiles that W is resolving set, but it is not necessarily the lower bound. Thus the upper bound is dim(P ) ≤ 3.

  n,m

  Now, we show that dim(P ) ≥ 3. Assume W is another minimum

  n,m

  1

  resolving set of P n,m for m = 1, n is even, n ≥ 4 with |W | < 3. If we choose

  1

  an ordered set W ⊆ W − {v }, so that there are the same representations

  1 k

  r (u |W ) = r(v |W ) = (1, 2), r(u n |W ) = r(v n |W ) = (2, 1), r(u k |W ) =

  2 1 −1

  r (v k |W ) = (2, 3), r(u k |W ) = r(v k |W ) = (3, 2). W

  1 is not a resolving −1 +1 +2

  set, a contradiction with assumption. Thus the lower bound is dim(P n,m ) ≥ 3.

  From the above proving, we conclude that dim(P n,m ) = 3. . This completes the proof of of Theorem 3.3

  

REFERENCES

1.

  A. Abdollahi, S. Janbaz & M. R. Oboudi. 2013. Graphs Cospectral With A Friendship Graph or Its Complement. Transactions on Combinatorics.

  2. G. Chartrand, D. Erwin, G. L. Johns, and P. Zhang. Boundary vertices in graphs. Discrete Math., 263(1-3):2534, 2003.

  3. Chartrand, G., Linda Lesniak. 1986. Applied and Algorithmic Graph

  Mulyono & Wulandari – Metric Dimension 4.

  Chartrand, G., Ping Zhang. 2012. A First Course in Graph Theory, Dover Publication, Inc., New York.

  

5. Goodaire, Edgar G., Michael M. Parmenter. 2008. Discrete Mathematics

with Graph Theory Third Edition, Pacific Graw, California.

  

6. Harary and Melter. 1976. On the Metric Dimension of A Graph. Com-

  binatoria , m, r, s

  

7. Hindayani. 2011. Dimensi Metrik Graf K r + mK s ∈ N . Cauchy

  

8. Ngurah, AAG., Edi Try Baskoro. 2003. On Magic and Antimagic Label-

  ing Generalized of Petersen Graph. Utilitas Math 9. Permana, Angga Budi, Darmaji. 2012. Dimensi Metrik Graf Pohon Ben- tuk Tertentu. Jurnal Teknik POMITS

  10. Siddiq, Muhammad. Parmenter. 2014. Pemberian Nomor Vertex pada

  Topologi Jaringan Graf Wheel, Graf Helm, dan Graf Lollipop [Skripsi], Universitas Sebelas Maret, Surakarta.

  

11. Sebo and Tannier. 2004. On Metric Generators of Graphs. Math. Oper.

  Res.

  12. S. Khuller, B. Raghavachari and A.Rosenfeld. 1996. Landmarks in Graphs. Appl. Math.

  

13. Vasudev, C. 2006. Graph Theory with Applications, New Age Interna-

Mulyono tional P. Limited, Bangalore.

  : Mathematics Department, Faculty Mathematics and Natural Science, Universitas Negeri Medan. Wulandari E-mail: mulyono mat@yahoo.com : Mathematics Department, Faculty Mathematics and Natural Science, Universitas Negeri Medan.

  E-mail: wulandari 0910@yahoo.com