Volume 5, No.

5, May 2018
Journal of Global Research in Mathematical Archives
UGC Approved Journal
RESEARCH PAPER
Available online at http://www.jgrma.info

Domination in Cube of Graphs

B. Basavanagoud∗ and Sujata Timmanaikar∗∗

∗Department of Mathematics, Karnatak University, Dharwad, India
E-mail: [email protected]
∗∗ Department of Mathematics, Government Engineering College, Haveri, India
E-mail: [email protected]

Abstract:Let 𝐺 = (𝑉, 𝐸) be a graph. The cube of a graph 𝐺 is denoted by 𝐺 3 has the same vertex set as in 𝐺
and every two vertices 𝑢, 𝑣 ∈ 𝑉(𝐺 3 ) are adjacent in 𝐺 3 if and only if they are joined by a path of length ≤ 3.
In this paper, we establish the bounds for the domination number of cube of 𝐺 in terms of 𝐺. Also we
investigate their relationship with other domination parameters.

Keywords: domination; square of a graph; cube of a graph.

AMS Subject Classification: 05C69.

1 Introduction
All graphs considered throughout this paper are simple and nontrivial. For undefined terms or
notations may be found in [4]. Let 𝐺 = (𝑉, 𝐸) be a graph. The number of vertices of 𝐺 we denote by 𝑛 and
the number of edges we denote by 𝑚, thus |𝑉(𝐺)| = 𝑛 and |𝐸(𝐺)| = 𝑚. By the open neighborhood of a
vertex 𝑣 of 𝐺 we mean the set 𝑁𝐺 (𝑣) = {𝑢 ∈ 𝑉(𝐺): 𝑢𝑣 ∈ 𝐸(𝐺)}. By the closed neighborhood of a vertex 𝑣
of 𝐺 we mean the set 𝑁𝐺 [𝑣] = 𝑁𝐺 (𝑣) ∪ {𝑣}. The degree of a vertex 𝑣, denoted by 𝑑𝑒𝑔(𝑣), is the cardinality
of its open neighborhood. A vertex is called isolated if it has no neighbors, while it is called universal if it is
adjacent to all other vertices. Let Δ(𝐺)(Δ′) mean the maximum degree(edge) among all vertices(edges) of 𝐺.
The path (cycle, respectively) on 𝑛 vertices we denote by 𝑃𝑛 (𝐶𝑛 , respectively). A wheel 𝑊𝑛 , where 𝑛 ≥ 4,
is a graph with 𝑛 vertices, formed by connecting a vertex to all vertices of a cycle 𝐶𝑛−1 . By 𝐾𝑝,𝑞 we denote a
complete bipartite graph with partite sets of cardinalities 𝑝 and 𝑞.

2 Motivation
M. A. Henning et al.[7] gave an interpretation for the k-distance domination and domination of the
power of a graph 𝐺, such as 𝛾𝑘 (𝐺) = 𝛾(𝐺 𝑘 ), for any connected 𝐺. F. Harary et al. [3] defined squares of
graph and using this M. H. Muddebihal et al.[8] used the notation for 2-distance domination 𝛾2 (𝐺) as
𝛾(𝐺 2 ) and called it as domination in 𝐺 2 of a graph 𝐺. Motivated by this we defined cubes of graphs and used
the notation for 3-distance domination 𝛾3 (𝐺) 𝑎𝑠 𝛾(𝐺 3 ) and called it as domination in 𝐺 3 of a graph 𝐺.
Since the distance version of domination have a strong background of applications in interconnection
networks, many efforts have been made by several authors to consider the distance parameters. The diameter
of the graph defined as the maximum distance between any two vertices, represents the maximum

© JGRMA 2018, All Rights Reserved 82

 B. Basavanagoud et al, Journal of Global Research in Mathematical Archives, 5(5), 82-89

communication link between any two vertices of the network. As a consequence, when finding
interconnection network models it is important for the diameter to be small as possible.
In this paper we establish bounds for the domination number of cubes of graphs.

3 Preliminaries
We now define domination and its related parameters which have been studied by different
mathematicians. These parameters are of great practical interest because of the applications of domination
theory in different fields.
A set 𝐷 of vertices in a graph 𝐺 is a dominating set if every vertex in 𝑉 − 𝐷 is adjacent to some
vertex in 𝐷. The domination number 𝛾(𝐺) of 𝐺 is the minimum cardinality of a dominating set of 𝐺. A
minimum dominating set of a graph 𝐺 is called a 𝛾 − set of 𝐺.

• A connected dominating set is a dominating set in which 〈𝐷〉 is connected.

• A dominating set 𝐷 is independent dominating set if 〈𝐷〉 is independent.

• A dominating set 𝐷 is called a total dominating set if there are no isolates in 〈𝐷〉.

• A dominating set 𝐷 is called a split dominating set if 〈𝑉 − 𝐷〉 is disconnected.

The minimum cardinality taken over all connected \ independent \ total \ split \ nonsplit \
cototal \ maximal gives the respective domination numbers 𝛾𝑐 (𝐺) \ 𝛾𝑖 (𝐺) \ 𝛾𝑡 (𝐺) \ 𝛾𝑠 (𝐺) respectively.
The distance 𝑑(𝑢, 𝑣) between two vertices 𝑢 and 𝑣 in 𝐺 is the length of a shortest path joining
them if any; otherwise 𝑑(𝑢, 𝑣) = ∞. A shortest 𝑢 − 𝑣 path is called a geodesic. The diameter of a connected
graph 𝐺 is the length of any longest geodesic and is denoted as 𝑑𝑖𝑎𝑚(𝐺). The eccentricity 𝑒(𝑣) of a vertex
𝑣 in a connected graph 𝐺 is 𝑚𝑎𝑥 𝑑(𝑢, 𝑣) for all 𝑢 in 𝐺. The radius 𝑟(𝐺) is the minimum eccentricity of
the points. The maximum eccentricity is the diameter. A point 𝑣 is a central point if 𝑒(𝑣) = 𝑟(𝐺) and the
center of 𝐺 is the set of all central vertices. The girth of a graph 𝐺 denoted by 𝑔(𝐺), is the length of a
shortest cycle in 𝐺. For more details about domination and distance concepts interested readers can refer [1, 2,
5, 6].

Definition 1. 1 Let 𝐺 = (𝑉, 𝐸) be a graph. The cube of a graph 𝐺 is denoted by 𝐺 3 has the same vertex set
as in 𝐺 and every two vertices 𝑢, 𝑣 ∈ 𝑉(𝐺 3 ) are adjacent in 𝐺 3 if and only if they are joined by a path of
length one or two or three.

4 Results
Since we are considering all connected graphs with diameter at least three. Therefore the domination
number of cube of complete graph, wheel graph, star graph, complete bipartite graph does not make any sense.
Therefore from the list of standard class of graphs we are considering only path graph (𝑃𝑛 ; 𝑛 ≥ 4) and cycle
graph (𝐶𝑛 ; 𝑛 ≥ 6).

Observation 1. 2 For any graph 𝐺, 𝛾(𝐺 2 ) ≤ 𝛾(𝐺) and 𝛾(𝐺 3 ) ≤ 𝛾(𝐺 2 ).

The following result gives the exact values of domination number of cube of path and cycle graph

© JGRMA 2018, All Rights Reserved 83

 B. Basavanagoud et al, Journal of Global Research in Mathematical Archives, 5(5), 82-89

respectively.

𝑛
Proposition 2. 3 If 𝐺 = 𝑃𝑛 or 𝐶𝑛 , then 𝛾(𝑃𝑛3 ) = = 𝛾(𝐶𝑛3 ).
7

Proof. Let 𝑣1 , 𝑣2 , ⋯ , 𝑣𝑛 be the vertices of a path 𝑃𝑛 . By the definition of cubes of a graph 𝑣1 is adjacent to
𝑣2 , 𝑣3 and 𝑣4 respectively, and 𝑣2 is adjacent to 𝑣1 , 𝑣3 , 𝑣4 and 𝑣5 and so on. In general every 𝑣𝑖 is
adjacent to 𝑣𝑖−1 , 𝑣𝑖−2 , 𝑣𝑖−3 , 𝑣𝑖+1 , 𝑣𝑖+2 and 𝑣𝑖+3 only when 𝑖 ≥ 2. Now we construct a vertex for 𝐺 3 as
𝑛
follows: 𝑋 = {𝑣4 , 𝑣8 , 𝑣12 , ⋯ 𝑣4𝑖 } where 1 ≤ 𝑖 ≤ . The above set 𝑋 is a dominating set for 𝐺 3 because
7
every vertex not in 𝑋 is adjacent to a vertex in 𝑋. Moreover for each vertex 𝑣𝑖 ∈ 𝑋 there exist a vertex
𝑢𝑖 ∈ 𝑉 − 𝑋 such that 𝑁(𝑢𝑖 ) ∩ 𝑋 = {𝑣}. Now each vertex in the set 𝑋 is of maximum degree in 𝐺 3 and the
𝑛
vertices 4𝑖 for 1 ≤ 𝑖 ≤ . being nonadjacent to each other will dominate maximum distinct vertices of
7
𝑛
𝐺 3 . Therefore the set 𝑋 is of minimum cardinality. Hence 𝛾(𝐺 3 ) = |𝑋| = . . Hence, the domination
7
𝑛 𝑛
number of cubes of path graph 𝑃𝑛 is . i.e 𝛾(𝑃𝑛 )3 = 7
..
7
The proof for cubes of cycle graph 𝐶𝑛 follows the same lines of the proof of cubes graph of path graph
𝑃𝑛 .

Theorem 3. 4 For any graph 𝐺, 𝛾(𝐺 3 ) = 1 if and only if there exist a central vertex 𝑣 ∈ 𝑉(𝐺) such that
𝑒(𝑣) ≤ 3.

Proof. Let 𝐺 be any graph with 𝑑𝑖𝑎𝑚(𝐺) ≥ 3. Let 𝐷 be a minimal dominating set of 𝐺 3 . If 𝛾(𝐺 3 ) = 1 and
𝐺 contains a central vertex 𝑣 ∈ 𝐺 such that 𝑒(𝑣) ≥ 4. Then there exist a vertex 𝑢 ∈ 𝑉\𝐷 such that 𝑢 is not
dominated by any vertex in 𝐷. Hence 𝐷′ = 𝐷 ∪ {𝑢} will form a minimal dominating set for 𝐺 3 , which is a
contradiction to the minimality of 𝐷. Hence 𝐺 must contain a vertex 𝑣 such that 𝑒(𝑣) ≤ 3.

Theorem 4. 5 If 𝐻 is a spanning subgraph of 𝐺 then 𝛾(𝐺) ≤ 𝛾(𝐻) and 𝛾(𝐺 3 ) ≤ 𝛾(𝐻 3 ).

Proof. It is well known fact that if 𝐻 is a spanning subgraph of 𝐺 then 𝛾(𝐺) ≤ 𝛾(𝐻). Now the only part we
have to prove is 𝐻 3 is spanning subgraph of 𝐺 3 . Since 𝐻 is spanning subgraph of 𝐺 therefore by the
definition of 𝐺 3 , 𝐻 3 must be a spanning subgraph of 𝐺 3 . Hence, 𝛾(𝐺 3 ) ≤ 𝛾(𝐻 3 ).

5 Bound for 𝜸(𝑮𝟑 )

First, we obtain an upper bound for 𝛾(𝐺 3 ) in terms of order and maximum degree Δ(𝐺 3 ).

Theorem 5. 6For any connected graph 𝐺, 𝛾(𝐺 3 ) ≤ 𝑛 − 𝛥(𝐺 3 ).

Proof. Let 𝐺 be any connected graph with 𝑑𝑖𝑎𝑚(𝐺) = 𝑘 where 𝑘 ≥ 3. Since 𝛾(𝐺 3 ) ≤ 𝛾(𝐺). Therefore,
the result follows from Observation 1.

Next, we obtain an upper bound for 𝛾(𝐺 3 ) in terms of vertex connectivity 𝜅(𝐺 3 ).
Theorem 6. 7 For any graph 𝐺, with 𝑛 ≥ 4,

© JGRMA 2018, All Rights Reserved 84

 B. Basavanagoud et al, Journal of Global Research in Mathematical Archives, 5(5), 82-89

𝛾(𝐺 3 ) ≤ 𝑛 − 𝜅(𝐺 3 ).

Proof. Since for disconnected graph 𝐺, we cannot define 𝐺 3 . So we shall assume that 𝐺 is connected graph
of order at least four with 𝑑𝑖𝑎𝑚(𝐺) ≥ 3. Which implies that 𝐺 is non-complete graph. Let {𝑣1 , 𝑣2 , ⋯ , 𝑣𝑘 } =
𝐹 be the set of vertices whose removal results in a disconnected graph of 𝐺 3 . That is 𝐹 is a minimum vertex
cut of 𝐺 3 (i.e., 𝜅(𝐺 3 ) = |𝐹|). Next we have to show that 𝑉(𝐺 3 ) − 𝐹 is a dominating set of 𝐺. Suppose this
is not true then there exists a vertex 𝑣𝑖 ∈ 𝐹; 1 ≤ 𝑖 ≤ 𝑘 such that 𝑁(𝑣𝑖 ) and 𝑉 − 𝐹 are disjoint. Also
𝑣𝑖 ∈ 𝐹 − 𝑁(𝑣𝑖 ) which implies that 𝑁(𝑣𝑖 ) is a proper subset of 𝐹. So |𝐹| ≤ 𝑛 − 2 and 𝑉(𝐺 3 ) − 𝐹 has at
least one element. Therefore the subgraph induced by the set 𝑉(𝐺 3 ) − 𝑁(𝑣𝑖 ) is disconnected. Therefore the
removal of neighborhood vertices of 𝑣𝑖 results in a disconnected graph of 𝐺 3 , that is 𝑁(𝑣𝑖 ) is the minimal
vertex cut of 𝐺 3 . But 𝐹 is a minimal vertex cut of 𝐺 3 which is a contradiction to the minimality 𝐹.
Therefore, 𝑉(𝐺 3 ) − 𝐹 is a dominating set 𝐺 3 . Thus,
𝛾(𝐺 3 ) ≤ |𝑉(𝐺 3 ) − 𝐹|
≤ 𝑛 − 𝜅(𝐺 3 )
.

Theorem 7. 8 Let 𝐺 3 be a cube graph of a graph 𝐻 of order 𝑛 such that 𝛾(𝐺 3 ) = 𝑛 − 𝜅(𝐺 3 ). Then 𝐺 3 is
regular graph.

Proof. By Theorem 5, we have 𝛾(𝐺 3 ) ≤ 𝑛 − Δ(𝐺 3 ) and since 𝛾(𝐺 3 ) = 𝑛 − 𝜅(𝐺 3 ). We have 𝑛Δ(𝐺 3 ) ≥
𝑛 − 𝜅(𝐺 3 ). Which gives, Δ(𝐺 3 ) ≥ 𝜅(𝐺 3 ). Since, 𝜅(𝐺 3 ) ≤ 𝛿(𝐺 3 ) therefore Δ(𝐺 3 ) ≤ 𝛿(𝐺 3 ) and 𝛿(𝐺 3 ) ≤
Δ(𝐺 3 ) is always true. Hence 𝐺 3 is a regular graph.

𝑛
Theorem 8. If 𝐺 3 is hamiltonian, then 𝛾(𝐺 3 ) ≤ .
7
Proof. Let 𝐺 be any connected graph of order at least four with 𝑑𝑖𝑎𝑚(𝐺) ≥ 3. Since 𝐺 3 is hamiltonian
𝑛
therefore 𝐺 3 contains a spanning cycle 𝐻 such that 𝛾(𝐻) = . By Observation 1, we have
3

𝛾(𝐺 3 ) ≤ 𝛾(𝐺)
≤ 𝛾(𝐻)
𝑛
≤ .
3

Next, we find the lower bound for cube of a graph 𝐺, in terms of diameter of a graph 𝐺.

𝑑𝑖𝑎𝑚(𝐺)+1
Theorem 9. 9 For any connected graph 𝐺, ≤ 𝛾(𝐺3 ). Further equality holds for 𝐺 = 𝑃𝑛 .
7

Proof. Let 𝐺 be any connected graph with 𝑑𝑖𝑎𝑚(𝐺) ≥ 3 and let 𝑉(𝐺) = {𝑣1 , 𝑣2 , 𝑣3 , ⋯ , 𝑣𝑛 } and 𝐸(𝐺) =
{𝑒1 , 𝑒2 , 𝑒3 , ⋯ , 𝑒𝑚 }. By the definition of cubes of graphs every vertex 𝑣𝑖 ∈ 𝑉(𝐺) is adjacent to its neighbors
and all the vertices at a distance 2 and 3 respectively. Clearly the degree of each pendant vertex in 𝐺 3 is 3. i.e,
𝑑𝑒𝑔𝐺 3 (𝑣𝑖 ) = 3 and degrees of all support vertices will be 4. Finally the degree of remaining vertices will be at
𝑛
least 5. Let 𝐷 = {𝑣1 , 𝑣2 , ⋯ , 𝑣𝑘 } for 1 ≤ 𝑘 ≤ be a dominating set of 𝐺 3 . Consider an arbitrary path of
7
length 𝑑𝑖𝑎𝑚(𝐺) and this diametral path induces at most six edges from the induced subgraph 〈𝑁[𝑣]〉 for
each 𝑣 ∈ 𝐷 because 𝑑𝑒𝑔𝐷 (𝑣𝑖 ) ≥ 6. Furthermore, the diametral path induces at most 𝛾(𝐺 3 ) − 1 edges

© JGRMA 2018, All Rights Reserved 85

 B. Basavanagoud et al, Journal of Global Research in Mathematical Archives, 5(5), 82-89

joining the neighborhoods of the vertices of 𝐷. Hence

𝑑𝑖𝑎𝑚(𝐺) ≤ 6𝛾(𝐺 3 ) + 𝛾(𝐺 3 ) − 1
𝑑𝑖𝑎𝑚 (𝐺)+1
≤ 𝛾(𝐺 3 ).
7

If 𝐺 = 𝑃𝑛 , then we know that 𝑑𝑖𝑎𝑚(𝑃𝑛 ) = 𝑛 − 1 and by Proposition 2, result follows.

In the following Theorems 5-11, we establish different upper bounds for 𝛾(𝐺 3 ).

𝑛
Theorem 10. 10 For any connected graph 𝐺, 𝛾(𝐺 3 ) ≤ . Further, equality holds if 𝐺 = 𝑃𝑛 or 𝐶𝑛 .
7

Proof. Let 𝐺 be any graph of order 𝑛 and size 𝑚 and let cubes of a graph 𝐺 be 𝐺 3 . Let 𝐷 be a dominating
𝑛
set of 𝐺 3 . Then we have to show that the cardinality of 𝐷 must not exceed ≤ . To show this, we have to
7
choose the vertices in 𝐺 which are adjacent to at least six vertices in 𝐺 3 . consider an induced path 𝑃7 say 𝑋1
in which the vertices are labeled as 𝑣1 𝑣2 𝑣3 𝑣4 𝑣5 𝑣6 𝑣7 in 𝐺. Here we have to select 𝑣4 which is adjacent at
least six vertices in 𝐺 3 . Therefore 𝑣4 ∈ 𝐷. Similarly select another edge disjoint induced path 𝑃7 say 𝑋2 in
𝐺 with labeled as 𝑣1 ′𝑣2 ′𝑣3 ′𝑣4 ′𝑣5 ′𝑣6 ′𝑣7 ′ here we have to choose 𝑣7 ′ to the set 𝐷. Continuing this process for
all induced paths 𝑋𝑖 ′𝑠 in 𝐺 and choose a vertex 𝑣4𝑖 to the set 𝐷.
𝑛
Hence 𝐷 = {𝑣4 , 𝑣4′ , ⋯ , 𝑣4𝑖′ } for 1 ≤ 𝑖 ≤ will form a dominating set for 𝐺 3 . Further, note that
7
𝑛
set 𝑋 is of maximum degree and 𝐺 3 and the vertices 4𝑖 for 1 ≤ 𝑖 ≤ being nonadjacent to each other
7
will dominate maximum distinct vertices of 𝐺 3 . Therefore the set 𝑋 is of minimum cardinality.
Thus
𝛾 𝐺 3 = |𝐷|
𝑛
= 𝑣4𝑖 , 1 ≤ 𝑖 ≤ 7
𝑛
≤ .
7
Suppose 𝐺 = 𝑃𝑛 or 𝐶𝑛 , then the result follows from Proposition 2.

Corollary 11. 11 Let 𝐺 be any connected graph. Suppose 𝑑𝑖𝑎𝑚(𝐺) = 𝑘 for some positive integer 𝑘, then
𝑘
𝛾(𝐺 3 ) ≤ .
7

𝑛−𝛥(𝐺)+2
Theorem 12. 12 For any connected graph 𝐺 with 𝛿(𝐺) ≥ 2, 𝛾(𝐺 3 ) ≤ . Further equality holds
7
for 𝐺 = 𝐶𝑛 .

Proof. Let 𝐺 be any (𝑛, 𝑚) −graph with diameter is at least 3. Then Δ(𝐺) ≤ 𝑛 − 2. Let 𝑣 be a vertex of
maximum degree in 𝐺 , then 𝑣 must be a maximum degree vertex in 𝐺 3 . Then 𝑣 is adjacent to
𝑁(𝑣), 𝑁(𝑁(𝑣)) and 𝑁(𝑁(𝑁(𝑣))) vertices. Such that Δ(𝐺 3 ) = |𝑁(𝑣)| ∪ |𝑁(𝑁(𝑣))| ∪ |𝑁(𝑁(𝑁(𝑣)))| .
Hence 𝑉(𝐺 3 ) − Δ(𝐺 3 ) will form a dominating set for 𝐺 3 . Further, note that the 𝛿(𝐺) ≥ 2, so every vertex in
𝐺 must be in some edge disjoint induced path 𝑃7 . Also, we know that 𝛿(𝐺) ≤ Δ(𝐺). Therefore, combining
𝑛−𝛥(𝐺)+2
all these properties of 𝐺 and by Theorem 7, we get 𝛾(𝐺 3 ) ≤ .
7
For equality, suppose 𝐺 is a cycle 𝐶𝑛 then the 𝛿(𝐺) = Δ(𝐺) = 2 by using Proposition 2, we get the

© JGRMA 2018, All Rights Reserved 86

 B. Basavanagoud et al, Journal of Global Research in Mathematical Archives, 5(5), 82-89

required result.

Theorem 13. 13 If every support vertex of a tree 𝑇 is adjacent to at least one pendant vertex, then 𝛾(𝑇 3 ) ≤
𝑛−𝑟
, where r is the number of pendant vertices in 𝑇.
4

Proof. Let 𝑋 = {𝑣1 , 𝑣2 , . . . , 𝑣𝑘 } be the set of all support vertices and 𝑌 = {𝑢1 , 𝑢2 , . . . , 𝑢𝑟 } be the set of
pendant vertices such that 𝑋 ∪ 𝑌 = 𝑉. Since each vertex of 𝑌 is adjacent with at least one vertex of 𝑋 in 𝑇,
we have 𝑋 itself a dominating set of 𝑇. Now, in graph 𝑇 3 , 𝐵 = {𝑣1 , 𝑣2 , . . . , 𝑣𝑖 }, 𝑖 ≤ 𝑘, denotes the
dominating set of 𝑇 3 such that 𝐵 ⊆ 𝑋 and every pendant vertex 𝑢𝑖 , 𝑖 = 1,2, . . . , 𝑟 in 𝐺 forms 𝐶3 𝑎𝑛𝑑 𝐶4
with at least two vertices in 𝑋 i.e.,
𝑋
𝐵 ⊆ 4
|𝑉|−|𝑌|
|𝐵| ≤ 4
𝑛−𝑟
≤ 4
3 𝑛−𝑟
𝛾(𝑇 ) ≤ .
4
Hence the proof.

𝑛−𝛼 0 (𝐺)
Theorem 14. 14 For any connected graph 𝐺 , 𝛾(𝐺 3 ) ≤ .
2

Proof. Let 𝑋 = {𝑣1 , 𝑣2 , . . . , 𝑣𝑚 } where 𝑑𝑒𝑔(𝑣𝑖 ) ≥ 2, 1 ≤ 𝑖 ≤ 𝑚 be the set of vertices which covers all the
edges of 𝐺, such that |𝑋| = 𝛼0 (𝐺). Now in 𝐺 3 , let 𝐷 = {𝑣1 , 𝑣2 , . . . , 𝑣𝑘 } be the minimal dominating set.
𝑋
Since 𝑉(𝐺 3 ) = 𝑉(𝐺) it follows that 𝐷 ⊆ 2 . Clearly,

𝑛−|𝑋|
|𝐷| ≤ 2
and hence
𝑛−𝛼 0 (𝐺)
𝛾(𝐺 3 ) ≤ .
2

We need the following theorem for our next result.

Theorem 15. 15 [4] For any graph 𝐺, 𝜒(𝐺) ≤ 𝛥(𝐺) + 1. Where 𝜒(𝐺) is the chromatic number of a graph
𝐺.

𝑛+7
Theorem 16. 16 For any connected graph 𝐺, 𝛾(𝐺 3 ) + 𝜒(𝐺 + 𝛥(𝐺). Where 𝜒(𝐺) is the chromatic
7
number of a graph 𝐺. Further equality holds for 𝐺 = 𝐶2𝑘+1 , for any positive integer 𝑘.

Proof. Let 𝐺 be any connected graph and let 𝑋 = {𝑥1 , 𝑥2 , ⋯ , 𝑥𝑖 } for some 1 ≤ 𝑖 ≤ 𝑛 be any proper coloring
𝑛+7
of 𝐺 . By combining the results of Theorem 7 we have 𝛾(𝐺 3 ) ≤ and Brook's Theorem, 𝜒(𝐺) ≤
7
Δ(𝐺) + 1. Hence
𝑛+7
𝛾(𝐺 3 ) + 𝜒(𝐺) ≤ + Δ(𝐺) + 1
7
𝑛+7
≤ + Δ(𝐺).
7

© JGRMA 2018, All Rights Reserved 87

 B. Basavanagoud et al, Journal of Global Research in Mathematical Archives, 5(5), 82-89

𝑛+7
For equality, suppose 𝐺 = 𝐶2𝑘+1 , then by Proposition 4, 𝛾(𝐶 3 ) = and we know that
7
𝜒(𝐶2𝑘+1 ) = Δ(𝐶2𝑘+1 ) + 1. Hence by combining these two facts, we get the equality for the above bound.

6 Relation with other domination parameters

Theorem 17. 17For any graph 𝐺,

𝛾(𝐺3 )+𝛾(𝐺)
≤ 𝛾𝑡 (𝐺).
2

Proof. To prove this result we consider the following two cases.

case 1: Suppose 𝐺 is a tree, 𝑋 = {𝑣1 , 𝑣2 , . . . , 𝑣𝑘 } be the set of all pendant vertices of 𝐺 and
𝑉′ = 𝑉 − 𝑋, then 𝐷′ ⊆ 𝑉′ is a minimal dominating set of 𝐺. Further, 𝐷′ ∪ 𝐻 , where 𝐻 ∈ 𝑁(𝐷′) and
𝐻 ⊆ 𝑉(𝐺) − 𝐷′ forms minimal total dominating set of 𝐺. If 𝑣𝑗 = {∅}, then there exist at least one vertex
𝑣 ∈ 𝑋 such that 𝐷′ ∪ {𝑣} forms a total dominating set of 𝐺 3 . Let 𝐷 = {𝑢1 , 𝑢2 , . . . , 𝑢𝑚 } be the dominating
set of 𝐺 3 . If the neighbors of each 𝑢𝑖 , 1 ≤ 𝑖 ≤ 𝑚 are at a distance of at most three which generates 𝐷 to be a
minimal dominating set of 𝐺 3 . Then we have,

|𝐷|+|𝐷′|
≤ |𝐷′ ∪ 𝐻|.
2
Gives the required result.
case 2: Suppose 𝐺 is not a tree and let 𝐷 be a minimal dominating set of 𝐺 and 𝑉′ = 𝑉 − 𝐷. Further,
𝐻 ∈ 𝑁(𝐷′), such that 𝐻 ⊆ 𝑉′. Now 𝐷′ ∪ 𝐻 forms minimal total dominating set of 𝐺. In 𝐺 3 let 𝐷 =
{𝑢1 , 𝑢2 , . . . , 𝑢𝑚 } be minimal dominating set which is also subset of 𝐷′. Now
|𝐷|+|𝐷′|
≤ |𝐷′ ∪ 𝐻|
2
gives the required result.

Theorem 18. 18 For any graph 𝐺, 𝛾(𝐺 3 ) < 𝛾𝑐 (𝐺).

Proof. Let 𝑃 be a diametral path such that 𝑃: 𝑣1 , 𝑣2 , . . . , 𝑣𝑘 . Suppose 𝐼 = {𝑣1 , 𝑣2 , . . . , 𝑣𝑖 }, 𝑖 ≤ 𝑘 connected

dominating set of 𝐺 such that 𝐼 ⊆ 𝑉(𝑃) so that |𝐼| = 𝛾𝑐 (𝐺). Now we consider 𝐷 = {𝑣𝑖−1 , 𝑣𝑖−2 , . . . , 𝑣𝑖−𝑚 },
𝑚 ≤ 𝑖 forms dominating set of 𝐺 3 and if each neighbor of 𝑣𝑖−𝑚 , 𝑚 ≤ 𝑖 is at a distance of at most three
forms 𝐷 to be a minimal dominating set. Hence |𝐷| = 𝛾(𝐺 3 ). Clearly, 𝛾(𝐺 3 ) < 𝛾𝑐 (𝐺).
Finally we establish Nordhaus-Gaddum type result.

Theorem 19. 19 For any connected graph 𝐺,

3 2n
(i). 𝛾(𝐺 3 ) + 𝛾(𝐺 ) ≤ and
7
3 n
(ii). 𝛾(𝐺 3 ) ⋅ 𝛾(𝐺 ) ≤ 7

References

© JGRMA 2018, All Rights Reserved 88

 B. Basavanagoud et al, Journal of Global Research in Mathematical Archives, 5(5), 82-89

[1] R. B. Allan, R. C. Laskar, On domination and independent domination numbers of a

graph, Discrete Math. 23 (1978), 73-76.
[2] F, Buckley, F. Harary, Distance in Graphs, Addison-Wesley Pub. Co. (1990).
[3] F. Harary, I. C. Ross, The square of a tree, Bell System Tech. J. 39 (1960), 641-647.
[4] F. Harary, Graph Theory, Addison-Wesley, Reading Mass(1969).
[5] T. Haynes, S. Hedetniemi and P. Slater, Fundamentals of Domination in Graphs,
Marcel Dekker, New York, (1998).
[6] T. Haynes, S. Hedetniemi and P. Slater (eds.), Domination in Graphs: Advanced
Topics, Marcel Dekker, New York, (1998).
[7] M. A. Henning, Distance domination in graphs, In. T.W. Haynes, S.T. Hedetniemi,
P.J. Slater, editors. Domination in graphs: Advanced Topics, Chapter 12, Marcel
Dekker, Inc., New Yprk (1998).
[8] M. H. Muddebihal, G. Shrinivasa, A.R. Sedamkar, Domination in squares of graphs,
Ultra Scientists, 23(3)A (2011), 795-800.

© JGRMA 2018, All Rights Reserved 89