International Journal of Computer Applications Technology and Research

Volume 6Issue 2, 106-108, 2017, ISSN:-23198656

Center Concepts on Distance k-Dominating Sets

Dr. A. Anto Kinsley V. Annie Vetha Joeshi
Department of Mathematics Department of Mathematics
St.Xaviers (Autonomous) College St.Xaviers (Autonomous) College
Palayamkottai-627002, India Palayamkottai-627002, India

Abstract: A set () is called a dominating set of if every vertex in () is adjacent to some vertex in . A set is
called a distance k-dominating set of G if each x V D is within distance k from some vertex of D. In this paper, we determine the
distance- domination number for a given graph using the k-center and link vector concepts. Using the k-center concept we present
some necessary and sufficient condition for distance-k dominating set.

Keywords: Distance, radius, domination number, distance k-domination number, k-center, reachable set, link vector.

Theorem 2.2
1. INTRODUCTION
Let = (, ) be a simple graph. The distance between Every central vertex with radius forms a distance k-
and , denoted by d(u, v), is the length of a shortest dominating set.
path. For a vertex and a positive integer k, the k-
neighborhood of in is defined as () ={ Proof
()/ (, ) = }. For = , () is the neighborhood
of and simply denoted by (). Let () = |()| be the Let G be a graph with radius . Let () be the
degree of G and and be the minimum and maximum center of the graph . Let () = { 1 , 2 ,, }. If
degree of , respectively. The set [] = () {} is C(G) , (1 ), then e( ) = k. Hence ( , ) = for
called the closed neighborhood in . some v in V. Therefore ( , ) for all i.e, Each
For a connected graph , the eccentricity () = is with distance to all other vertices in . Hence each can
max{(, ): ()} and the eccentric set () = { dominate all the vertices of G with distance k. Hence every
: (, ) = ()}. The minimum eccentricity among the center vertex with radius forms a distance -dominating set.
vertices of is its radius and the maximum eccentricity is its
diameter, which is denoted by () and (), Theorem 2.3
respectively. A vertex in is a central vertex if () =
() and the subgraph induced by the central vertices of For any connected graph with radius , () = 1.
is the center () of . In this paper, we present the relation
between distance-k dominating set and k-center of the given Proof
graph. We study the binary operations , in [1]. Using these
operations we construct algorithm to find the distance k- Let be any connected graph. Let be the radius of
dominating set. . Let C(G) be the center of G. Then e(v) = k for all
(). Let () then max{d(v, u) = k; }. Also we
Definition 1.1: A set D V(G) is called a dominating set of have d(v, u) k; . Hence v dominates every vertex within
if every vertex in V D is adjacent to some vertex in . distance k. So () = 1.

The domination number () is the minimum cardinality of a Theorem 2.4

dominating set. We call the set of vertices as a -set if it is a
dominating set with cardinality (). For any connected graph G, () = 1 if and only if there
exists a vertex in with eccentricity .
Definition 1.2: A set D V(G) is called a distance k-
dominating set of if [] = . The distance k- domination Proof
number () of .
Let G be any connected graph and V(G) ={1 , 2 ,, } .
2. k-center [1] Suppose that () = 1. Let D be a minimum distance k-
Definition 2.1: Let be a subset of with vertices. Let dominating set. Let .
. Then the distance of from is defined as (, ) =
Case (i): If () (), then there exists a vertex
{ (, ) / }. If then d(S, v) = 0. The can dominate all the vertices within distance k. So that
eccentricity of is the maximum of (, ) over all . ( , ) { } . Which implies that e(vi)=
That is, () = { (, ) / }.Consider the family max{d(vi, v) , vV-{vi}} = k.
of the subset S of vertices (1 k n-1) of . The -center of
the graph is the set of vertices of such that, ( ) = Case(ii): If > (), then there exists a vertex vi can
dominate all other vertices of G with distance less than k.
{(), }. This minimum eccentricity is called the
Hence ( , ) < { } . Which implies that
radius of k-center and it is denoted by (). e(vi)= max{d(vi, v) , v V-{vi}} < k.

www.ijcat.com 106
 International Journal of Computer Applications Technology and Research
Volume 6Issue 2, 106-108, 2017, ISSN:-23198656

Conversely, assume that there exists a vertex with
eccentricity k. Let vi be a vertex with e(vi) k. Then
obviously, d(vi, v) k v V-{vi }. Hence D = { } can
dominate all other vertices of G. Then D is a minimum distance
k-dominating set of G. Hence () = 1.
= (). It implies that D is a k-center. Conversely, suppose
that D is a k-center with radius i. By theorem 2.5, D is a
distance i- dominating set.
Now we take i =1, then we have the following corollary.

Theorem 2.5 Corollary 2.10 [1]

Every -center of G with radius is a distance i- In any graph G with radius 1, a set D of k vertices 1
dominating set. 1 is a dominating set if and only if D is a k-center.

Proof Theorem 2.11

Let be any connected graph with n vertices. Let Let G be a graph with n vertices. Then =1 is full
be the -center of G with radius . Hence Sk = k and e( )= for a least integer k in G for = if and only if =
i. That is the distance of from the farthest vertex is i. {1 , 2 , , } is a minimum distance -dominating set.
Therefore, dominates the farthest vertex with distance i.
Hence dominates all vertices of V within distance i and so Proof
is a distance i-dominating set.
Consider the amount of penetration = . Suppose that
Definition 2.6: The set of all vertices of the graph G, from =1 is full where is the LV of . Take =
which the vertex x is connected within a minimum distance (1 , 2 , , ) for a least integer k. Now=1 =
is called as a reachable set of x within a distance and is
denoted as () = { (, ) }. Call this distance (11 , 12 , , 1 ) (21 , 22 , , 2 )
as penetration. (1 , 2 , , ). Since =1 is full, then
(11 , 12 , , 1 ) (21 , 22 , , 2 )
Definition 2.7: Characterize each vertex as a n-tuple. Each (1 , 2 , , ) = (1, 1, .1). Hence = {1 , 2 , , }
place of n-tuple can be represented by a binary zero or one. dominates V and it is a minimum distance i-dominating set.
Call this n-tuple as a link vector simply LV of a vertex. Since k is minimum.
Thus a link vector (1, 2 , . , ) represent a vertex Conversely, suppose that = {1 , 2 , , } is a
where = 1 if is reachable within the penetration from minimum distance -dominating set. Then a vertex not in D is
and zero otherwise. Denote a link vector of the vertex by adjacent to at least one vertex of D within distance = . That
and denote the set of all link vectors as . is, d(D, y) i y V-D. Thus all coordinates of any one of
1 , 2 , . . , is 1. Hence 1 2 . . is full, that
If all the coordinate of a link vector of a vertex are equal is =1 is full. It completes the proof.
to one then the link vector is said to be full and is denoted as
(1). If all the coordinates of a link vector of a vertex are equal Theorem 2.12
to zero then the link vector is said to be null and it is denoted
by (0). Let G be a graph with n vertices. Then there exists a
vertex whose link vector is full with = if and only if
Definition 2.8: Let G be a graph. Let be the set of LVs of () = 1.
all vertices. Define two binary operations (cup) and (cap)
as follows: Proof
, : by Let G be a graph with n vertices. Suppose that there
exists a vertex whose link vector is full with penetration k.
(1, 2 , . , )(1, 2 , . , ) = (1, 2 , . , ) That is, the jth coordinate of is 1 for every j (1 ).
Hence the vertex is reachable to all other vertices of G with
where = { , } & i= 1 to n penetration k. Hence this vertex alone forms a distance k
dominating set. Hence D is a minimum distance k -dominating
(1, 2 , . , )(1, 2 , . , ) = (1, 2 , . , ) set and so () = 1. Conversely, assume that () = 1. Let
D be a -set of G. Take D = { }. Then the vertex
where = { , } & i = 1 to n dominates all other vertices within distance k. Hence is
reachable to all other vertices of G with = . Then the LV
Theorem 2.9 of is full.
Let G be a graph with n vertices and () = . Let
D of k vertices (1 k n-1). Then D is a distance i-
dominating set if and only if D is a k-center.

Proof Theorem 2.13

Let D V be a set of k vertices with () =i. Suppose If () (), then there exist a LV which
that D is a distance i-dominating set. Then there exists a vertex is full with = .
v in D such that (, ) , for every . () =

www.ijcat.com 107
 International Journal of Computer Applications Technology and Research
Volume 6Issue 2, 106-108, 2017, ISSN:-23198656

Proof Step 2.

Let G be a graph with n vertices. Assume that 2.1. Take w' is the LV of w. Initialize w'(0) and D =
() () . Then there exists a vertex of G with 2.2. For j = 1 to n
eccentricity k. By theorem 2.4, () = 1. Then by theorem
2.12, is full with = k. w' = w' xj'
D = D {xj}
Theorem 2.14
2.3. If is not full then go to step 3.2
Let G be a connected graph with n vertices. Then
2.4. Print D is a minimum distance i -dominating set.
=1 is full with = 1 if and only if G is complete.
Otherwise go to step 3.2
Proof
2.5 Go to step 1.
Let G be a connected graph with n vertices. Suppose that This algorithm finds d number of minimum distance i-
=1 is full with = 1. dominating sets, for i = 1 to d. It works with a for loop j = 1
to n, for a fixed i. To find a minimum distance i-dominating
Then =1 = [(11 , 12 , , 1 ) set this algorithm works with the operations and in 2n
(21 , 22 , , 2 ) . . (1 , 2 , , )= (1, 1,,1). times. Totally it works in 2nd times and so it is a polynomial
Since the jth coordinate of is full for all i, j (1 , ), time algorithm.
then the vertex is reachable to all other vertices. Hence G is
complete. 3. REFERENCES
[1] A. Anto Kinsley and S. Somasundaram, Domination
Conversely, Take=1 = [(11 , 12 , , 1 ) based algorithm to k-center problem, Journal of Discrete
(21 , 22 , , 2 ) . . (1 , 2 , , )]. Since G is Mathematical Sciences and Cryptography, Vol. 9 (2006),
complete, link vector of every vertex is full. Hence =1 is No.3, pp. 403-416.
full. Thus the link vector concept is very useful to prove many [2] F. Buckley and F. Harary, Distance in Graphs, Addison
results. Wesley Publishing Company, New York, (1990).
[3] G. Chartrand and P. Zhang , Introduction to Graph
Algorithm 2.15
Theory, Tata McGrew Hill Education Private Ltd, New
Algorithm to find a minimum distance i-dominating set Delhi (2006), 327-333.
[4] T. W. Haynes, S. T. Hedetniemi and P.J. Slater,
Input. A graph G = (V, E) with V(G) = {x1, x2, , xn} with Fundamentals of Domination in Graphs, Marcel Dekker,
Inc., New York, (1998).
distance matrix and diam(G) = d. Find all reachable
[5] P. J. Slater, Maximin facility location, J. Res. Net
sets (xj) of xj and find the link vector xj' of xj Burstandards, 79B, (1975), 107-115.
Output. Minimum distance i-dominating set. [6] H. S. Wilf, Algorithm and Complexity, Prentice Hall
Step 1. i = 1 to d International, Inc., U.S.A. (1986).

www.ijcat.com 108