1. Introduction
In general, we use the standard terminology and notation of graph theory. For concepts not defined here, see [
1].
Let G be a finite, simple, undirected graph with the vertex set and the edge set . By , we denote the degree of a vertex x.
Let G be a graph such that , and let be a sequence of arbitrary graphs. The generalized corona of a graph G and the sequence is the graph such that and If , for ; then, we obtain the definition of corona of two graphs G and H.
A subset
is dominating, if every vertex of
G is either in
D or it is adjacent to at least one vertex of
D. The minimum cardinality of all dominating sets of
G is the domination number
. For domination-related concepts not defined here, see [
2].
A subset
is called an independent set of
G if no two vertices from
S are adjacent. A subset of
being independent and dominating simultaneously is an independent dominating set (also named a kernel) of G. The minimum cardinality of an independent dominating set is the independent domination number
. An independent set with minimum cardinality is called an
-set. The theory of kernels was introduced for digraphs by Neumann and Morgenstern in 1953 in the context of game theory, see [
3], and next studied because of its relations with distinct graph problems, for example, in the list colorings and perfectness. Note that Berge was a pioneer in studying independent dominating sets in digraphs; see [
4]. Next this theory was generalized into two directions in the distance sense by considering so-called
-kernels, which were introduced by Borowiecki and Kuzak in [
5], or kernels by the monochromatic paths, introduced by Sands, Sauer and Woodrow in [
6]. Details can be found in a number of papers; for example, see the results of Galeana-Sanchez et al., Kucharska, Szumny et al. and others [
7,
8,
9,
10,
11]. Different types of kernels have been introduced by adding some new restrictions to the classical definition of kernels—for example, restrained independent dominating sets [
12].
In undirected graphs, the problem of the existence of independent dominating sets is trivial because every maximal independent set is dominating. The problem is more complicated and interesting if we add a restriction related to the domination or independence.
Different variants and extensions of dominating sets, for example, total, perfect, efficient, secondary domination, have been introduced and studied in both combinatorial and algoritmic senses; see [
13,
14,
15,
16,
17,
18]. Dominating sets have also various applications in many practical problems, such as electrical networks, monitoring communication and others; for example, see [
19].
Some considered types of independent dominating sets in graphs are based on multiple domination introduced by Fink and Jacobson in [
20]. Let
k be a positive integer. A subset
is
k-dominating if every vertex from
has at least
k neighbors in
D. If
, then we obtain a dominating set in a classical sense. If
, then we obtain a definition of 2-dominating sets. In [
21], A. Włoch introduced the concept of independent 2-dominating sets (named as
-d)-kernels), and next, their some properties were studied in [
16,
22,
23,
24,
25,
26]. Based on the definition of
-d)-kernel, Nagy introduced independent
k-dominating sets as a subset being independent and
k-dominating; details can be found in [
27,
28].
Another type of dominating sets related to multiple domination is -dominating, which we study in this paper.
Let us recall that the set of all vertices which are adjacent to
is named the neighborhood of
x and denoted by
. Let
j and
k be non-negative integers. A subset
is a
-dominating set (in short: a
-dset) if for every vertex
,
. In other words, every vertex
is adjacent to at least
j but not more than
k vertices in
S. Clearly, for
, the set
S does not have to be dominating. Moreover, every
-dset is a
j-dominating set and a dominating set in a classical sense, for all
. The concept of the
-dset was introduced in 2013 by Chellali et al. in [
29]. In earlier work, Dejter in [
30] considered quasiperfect dominating sets, which are basically
-dsets. Moreover, in [
29], it was noted that
-dsets are related to several other types of dominating sets, such as nearly perfect sets (
-dsets), perfect dominating sets (
-dsets) and nearly 2-perfect sets (
-dsets); see, for example, [
15,
31]. In recent years,
-dsets were broadly studied in the literature; see [
32,
33,
34,
35,
36,
37].
A
-dset which is independent we call an independent
-dominating set (in short: an independent
-dset). Clearly, an independent
-dset is an independent dominating set in a classical sense for all
. If
and
, we obtain a definition of an efficient dominating set, which were introduced by Bange, Barkauskas and Slater in [
13] and later studied, for example, in [
38,
39,
40]. In [
41], independent
-dsets were considered. It was noted that not every graph has an independent
-dset, and it was shown that the related decision problem is
-complete. Moreover, the trees with an independent
-dset were characterized.
The graphs products are operations which enable one to construct new classes of graphs. We can describe their properties based on their components. In the context of existence-independent
-dsets, the corona of two graphs was studied. In [
41], the following theorem was proved, which gives sufficient and necessary conditions for the existence of an independent
-dset.
Theorem 1 ([
41]).
Let k be a positive integer and be a corona of graphs G and H. Then, has an independent -dset if and only if each component of G is an isolated vertex or . In this paper, we generalize this result by studying the existence of independent
-dsets in the generalized corona
of a graph
G and a sequence
of graphs. The product
was studied with respect to distinct kinds of domination and together with the independence restriction; see for details [
16]. In this paper, we will give a complete characterization of the generalized corona
with an independent
-dset, for arbitrary positive integers
j and
k.
2. Main Results
In this section, we study and completely solve the problem of the independent -dominating sets in . We begin with the existence of independent -dsets, .
Let us begin with proving lemma, which will be useful in next theorems.
Lemma 1. Let G be an arbitrary graph with vertices and be a sequence of n arbitrary graphs. Moreover, let k be a positive integer and . If the graph has an independent -dset J, then .
Proof. Let J be an independent -dset. By contradiction, let us suppose that there exists such that and . Since the set J is dominating and independent, it contains at least vertices from the set . It means that the vertex is adjacent to at least vertices from the set J, a contradiction. □
Firstly, we provide theorem which provides the complete characterization of with an independent -dset in the case where , for all .
Theorem 2. Let G be an arbitrary graph with vertices and be a sequence of n nonempty graphs. The graph has an independent -dset if and only if, for every , or is an isolated vertex of graph G.
Proof. Let us assume that for every , it holds that or is an isolated vertex of graph G. We will prove that graph has an independent -dset, say, , where and is the minimum dominating set of a graph . If is an isolated vertex of G, then and . Then, the vertices from the set are adjacent to exactly one vertex from the set J. Otherwise, an arbitrary minimum dominating set of graph is included in the set J. Since , the vertices from the set have exactly one neighbor in the set . Hence, the set J is an independent -dset of the graph .
Conversely, let us assume that the graph has an independent -dset J and there exists such that is not an isolated vertex of G and . Since is nonempty graph, . This means that . Let . Then, . By Lemma 1, we obtain that . Since is not an isolated vertex of G, there exists , , such that . The independence of the set J implies that . Since the vertices from the set must be dominated by the set J, . Consequently, we obtain that , which means that the vertex has at least two neighbors in the set J, a contradiction. □
The next theorem completely characterizes the generalized corona having an independent -dset, , where graphs from the sequence are nonempty.
Theorem 3. Let G be an arbitrary graph with vertices and be a sequence of n nonempty graphs. Moreover, let k be a positive integer and . The graph has an independent -dset if and only if following conditions hold:
- (1)
S is an independent set of a graph G and
- (2)
for all vertices .
Proof. Let us suppose that has an independent -dset J. We will show that conditions (1) and (2) must hold.
First, assume that condition (1) does not hold. Since S is not an independent set, there exist vertices and such that . By Lemma 1 we obtain that and . This means that J is not an independent set, a contradiction.
Assume now that condition (2) does not hold. Then, there exists a vertex such that . Since and by Lemma 1, we obtain that . Then, the vertex is adjacent to at least vertices from the set J. Clearly, . Since the vertices from the set must be dominated by the independent set J, at least vertices from the set belong to J. Then, the vertex has at least neighbors in J, a contradiction.
We obtain that if the generalized corona has an independent dset, , then it satisfies conditions (1) and (2).
Conversely, let us assume that conditions (1) and (2) hold. We will show that the graph has an independent -dset , where is an arbitrary of graph . We will prove that J is a -dset.
If , then . Consequently, and . This means that every vertex from the set is adjacent to exactly one vertex from the set J.
If , then , where is an arbitrary -set of . Since graph is nonempty, . Then, every vertex from the set has at most neighbors in the set J. Moreover, the vertex is adjacent to vertices from the set J. If , then it has exactly neighbors in the set J. If , then from condition (2) we know that the vertex has at most k neighbors in J.
We obtain that J is a dset of . Moreover, the definition of the generalized corona implies the independence of the set J. Hence, J is an independent -dset of the graph , which ends the proof. □
From Theorem 3, we obtain the following corollary.
Corollary 1. Let G be an arbitrary graph with vertices and be a sequence of n nonempty graphs. If for every it holds that or is an isolated vertex of G, then graph has an independent -dset.
Proof. Let us assume that or is an isolated vertex of G for every . Let . If , then . This means that S is an independent set and . By Theorem 3 we obtain that has an independent -dset. □
Now we are ready to give equivalent conditions for the existence of an independent -dset in generalized corona.
Corollary 2. Let G be an arbitrary graph with vertices and be a sequence of n nonempty graphs. Let . The graph has an independent -dset if and only if the following conditions hold:
- (1)
S is an independent -dset of a graph G and
- (2)
for all .
Proof. We will show that conditions (1) and (2) from Theorem 3 are equivalent to conditions (1) and (2) from Corollary 2. If conditions (1) and (2) from Corollary 2 hold, the result is obvious. Conversely, by Theorem 3 we know that for all . The graph is nonempty, so . Hence, . Since , we obtain and . Then, for every . Since S is an independent set, it is an independent -dset. □
It is worth emphasizing that if
for every
, then we obtain Theorem 1, which is the result from [
41].
Now we give necessary and sufficient conditions for the existence an independent -dset in , where graphs from the sequence are arbitrary.
Theorem 4. Let G be an arbitrary graph with vertices and be a sequence of n arbitrary graphs. Moreover, let k be a positive integer and . The graph has an independent -dset if and only if the following conditions hold:
- (1)
S is an independent set of a graph G,
- (2)
the subgraph induced by the set has an independent -dset and
- (3)
for all .
Proof. Let us assume that a graph has an independent -dset J. We will show that conditions (1)–(3) must hold. The independence of the set S follows from Lemma 1. Otherwise, the set J is not independent. Hence, condition (1) holds. We will show that the subgraph induced by the set has an independent -dset. If , then . We obtain . Otherwise, J is not independent. Let . Since J is an independent -dset of the graph and , is an independent -dset of the subgraph induced by the set . Thus, condition (2) holds. Now let us suppose that there exists such that . Since , at least vertices from the set belong to the set J. Otherwise, the vertices from the set are not dominated by the set J. Then, the vertex has at least neighbors in the set J, a contradiction. Hence, condition (3) holds.
Conversely, let us assume that conditions (1)–(3) hold. We will show that the graph has an independent -dset , where is an arbitrary -set of . The definition of the generalized corona implies the independence of the set J. It is sufficient to show that J is a -dset. Firstly, if , then . By Lemma 1, we obtain . Then, . The vertices from the set have exactly one neighbor in the set J, i.e., the vertex . Secondly, if and . Since and , and from condition (2), we obtain that vertices from the set belong to the set J or are -dominated by the set J. Finally, if and , then . Hence, . Then every vertex from the set is dominated by the set J and has at most neighbors in the set J. Moreover, the vertex is dominated, for example, by vertices from the set . By condition (3), we obtain that is adjacent to at most k vertices from the set J. Hence, J is an independent -dset of the graph , which ends the proof. □
Now we study the existence of independent -dsets, , in the generalized corona . Let us denote . Clearly, . Let us begin with proving an helpful lemma.
Lemma 2. Let J be an independent -dset, , in the generalized corona . If , then .
Proof. Let J be an independent -dset. By contradiction let us assume that there exists such that . Since J is independent, . This means that vertices from the set have exactly one neighbor in the set J, i.e., the vertex , a contradiction. □
Now we give the necessary and sufficient conditions for the existence independent -dsets, , in the generalized corona of graphs in the case where , for all .
Theorem 5. Let G be an arbitrary graph with vertices and be a sequence of n nonempty graphs. Let . The graph has an independent -dset if and only if all graphs , have an independent -dset such that .
Proof. Let us assume that all graphs , , have an independent -dset such that . We will show that the set is an independent -dset in the graph . From the definition of the generalized corona, we obtain that J is an independent set. Moreover, every vertex has exactly neighbors in the set J. Hence, J is an independent -dset of .
Now, conversely, let us suppose that the graph has an independent -dset J. Note that . By Lemma 2, we obtain that . Then, all graphs , , must have an independent -dset . Since , , and has an independent -dset, the vertex is adjacent to at least j and at most k vertices from the set J. Hence, , which ends the proof. □
The next theorem characterizes generalized corona having an independent -dset, , where graphs from the sequence are arbitrary.
Theorem 6. Let G be an arbitrary graph with vertices and be a sequence of n arbitrary graphs. Let . The graph has an independent -dset if and only if the following conditions hold:
- (1)
every graph , , has an independent -dset ,
- (2)
the subgraph induced by the set has an independent -dset and
- (3)
for all .
Proof. Let us assume that the graph has an independent -dset J. Firstly, by Lemma 2 we obtain that for every . Then, the graph must have an independent -dset . Hence, the condition holds. Secondly, from Lemma 2 we know that if , then . Let . Since J is an independent -dset in generalized corona , is an independent -dset in the subgraph induced by the set . Thus, condition (2) holds. Finally, we will show that condition (3) must hold. Since , the vertex is -dominated by the set J, i.e., . Note that , and . Hence, . Since the sets and are disjoint, . We obtain that condition (3) hold.
Conversely, let us suppose that conditions (1)–(3) hold. We will show that the generalized corona has an independent -dset , where is an independent -dset of graph and is an independent -dset in the subgraph induced by the set . From the definition of , we obtain that J is independent. It is sufficient to show that J is a -dset. Let . If , , then the vertex y is -dominated by the set . If , , then is -dominated by the set . If , , then it is dominated by the vertices from the set and . From condition (3), we obtain that the vertex is -dominated. Hence, the set J is an independent -dset of the graph , which ends the proof. □